Skip to content

Add in-place buffer reuse for arithmetic and bitwise binary expressions#21339

Closed
Dandandan wants to merge 4 commits intoapache:mainfrom
Dandandan:worktree-mutable-binary-eval
Closed

Add in-place buffer reuse for arithmetic and bitwise binary expressions#21339
Dandandan wants to merge 4 commits intoapache:mainfrom
Dandandan:worktree-mutable-binary-eval

Conversation

@Dandandan
Copy link
Copy Markdown
Contributor

Summary

  • When evaluating arithmetic binary expressions (+, -, *, /, %), reuse the left operand's buffer in-place when its Arc reference count is 1, avoiding a buffer allocation
  • Uses PrimitiveArray::unary_mut for array-scalar and into_builder for array-array cases on all integer (i8–u64) and float (f16–f64) types
  • Only wrapping (infallible) ops use in-place mutation; checked overflow ops and division by zero fall back to standard Arrow kernels

Test plan

  • All 1367 existing datafusion-physical-expr tests pass
  • All 26 datafusion-physical-expr-common tests pass
  • cargo clippy clean with -D warnings
  • Benchmark arithmetic-heavy queries (e.g. SELECT a+b+c+d FROM t) to measure allocation savings

🤖 Generated with Claude Code

@github-actions github-actions Bot added the physical-expr Changes to the physical-expr crates label Apr 3, 2026
@Dandandan
Copy link
Copy Markdown
Contributor Author

run benchmarks

@adriangbot
Copy link
Copy Markdown

Benchmark for this request failed.

Last 20 lines of output:

Click to expand
 * [new tag]             4.0.0                -> 4.0.0
 * [new tag]             44.0.0-rc1           -> 44.0.0-rc1
 * [new tag]             5.0.0                -> 5.0.0
 * [new tag]             5.0.0-rc1            -> 5.0.0-rc1
 * [new tag]             5.0.0-rc3            -> 5.0.0-rc3
 * [new tag]             6.0.0                -> 6.0.0
 * [new tag]             6.0.0-rc0            -> 6.0.0-rc0
 * [new tag]             7.0.0                -> 7.0.0
 * [new tag]             7.0.0-rc2            -> 7.0.0-rc2
 * [new tag]             8.0.0                -> 8.0.0
 * [new tag]             8.0.0-rc1            -> 8.0.0-rc1
 * [new tag]             8.0.0-rc2            -> 8.0.0-rc2
 * [new tag]             9.0.0                -> 9.0.0
 * [new tag]             9.0.0-rc1            -> 9.0.0-rc1
 * [new tag]             ballista-0.5.0       -> ballista-0.5.0
 * [new tag]             ballista-0.6.0       -> ballista-0.6.0
 * [new tag]             ballista-0.7.0       -> ballista-0.7.0
 * [new tag]             python-0.3.0         -> python-0.3.0
 * [new tag]             python-0.4.0         -> python-0.4.0
Switched to branch 'worktree-mutable-binary-eval'

File an issue against this benchmark runner

2 similar comments
@adriangbot
Copy link
Copy Markdown

Benchmark for this request failed.

Last 20 lines of output:

Click to expand
 * [new tag]             4.0.0                -> 4.0.0
 * [new tag]             44.0.0-rc1           -> 44.0.0-rc1
 * [new tag]             5.0.0                -> 5.0.0
 * [new tag]             5.0.0-rc1            -> 5.0.0-rc1
 * [new tag]             5.0.0-rc3            -> 5.0.0-rc3
 * [new tag]             6.0.0                -> 6.0.0
 * [new tag]             6.0.0-rc0            -> 6.0.0-rc0
 * [new tag]             7.0.0                -> 7.0.0
 * [new tag]             7.0.0-rc2            -> 7.0.0-rc2
 * [new tag]             8.0.0                -> 8.0.0
 * [new tag]             8.0.0-rc1            -> 8.0.0-rc1
 * [new tag]             8.0.0-rc2            -> 8.0.0-rc2
 * [new tag]             9.0.0                -> 9.0.0
 * [new tag]             9.0.0-rc1            -> 9.0.0-rc1
 * [new tag]             ballista-0.5.0       -> ballista-0.5.0
 * [new tag]             ballista-0.6.0       -> ballista-0.6.0
 * [new tag]             ballista-0.7.0       -> ballista-0.7.0
 * [new tag]             python-0.3.0         -> python-0.3.0
 * [new tag]             python-0.4.0         -> python-0.4.0
Switched to branch 'worktree-mutable-binary-eval'

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

Benchmark for this request failed.

Last 20 lines of output:

Click to expand
 * [new tag]             4.0.0                -> 4.0.0
 * [new tag]             44.0.0-rc1           -> 44.0.0-rc1
 * [new tag]             5.0.0                -> 5.0.0
 * [new tag]             5.0.0-rc1            -> 5.0.0-rc1
 * [new tag]             5.0.0-rc3            -> 5.0.0-rc3
 * [new tag]             6.0.0                -> 6.0.0
 * [new tag]             6.0.0-rc0            -> 6.0.0-rc0
 * [new tag]             7.0.0                -> 7.0.0
 * [new tag]             7.0.0-rc2            -> 7.0.0-rc2
 * [new tag]             8.0.0                -> 8.0.0
 * [new tag]             8.0.0-rc1            -> 8.0.0-rc1
 * [new tag]             8.0.0-rc2            -> 8.0.0-rc2
 * [new tag]             9.0.0                -> 9.0.0
 * [new tag]             9.0.0-rc1            -> 9.0.0-rc1
 * [new tag]             ballista-0.5.0       -> ballista-0.5.0
 * [new tag]             ballista-0.6.0       -> ballista-0.6.0
 * [new tag]             ballista-0.7.0       -> ballista-0.7.0
 * [new tag]             python-0.3.0         -> python-0.3.0
 * [new tag]             python-0.4.0         -> python-0.4.0
Switched to branch 'worktree-mutable-binary-eval'

File an issue against this benchmark runner

When evaluating arithmetic binary expressions (+, -, *, /, %), reuse
the left operand's buffer in-place when its reference count is 1,
avoiding a buffer allocation. This benefits expression chains like
a + b + c where intermediate results are consumed only once.

Uses PrimitiveArray::unary_mut for array-scalar and into_builder for
array-array cases. Only wrapping (infallible) ops use in-place
mutation; checked ops fall back to standard Arrow kernels to avoid
corrupting buffers on overflow.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@Dandandan Dandandan force-pushed the worktree-mutable-binary-eval branch from a1e5de3 to 418ccdd Compare April 3, 2026 12:22
@Dandandan
Copy link
Copy Markdown
Contributor Author

run benchmarks

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183247111-728-fl4x4 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (418ccdd) to 1e93a67 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183247111-730-lxh22 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (418ccdd) to 1e93a67 (merge-base) diff using: tpch
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183247111-729-q2bsg 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (418ccdd) to 1e93a67 (merge-base) diff using: tpcds
Results will be posted here when complete


File an issue against this benchmark runner

When the left operand's buffer cannot be reused (shared reference),
try the right operand for in-place mutation. This covers cases like
Scalar-Array and Array-Array where the right side has refcount 1.

For non-commutative ops (sub, rem), the argument order is preserved
correctly: result[i] = op(left[i], right[i]).

Also refactors type dispatch into shared macros. Decimal types are
excluded from in-place mutation because the result precision/scale
differs from the input.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@Dandandan Dandandan force-pushed the worktree-mutable-binary-eval branch from 418ccdd to 4c6f444 Compare April 3, 2026 12:30
@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃   worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 46.68 / 47.29 ±0.80 / 48.84 ms │ 46.60 / 47.73 ±1.18 / 49.96 ms │     no change │
│ QQuery 2  │ 21.99 / 22.84 ±0.95 / 24.28 ms │ 21.99 / 22.17 ±0.12 / 22.34 ms │     no change │
│ QQuery 3  │ 32.97 / 33.67 ±0.50 / 34.42 ms │ 31.61 / 32.43 ±0.85 / 33.84 ms │     no change │
│ QQuery 4  │ 21.64 / 22.03 ±0.34 / 22.65 ms │ 21.00 / 21.51 ±0.69 / 22.75 ms │     no change │
│ QQuery 5  │ 50.64 / 51.75 ±1.01 / 53.61 ms │ 48.47 / 49.06 ±0.66 / 50.34 ms │ +1.05x faster │
│ QQuery 6  │ 17.61 / 18.38 ±0.91 / 20.16 ms │ 16.92 / 17.20 ±0.23 / 17.64 ms │ +1.07x faster │
│ QQuery 7  │ 58.25 / 59.40 ±0.94 / 60.95 ms │ 55.82 / 57.93 ±1.24 / 59.57 ms │     no change │
│ QQuery 8  │ 49.96 / 50.58 ±0.46 / 51.11 ms │ 48.61 / 50.33 ±1.22 / 51.94 ms │     no change │
│ QQuery 9  │ 56.21 / 57.34 ±1.62 / 60.57 ms │ 54.68 / 55.35 ±0.77 / 56.79 ms │     no change │
│ QQuery 10 │ 70.42 / 74.41 ±2.44 / 78.10 ms │ 70.82 / 72.65 ±1.82 / 75.74 ms │     no change │
│ QQuery 11 │ 14.07 / 14.20 ±0.14 / 14.47 ms │ 15.00 / 15.81 ±0.93 / 17.48 ms │  1.11x slower │
│ QQuery 12 │ 27.44 / 27.86 ±0.31 / 28.38 ms │ 28.65 / 28.85 ±0.15 / 29.02 ms │     no change │
│ QQuery 13 │ 39.79 / 40.47 ±0.70 / 41.36 ms │ 38.64 / 39.86 ±0.75 / 40.87 ms │     no change │
│ QQuery 14 │ 29.04 / 29.46 ±0.27 / 29.76 ms │ 28.41 / 28.55 ±0.17 / 28.83 ms │     no change │
│ QQuery 15 │ 33.46 / 34.15 ±0.56 / 35.03 ms │ 33.29 / 33.54 ±0.18 / 33.84 ms │     no change │
│ QQuery 16 │ 15.93 / 16.36 ±0.27 / 16.75 ms │ 15.61 / 16.26 ±0.53 / 17.09 ms │     no change │
│ QQuery 17 │ 80.78 / 82.14 ±1.38 / 84.56 ms │ 73.57 / 79.20 ±3.74 / 83.07 ms │     no change │
│ QQuery 18 │ 77.78 / 82.09 ±2.65 / 85.73 ms │ 76.45 / 79.99 ±2.41 / 83.63 ms │     no change │
│ QQuery 19 │ 37.10 / 38.08 ±0.62 / 38.85 ms │ 37.20 / 37.47 ±0.22 / 37.73 ms │     no change │
│ QQuery 20 │ 42.71 / 43.57 ±0.58 / 44.37 ms │ 40.65 / 41.59 ±0.72 / 42.75 ms │     no change │
│ QQuery 21 │ 63.56 / 66.88 ±3.23 / 71.76 ms │ 69.60 / 70.76 ±0.99 / 72.16 ms │  1.06x slower │
│ QQuery 22 │ 17.67 / 18.19 ±0.47 / 18.91 ms │ 17.80 / 19.03 ±0.67 / 19.79 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 931.15ms │
│ Total Time (worktree-mutable-binary-eval)   │ 917.25ms │
│ Average Time (HEAD)                         │  42.33ms │
│ Average Time (worktree-mutable-binary-eval) │  41.69ms │
│ Queries Faster                              │        2 │
│ Queries Slower                              │        2 │
│ Queries with No Change                      │       18 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 4.1 GiB
Avg memory 3.6 GiB
CPU user 35.1s
CPU sys 2.7s
Disk read 0 B
Disk write 140.0 KiB

tpch — branch

Metric Value
Wall time 4.8s
Peak memory 4.1 GiB
Avg memory 3.6 GiB
CPU user 34.6s
CPU sys 2.5s
Disk read 0 B
Disk write 60.0 KiB

File an issue against this benchmark runner

When evaluating boolean AND/OR expressions, try to reuse the left
buffer in-place via Buffer::into_mutable. If the left buffer is
shared, try the right buffer (AND/OR are commutative). Falls back
to standard and_kleene/or_kleene when both buffers are shared or
when nulls are present.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.29 / 4.53 ±6.36 / 17.25 ms │          1.30 / 4.55 ±6.41 / 17.36 ms │     no change │
│ QQuery 1  │        14.48 / 14.67 ±0.13 / 14.85 ms │        14.26 / 14.67 ±0.24 / 14.95 ms │     no change │
│ QQuery 2  │        44.02 / 44.48 ±0.36 / 45.08 ms │        43.73 / 44.31 ±0.45 / 45.02 ms │     no change │
│ QQuery 3  │        43.18 / 45.54 ±1.22 / 46.65 ms │        42.76 / 45.89 ±2.24 / 48.45 ms │     no change │
│ QQuery 4  │    291.54 / 304.09 ±11.09 / 324.21 ms │     289.68 / 299.61 ±8.79 / 313.60 ms │     no change │
│ QQuery 5  │     344.86 / 347.95 ±2.60 / 350.95 ms │     348.84 / 351.29 ±2.23 / 354.71 ms │     no change │
│ QQuery 6  │           5.71 / 6.20 ±0.38 / 6.67 ms │           5.54 / 7.03 ±1.28 / 9.09 ms │  1.13x slower │
│ QQuery 7  │        17.01 / 17.50 ±0.62 / 18.73 ms │        16.92 / 19.45 ±4.68 / 28.81 ms │  1.11x slower │
│ QQuery 8  │     420.61 / 429.40 ±5.68 / 436.14 ms │    421.62 / 436.81 ±10.82 / 452.77 ms │     no change │
│ QQuery 9  │    630.22 / 644.65 ±10.49 / 657.89 ms │     635.00 / 647.42 ±8.57 / 659.33 ms │     no change │
│ QQuery 10 │        91.34 / 95.52 ±3.05 / 99.34 ms │        92.56 / 95.18 ±2.52 / 99.82 ms │     no change │
│ QQuery 11 │     104.68 / 105.63 ±0.82 / 107.03 ms │     103.94 / 106.19 ±1.42 / 108.09 ms │     no change │
│ QQuery 12 │     345.68 / 350.77 ±3.53 / 355.52 ms │     351.51 / 354.16 ±2.28 / 357.63 ms │     no change │
│ QQuery 13 │    454.95 / 469.03 ±13.50 / 487.77 ms │    454.83 / 470.57 ±10.10 / 483.22 ms │     no change │
│ QQuery 14 │     347.50 / 353.57 ±5.76 / 364.26 ms │     352.37 / 355.65 ±3.17 / 361.40 ms │     no change │
│ QQuery 15 │    374.14 / 382.07 ±11.10 / 403.89 ms │    362.29 / 377.99 ±19.71 / 416.12 ms │     no change │
│ QQuery 16 │    723.19 / 739.85 ±13.61 / 755.33 ms │    736.65 / 767.74 ±34.04 / 812.93 ms │     no change │
│ QQuery 17 │     714.57 / 721.04 ±6.10 / 728.67 ms │    733.37 / 746.77 ±12.12 / 766.63 ms │     no change │
│ QQuery 18 │ 1445.82 / 1493.42 ±38.83 / 1549.95 ms │ 1407.80 / 1460.71 ±43.17 / 1518.91 ms │     no change │
│ QQuery 19 │        35.06 / 37.10 ±2.64 / 42.04 ms │       35.72 / 45.05 ±15.11 / 75.17 ms │  1.21x slower │
│ QQuery 20 │    714.39 / 743.64 ±25.39 / 787.96 ms │    724.10 / 738.28 ±13.53 / 757.19 ms │     no change │
│ QQuery 21 │     773.45 / 778.94 ±7.48 / 793.73 ms │     774.23 / 779.65 ±6.49 / 791.47 ms │     no change │
│ QQuery 22 │ 1130.47 / 1144.69 ±11.40 / 1161.72 ms │  1133.77 / 1137.41 ±2.00 / 1139.82 ms │     no change │
│ QQuery 23 │ 3033.89 / 3071.17 ±19.78 / 3088.35 ms │ 3091.75 / 3123.63 ±25.69 / 3162.07 ms │     no change │
│ QQuery 24 │     100.24 / 104.13 ±2.47 / 107.59 ms │     102.43 / 103.20 ±0.79 / 104.48 ms │     no change │
│ QQuery 25 │     138.54 / 140.63 ±1.22 / 141.87 ms │     142.75 / 143.67 ±1.05 / 145.38 ms │     no change │
│ QQuery 26 │     100.37 / 102.76 ±1.63 / 105.34 ms │     100.59 / 103.88 ±2.56 / 107.54 ms │     no change │
│ QQuery 27 │     853.77 / 858.83 ±5.39 / 868.99 ms │     852.90 / 860.57 ±5.97 / 868.97 ms │     no change │
│ QQuery 28 │ 7710.55 / 7751.14 ±26.13 / 7784.94 ms │ 7744.63 / 7777.79 ±27.11 / 7813.49 ms │     no change │
│ QQuery 29 │        50.87 / 57.47 ±7.09 / 69.28 ms │        51.39 / 53.87 ±4.08 / 61.96 ms │ +1.07x faster │
│ QQuery 30 │     361.18 / 369.55 ±5.28 / 376.43 ms │     363.57 / 372.78 ±5.51 / 378.19 ms │     no change │
│ QQuery 31 │    352.96 / 375.06 ±11.97 / 387.24 ms │    363.65 / 376.71 ±11.74 / 394.93 ms │     no change │
│ QQuery 32 │ 1109.59 / 1253.95 ±87.06 / 1371.70 ms │ 1025.23 / 1043.27 ±17.78 / 1074.74 ms │ +1.20x faster │
│ QQuery 33 │ 1573.43 / 1613.68 ±26.29 / 1646.45 ms │  1454.97 / 1464.70 ±9.82 / 1482.96 ms │ +1.10x faster │
│ QQuery 34 │ 1521.24 / 1593.54 ±45.74 / 1652.34 ms │ 1453.16 / 1476.92 ±25.79 / 1526.71 ms │ +1.08x faster │
│ QQuery 35 │     384.85 / 392.72 ±4.76 / 399.69 ms │     381.27 / 390.38 ±5.19 / 396.29 ms │     no change │
│ QQuery 36 │     118.61 / 123.03 ±3.20 / 127.75 ms │     120.75 / 124.38 ±2.06 / 126.51 ms │     no change │
│ QQuery 37 │        48.36 / 49.89 ±1.58 / 52.86 ms │        47.47 / 49.59 ±1.63 / 51.57 ms │     no change │
│ QQuery 38 │        75.58 / 76.62 ±1.46 / 79.48 ms │        75.33 / 78.39 ±1.81 / 80.54 ms │     no change │
│ QQuery 39 │     205.01 / 212.76 ±9.94 / 231.67 ms │     210.00 / 221.88 ±6.06 / 226.42 ms │     no change │
│ QQuery 40 │        23.65 / 25.16 ±0.99 / 26.24 ms │        25.01 / 27.41 ±2.43 / 31.71 ms │  1.09x slower │
│ QQuery 41 │        19.35 / 20.14 ±0.56 / 20.89 ms │        20.69 / 23.12 ±1.98 / 26.15 ms │  1.15x slower │
│ QQuery 42 │        19.63 / 20.34 ±0.92 / 22.08 ms │        19.36 / 20.04 ±0.68 / 20.91 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 27486.84ms │
│ Total Time (worktree-mutable-binary-eval)   │ 27142.54ms │
│ Average Time (HEAD)                         │   639.23ms │
│ Average Time (worktree-mutable-binary-eval) │   631.22ms │
│ Queries Faster                              │          4 │
│ Queries Slower                              │          5 │
│ Queries with No Change                      │         34 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 138.6s
Peak memory 38.8 GiB
Avg memory 28.9 GiB
CPU user 1286.5s
CPU sys 107.6s
Disk read 0 B
Disk write 6.0 GiB

clickbench_partitioned — branch

Metric Value
Wall time 136.9s
Peak memory 42.4 GiB
Avg memory 32.1 GiB
CPU user 1286.4s
CPU sys 90.0s
Disk read 0 B
Disk write 776.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃             worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           43.47 / 44.38 ±0.95 / 46.10 ms │           43.25 / 43.91 ±0.66 / 45.10 ms │     no change │
│ QQuery 2  │        148.61 / 149.37 ±0.53 / 150.00 ms │        146.37 / 146.97 ±0.51 / 147.63 ms │     no change │
│ QQuery 3  │        115.28 / 115.84 ±0.45 / 116.45 ms │        114.49 / 114.81 ±0.35 / 115.48 ms │     no change │
│ QQuery 4  │    1371.87 / 1393.91 ±17.31 / 1410.89 ms │    1316.28 / 1346.34 ±24.19 / 1381.53 ms │     no change │
│ QQuery 5  │        173.67 / 175.27 ±1.33 / 177.63 ms │        175.22 / 176.29 ±1.10 / 178.18 ms │     no change │
│ QQuery 6  │    1010.97 / 1030.49 ±15.54 / 1049.35 ms │    1021.78 / 1048.49 ±17.95 / 1077.70 ms │     no change │
│ QQuery 7  │        352.07 / 355.56 ±2.13 / 358.00 ms │        352.68 / 356.46 ±3.99 / 363.78 ms │     no change │
│ QQuery 8  │        116.49 / 118.13 ±0.87 / 118.98 ms │        115.46 / 116.70 ±0.70 / 117.44 ms │     no change │
│ QQuery 9  │        102.12 / 106.20 ±2.69 / 109.24 ms │        101.14 / 107.98 ±9.82 / 127.24 ms │     no change │
│ QQuery 10 │        109.39 / 110.34 ±0.85 / 111.73 ms │        105.52 / 107.41 ±1.47 / 109.40 ms │     no change │
│ QQuery 11 │      938.87 / 986.49 ±35.81 / 1041.69 ms │       913.18 / 937.03 ±17.08 / 965.35 ms │ +1.05x faster │
│ QQuery 12 │           45.51 / 46.99 ±1.69 / 49.85 ms │           44.11 / 47.84 ±1.89 / 49.23 ms │     no change │
│ QQuery 13 │        402.30 / 405.76 ±2.97 / 410.04 ms │        399.69 / 404.11 ±4.13 / 411.90 ms │     no change │
│ QQuery 14 │    1018.54 / 1034.81 ±10.39 / 1045.37 ms │    1009.78 / 1033.00 ±13.69 / 1052.12 ms │     no change │
│ QQuery 15 │           16.43 / 17.30 ±0.81 / 18.84 ms │           16.27 / 17.46 ±1.25 / 19.84 ms │     no change │
│ QQuery 16 │           40.76 / 41.69 ±0.98 / 43.48 ms │           40.68 / 41.64 ±0.75 / 42.72 ms │     no change │
│ QQuery 17 │        249.67 / 252.09 ±1.57 / 254.38 ms │        237.62 / 239.96 ±2.00 / 243.46 ms │     no change │
│ QQuery 18 │        131.63 / 133.09 ±0.94 / 134.09 ms │        127.23 / 128.37 ±0.89 / 129.60 ms │     no change │
│ QQuery 19 │        159.50 / 161.07 ±1.44 / 163.66 ms │        153.88 / 156.30 ±1.29 / 157.56 ms │     no change │
│ QQuery 20 │           14.06 / 14.43 ±0.36 / 15.07 ms │           13.87 / 14.34 ±0.31 / 14.77 ms │     no change │
│ QQuery 21 │           20.21 / 20.61 ±0.32 / 21.15 ms │           19.13 / 19.45 ±0.24 / 19.87 ms │ +1.06x faster │
│ QQuery 22 │        492.70 / 494.07 ±1.30 / 496.43 ms │        489.78 / 492.84 ±2.68 / 497.28 ms │     no change │
│ QQuery 23 │       891.60 / 905.85 ±11.42 / 922.88 ms │       875.61 / 913.28 ±34.64 / 974.49 ms │     no change │
│ QQuery 24 │        414.56 / 417.05 ±1.61 / 418.83 ms │        424.83 / 430.23 ±4.64 / 438.15 ms │     no change │
│ QQuery 25 │        352.19 / 357.29 ±3.10 / 360.52 ms │        355.19 / 357.54 ±2.00 / 360.64 ms │     no change │
│ QQuery 26 │           80.94 / 82.12 ±1.43 / 84.84 ms │           81.45 / 83.41 ±1.35 / 84.73 ms │     no change │
│ QQuery 27 │        352.48 / 354.76 ±1.64 / 356.56 ms │        345.60 / 348.66 ±1.81 / 351.05 ms │     no change │
│ QQuery 28 │        149.88 / 151.17 ±1.69 / 154.43 ms │        149.59 / 150.60 ±1.01 / 152.37 ms │     no change │
│ QQuery 29 │        298.61 / 301.36 ±2.81 / 306.33 ms │        295.67 / 298.98 ±3.11 / 302.86 ms │     no change │
│ QQuery 30 │           42.28 / 45.14 ±1.93 / 47.04 ms │           42.75 / 45.43 ±1.77 / 47.97 ms │     no change │
│ QQuery 31 │        169.32 / 172.59 ±2.54 / 175.47 ms │        171.40 / 173.69 ±1.49 / 176.02 ms │     no change │
│ QQuery 32 │           57.15 / 57.69 ±0.47 / 58.36 ms │          58.64 / 67.54 ±15.85 / 99.20 ms │  1.17x slower │
│ QQuery 33 │        140.54 / 141.96 ±1.04 / 143.17 ms │        140.61 / 142.33 ±0.93 / 143.15 ms │     no change │
│ QQuery 34 │        106.65 / 107.72 ±1.34 / 110.26 ms │        107.17 / 107.90 ±0.59 / 108.64 ms │     no change │
│ QQuery 35 │        106.99 / 109.33 ±1.76 / 111.74 ms │        106.38 / 109.20 ±2.11 / 112.90 ms │     no change │
│ QQuery 36 │        219.99 / 223.82 ±2.36 / 226.32 ms │        216.89 / 220.79 ±2.42 / 223.39 ms │     no change │
│ QQuery 37 │        175.92 / 178.42 ±3.08 / 184.20 ms │        176.84 / 179.81 ±2.29 / 183.31 ms │     no change │
│ QQuery 38 │           85.57 / 88.24 ±1.96 / 91.07 ms │           87.52 / 89.83 ±2.11 / 93.78 ms │     no change │
│ QQuery 39 │        127.96 / 129.63 ±1.25 / 131.34 ms │        137.03 / 138.66 ±1.28 / 140.89 ms │  1.07x slower │
│ QQuery 40 │        115.09 / 120.26 ±6.74 / 133.50 ms │        111.50 / 120.14 ±8.99 / 137.20 ms │     no change │
│ QQuery 41 │           14.01 / 14.85 ±0.80 / 15.89 ms │           16.42 / 17.88 ±2.04 / 21.91 ms │  1.20x slower │
│ QQuery 42 │        107.79 / 109.51 ±1.02 / 110.85 ms │        109.46 / 111.84 ±1.41 / 113.65 ms │     no change │
│ QQuery 43 │           82.67 / 83.91 ±0.84 / 84.68 ms │           83.60 / 84.90 ±1.56 / 87.84 ms │     no change │
│ QQuery 44 │           11.20 / 11.61 ±0.31 / 12.02 ms │           11.93 / 12.23 ±0.15 / 12.37 ms │  1.05x slower │
│ QQuery 45 │           52.51 / 54.20 ±1.14 / 55.64 ms │           52.19 / 53.60 ±1.04 / 54.98 ms │     no change │
│ QQuery 46 │        228.42 / 235.57 ±4.40 / 241.46 ms │        230.82 / 233.43 ±1.80 / 235.38 ms │     no change │
│ QQuery 47 │       718.11 / 770.24 ±40.70 / 837.01 ms │        696.98 / 702.51 ±8.39 / 719.22 ms │ +1.10x faster │
│ QQuery 48 │        289.13 / 295.39 ±5.54 / 303.23 ms │        291.31 / 294.58 ±3.47 / 301.28 ms │     no change │
│ QQuery 49 │        253.81 / 257.41 ±2.40 / 261.35 ms │        253.72 / 255.42 ±1.50 / 257.42 ms │     no change │
│ QQuery 50 │        231.34 / 237.55 ±4.91 / 245.62 ms │        232.70 / 235.72 ±3.99 / 243.58 ms │     no change │
│ QQuery 51 │        188.60 / 190.71 ±1.56 / 192.32 ms │        181.52 / 184.70 ±2.42 / 187.60 ms │     no change │
│ QQuery 52 │        107.62 / 109.39 ±1.18 / 110.96 ms │        108.32 / 108.60 ±0.24 / 108.95 ms │     no change │
│ QQuery 53 │        102.85 / 103.92 ±0.88 / 105.23 ms │        102.16 / 102.86 ±0.59 / 103.76 ms │     no change │
│ QQuery 54 │        148.65 / 150.11 ±1.03 / 151.63 ms │        147.96 / 148.83 ±0.90 / 150.36 ms │     no change │
│ QQuery 55 │        107.64 / 108.77 ±1.40 / 111.45 ms │        107.74 / 108.82 ±0.98 / 110.15 ms │     no change │
│ QQuery 56 │        141.43 / 142.61 ±1.49 / 145.46 ms │        142.49 / 145.72 ±1.99 / 147.63 ms │     no change │
│ QQuery 57 │        173.59 / 175.31 ±1.11 / 177.06 ms │        175.77 / 177.89 ±1.66 / 180.78 ms │     no change │
│ QQuery 58 │        296.47 / 303.56 ±5.93 / 314.12 ms │        291.32 / 301.02 ±6.72 / 309.34 ms │     no change │
│ QQuery 59 │        199.77 / 202.23 ±1.33 / 203.80 ms │        197.16 / 200.42 ±3.46 / 207.11 ms │     no change │
│ QQuery 60 │        143.48 / 145.37 ±1.19 / 146.94 ms │        143.31 / 144.74 ±0.99 / 146.38 ms │     no change │
│ QQuery 61 │        171.97 / 173.62 ±1.30 / 175.68 ms │        170.50 / 171.94 ±1.28 / 173.78 ms │     no change │
│ QQuery 62 │      914.29 / 952.30 ±30.28 / 1000.49 ms │      914.31 / 958.50 ±41.38 / 1016.11 ms │     no change │
│ QQuery 63 │        102.37 / 106.50 ±4.34 / 114.89 ms │        103.34 / 105.42 ±1.64 / 108.24 ms │     no change │
│ QQuery 64 │        694.40 / 700.99 ±8.47 / 717.56 ms │        696.81 / 700.49 ±2.43 / 702.66 ms │     no change │
│ QQuery 65 │        251.09 / 256.51 ±5.97 / 267.53 ms │        255.45 / 258.55 ±2.28 / 262.31 ms │     no change │
│ QQuery 66 │       248.37 / 256.68 ±10.32 / 274.65 ms │       244.32 / 259.63 ±13.95 / 279.22 ms │     no change │
│ QQuery 67 │        307.46 / 314.28 ±7.11 / 327.98 ms │        324.79 / 336.36 ±7.05 / 347.05 ms │  1.07x slower │
│ QQuery 68 │        279.41 / 289.19 ±6.73 / 299.60 ms │        283.84 / 289.71 ±4.22 / 296.02 ms │     no change │
│ QQuery 69 │        102.59 / 105.71 ±2.91 / 110.89 ms │        102.41 / 103.87 ±1.09 / 105.46 ms │     no change │
│ QQuery 70 │       336.63 / 358.74 ±13.10 / 377.36 ms │       332.64 / 346.56 ±10.74 / 359.76 ms │     no change │
│ QQuery 71 │        135.42 / 139.89 ±4.21 / 146.81 ms │        134.78 / 135.28 ±0.39 / 135.91 ms │     no change │
│ QQuery 72 │       706.62 / 723.85 ±11.60 / 738.41 ms │       698.66 / 718.03 ±10.62 / 726.81 ms │     no change │
│ QQuery 73 │        101.33 / 105.08 ±2.29 / 108.30 ms │        101.38 / 104.23 ±2.09 / 107.03 ms │     no change │
│ QQuery 74 │        574.77 / 583.90 ±7.09 / 593.72 ms │       566.97 / 592.49 ±32.37 / 656.08 ms │     no change │
│ QQuery 75 │        277.96 / 279.46 ±0.98 / 281.04 ms │        282.21 / 284.54 ±2.83 / 289.86 ms │     no change │
│ QQuery 76 │        131.61 / 133.07 ±1.48 / 135.82 ms │        135.07 / 137.42 ±1.56 / 139.70 ms │     no change │
│ QQuery 77 │        188.47 / 190.07 ±1.09 / 191.66 ms │        189.50 / 192.28 ±2.26 / 195.76 ms │     no change │
│ QQuery 78 │        352.84 / 357.71 ±4.95 / 366.74 ms │        351.86 / 357.23 ±3.69 / 361.57 ms │     no change │
│ QQuery 79 │        247.20 / 253.82 ±4.36 / 258.86 ms │        230.63 / 234.37 ±3.47 / 239.41 ms │ +1.08x faster │
│ QQuery 80 │        331.75 / 335.92 ±2.86 / 339.11 ms │        327.56 / 330.82 ±2.14 / 333.19 ms │     no change │
│ QQuery 81 │           27.53 / 28.67 ±1.29 / 30.98 ms │           26.09 / 27.15 ±0.76 / 28.00 ms │ +1.06x faster │
│ QQuery 82 │        199.02 / 200.96 ±2.12 / 204.54 ms │        197.06 / 199.35 ±2.47 / 203.71 ms │     no change │
│ QQuery 83 │           40.06 / 42.05 ±1.02 / 42.81 ms │           38.87 / 40.56 ±1.42 / 42.26 ms │     no change │
│ QQuery 84 │           48.70 / 49.56 ±0.86 / 50.86 ms │           48.00 / 49.19 ±0.72 / 50.22 ms │     no change │
│ QQuery 85 │        147.98 / 150.00 ±1.53 / 151.77 ms │        150.65 / 151.06 ±0.35 / 151.49 ms │     no change │
│ QQuery 86 │           39.03 / 39.98 ±0.69 / 41.17 ms │           39.43 / 40.23 ±0.59 / 40.97 ms │     no change │
│ QQuery 87 │           87.64 / 90.77 ±2.76 / 95.36 ms │           86.78 / 88.98 ±3.11 / 95.10 ms │     no change │
│ QQuery 88 │         99.97 / 101.17 ±0.92 / 102.75 ms │         99.47 / 100.26 ±0.62 / 101.22 ms │     no change │
│ QQuery 89 │        117.19 / 118.83 ±1.04 / 119.98 ms │        118.61 / 120.55 ±1.85 / 123.99 ms │     no change │
│ QQuery 90 │           22.75 / 23.82 ±0.73 / 24.68 ms │           23.76 / 25.01 ±1.15 / 27.02 ms │     no change │
│ QQuery 91 │           62.65 / 63.97 ±1.10 / 65.90 ms │           62.39 / 65.80 ±1.96 / 67.97 ms │     no change │
│ QQuery 92 │           57.37 / 57.97 ±0.56 / 59.01 ms │           57.42 / 58.44 ±0.74 / 59.59 ms │     no change │
│ QQuery 93 │        194.25 / 196.67 ±2.05 / 199.78 ms │        190.93 / 193.59 ±2.58 / 196.95 ms │     no change │
│ QQuery 94 │           61.20 / 62.33 ±0.92 / 63.72 ms │           60.53 / 61.56 ±0.72 / 62.53 ms │     no change │
│ QQuery 95 │        136.13 / 138.19 ±1.92 / 141.76 ms │        133.71 / 136.04 ±1.40 / 138.06 ms │     no change │
│ QQuery 96 │           72.87 / 75.52 ±1.88 / 78.76 ms │           72.32 / 79.53 ±7.95 / 95.01 ms │  1.05x slower │
│ QQuery 97 │        132.12 / 134.00 ±1.82 / 137.43 ms │        132.58 / 135.59 ±1.53 / 136.68 ms │     no change │
│ QQuery 98 │        151.12 / 154.89 ±2.29 / 157.41 ms │        155.73 / 157.41 ±1.32 / 159.70 ms │     no change │
│ QQuery 99 │ 10713.08 / 10768.08 ±36.82 / 10825.38 ms │ 10760.06 / 10831.93 ±48.10 / 10886.67 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 33969.21ms │
│ Total Time (worktree-mutable-binary-eval)   │ 33881.35ms │
│ Average Time (HEAD)                         │   343.12ms │
│ Average Time (worktree-mutable-binary-eval) │   342.24ms │
│ Queries Faster                              │          5 │
│ Queries Slower                              │          6 │
│ Queries with No Change                      │         88 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 170.2s
Peak memory 5.6 GiB
Avg memory 4.5 GiB
CPU user 273.5s
CPU sys 18.0s
Disk read 0 B
Disk write 638.2 MiB

tpcds — branch

Metric Value
Wall time 169.7s
Peak memory 5.4 GiB
Avg memory 4.6 GiB
CPU user 272.9s
CPU sys 17.9s
Disk read 0 B
Disk write 152.0 KiB

File an issue against this benchmark runner

@Dandandan
Copy link
Copy Markdown
Contributor Author

run benchmarks

@Dandandan
Copy link
Copy Markdown
Contributor Author

run benchmark tpch10

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183330026-735-vrtg4 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (01f8fa6) to 1e93a67 (merge-base) diff using: tpcds
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183330026-734-8kc9b 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (01f8fa6) to 1e93a67 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183330315-737-4pd86 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (01f8fa6) to 1e93a67 (merge-base) diff using: tpch10
Results will be posted here when complete


File an issue against this benchmark runner

@Dandandan Dandandan changed the title Add in-place buffer reuse for arithmetic binary expressions Add in-place buffer reuse for arithmetic and boolean binary expressions Apr 3, 2026
@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183330026-736-jb92r 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (01f8fa6) to 1e93a67 (merge-base) diff using: tpch
Results will be posted here when complete


File an issue against this benchmark runner

When evaluating boolean AND/OR expressions without nulls, use
BooleanBuffer's BitAndAssign/BitOrAssign which internally attempts
Buffer::into_mutable() for in-place mutation. Falls back to
and_kleene/or_kleene when nulls are present.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@Dandandan Dandandan force-pushed the worktree-mutable-binary-eval branch from 01f8fa6 to f70c9e7 Compare April 3, 2026 12:47
@Dandandan Dandandan changed the title Add in-place buffer reuse for arithmetic and boolean binary expressions Add in-place buffer reuse for arithmetic and bitwise binary expressions Apr 3, 2026
@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃   worktree-mutable-binary-eval ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 45.17 / 45.80 ±0.82 / 47.33 ms │ 45.38 / 46.05 ±0.87 / 47.76 ms │ no change │
│ QQuery 2  │ 21.00 / 21.29 ±0.19 / 21.54 ms │ 21.09 / 21.37 ±0.23 / 21.66 ms │ no change │
│ QQuery 3  │ 31.44 / 31.80 ±0.24 / 32.15 ms │ 31.64 / 32.01 ±0.35 / 32.64 ms │ no change │
│ QQuery 4  │ 20.99 / 21.58 ±0.34 / 22.00 ms │ 20.62 / 20.98 ±0.25 / 21.32 ms │ no change │
│ QQuery 5  │ 50.38 / 52.32 ±1.06 / 53.42 ms │ 50.43 / 52.35 ±1.46 / 53.88 ms │ no change │
│ QQuery 6  │ 16.71 / 17.37 ±0.62 / 18.54 ms │ 17.01 / 17.79 ±0.76 / 18.83 ms │ no change │
│ QQuery 7  │ 54.67 / 55.51 ±0.77 / 56.48 ms │ 55.93 / 58.12 ±1.77 / 59.92 ms │ no change │
│ QQuery 8  │ 48.15 / 48.70 ±0.59 / 49.56 ms │ 47.63 / 47.89 ±0.17 / 48.15 ms │ no change │
│ QQuery 9  │ 56.64 / 57.82 ±1.63 / 61.00 ms │ 54.00 / 55.65 ±1.74 / 58.32 ms │ no change │
│ QQuery 10 │ 71.21 / 72.80 ±1.20 / 74.46 ms │ 72.46 / 75.32 ±2.56 / 79.26 ms │ no change │
│ QQuery 11 │ 13.89 / 14.21 ±0.23 / 14.55 ms │ 14.33 / 14.60 ±0.38 / 15.36 ms │ no change │
│ QQuery 12 │ 27.15 / 27.98 ±0.64 / 29.00 ms │ 27.69 / 28.71 ±0.76 / 29.97 ms │ no change │
│ QQuery 13 │ 39.44 / 40.15 ±0.54 / 40.65 ms │ 39.54 / 40.05 ±0.47 / 40.87 ms │ no change │
│ QQuery 14 │ 29.16 / 30.01 ±0.85 / 31.28 ms │ 28.92 / 29.38 ±0.66 / 30.70 ms │ no change │
│ QQuery 15 │ 33.88 / 34.86 ±0.59 / 35.54 ms │ 35.00 / 35.72 ±0.71 / 37.06 ms │ no change │
│ QQuery 16 │ 16.46 / 17.43 ±0.75 / 18.44 ms │ 17.74 / 18.17 ±0.45 / 18.75 ms │ no change │
│ QQuery 17 │ 75.85 / 78.54 ±2.52 / 82.71 ms │ 77.62 / 79.94 ±2.60 / 84.66 ms │ no change │
│ QQuery 18 │ 77.02 / 82.44 ±3.72 / 87.23 ms │ 81.38 / 81.96 ±0.56 / 82.80 ms │ no change │
│ QQuery 19 │ 37.27 / 38.96 ±1.08 / 40.48 ms │ 39.94 / 40.90 ±0.63 / 41.74 ms │ no change │
│ QQuery 20 │ 39.31 / 40.94 ±1.79 / 44.20 ms │ 40.95 / 42.65 ±1.15 / 44.08 ms │ no change │
│ QQuery 21 │ 62.47 / 65.11 ±1.54 / 67.16 ms │ 62.69 / 64.66 ±1.57 / 66.83 ms │ no change │
│ QQuery 22 │ 17.81 / 18.84 ±0.56 / 19.39 ms │ 17.99 / 18.41 ±0.39 / 19.08 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 914.47ms │
│ Total Time (worktree-mutable-binary-eval)   │ 922.70ms │
│ Average Time (HEAD)                         │  41.57ms │
│ Average Time (worktree-mutable-binary-eval) │  41.94ms │
│ Queries Faster                              │        0 │
│ Queries Slower                              │        0 │
│ Queries with No Change                      │       22 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 4.9s
Peak memory 4.1 GiB
Avg memory 3.6 GiB
CPU user 34.2s
CPU sys 2.6s
Disk read 0 B
Disk write 136.0 KiB

tpch — branch

Metric Value
Wall time 4.9s
Peak memory 4.1 GiB
Avg memory 3.6 GiB
CPU user 34.7s
CPU sys 2.6s
Disk read 0 B
Disk write 60.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃       worktree-mutable-binary-eval ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │  373.03 / 373.76 ±0.76 / 375.21 ms │  377.11 / 378.94 ±1.05 / 380.38 ms │ no change │
│ QQuery 2  │  140.38 / 141.91 ±1.77 / 144.22 ms │  140.31 / 142.81 ±1.99 / 145.94 ms │ no change │
│ QQuery 3  │  296.84 / 303.65 ±7.21 / 315.84 ms │  298.25 / 303.67 ±3.05 / 307.74 ms │ no change │
│ QQuery 4  │  155.58 / 159.33 ±2.34 / 162.70 ms │  156.38 / 159.48 ±1.58 / 160.60 ms │ no change │
│ QQuery 5  │ 450.90 / 458.83 ±10.36 / 478.30 ms │  440.88 / 452.23 ±7.44 / 461.02 ms │ no change │
│ QQuery 6  │  131.09 / 132.41 ±0.77 / 133.24 ms │  131.97 / 133.23 ±0.88 / 134.51 ms │ no change │
│ QQuery 7  │  577.50 / 584.89 ±4.70 / 591.29 ms │ 571.66 / 584.10 ±12.94 / 604.69 ms │ no change │
│ QQuery 8  │  483.05 / 491.98 ±5.96 / 500.48 ms │  486.19 / 492.05 ±3.96 / 498.43 ms │ no change │
│ QQuery 9  │  710.49 / 716.42 ±5.31 / 726.31 ms │  704.42 / 715.45 ±7.73 / 724.96 ms │ no change │
│ QQuery 10 │  345.85 / 349.91 ±3.75 / 356.67 ms │  341.53 / 347.79 ±4.76 / 354.33 ms │ no change │
│ QQuery 11 │  110.91 / 113.50 ±2.52 / 117.49 ms │  109.40 / 114.37 ±2.90 / 117.46 ms │ no change │
│ QQuery 12 │  205.72 / 207.72 ±1.88 / 210.74 ms │  207.45 / 211.32 ±3.39 / 216.89 ms │ no change │
│ QQuery 13 │  319.41 / 332.64 ±7.29 / 339.82 ms │  325.95 / 332.36 ±4.68 / 338.61 ms │ no change │
│ QQuery 14 │  181.59 / 184.09 ±1.99 / 186.53 ms │  187.66 / 190.08 ±2.08 / 192.89 ms │ no change │
│ QQuery 15 │  333.60 / 337.11 ±2.20 / 340.30 ms │  340.17 / 343.38 ±2.63 / 348.11 ms │ no change │
│ QQuery 16 │     88.37 / 89.22 ±0.67 / 90.35 ms │     90.48 / 91.25 ±0.62 / 92.32 ms │ no change │
│ QQuery 17 │  796.93 / 803.75 ±6.26 / 815.50 ms │ 784.30 / 797.74 ±11.52 / 814.74 ms │ no change │
│ QQuery 18 │ 874.38 / 909.60 ±20.75 / 931.34 ms │ 888.01 / 933.31 ±31.35 / 984.42 ms │ no change │
│ QQuery 19 │  274.86 / 281.80 ±4.81 / 289.22 ms │  278.47 / 282.88 ±5.27 / 292.86 ms │ no change │
│ QQuery 20 │ 330.29 / 341.34 ±11.92 / 363.74 ms │  340.70 / 345.35 ±6.27 / 357.39 ms │ no change │
│ QQuery 21 │ 865.47 / 884.39 ±16.45 / 908.87 ms │ 876.62 / 889.92 ±10.72 / 904.39 ms │ no change │
│ QQuery 22 │     83.21 / 84.36 ±1.34 / 86.87 ms │     85.79 / 88.30 ±2.01 / 90.65 ms │ no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 8282.62ms │
│ Total Time (worktree-mutable-binary-eval)   │ 8330.02ms │
│ Average Time (HEAD)                         │  376.48ms │
│ Average Time (worktree-mutable-binary-eval) │  378.64ms │
│ Queries Faster                              │         0 │
│ Queries Slower                              │         0 │
│ Queries with No Change                      │        22 │
│ Queries with Failure                        │         0 │
└─────────────────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric Value
Wall time 41.8s
Peak memory 10.2 GiB
Avg memory 7.7 GiB
CPU user 439.4s
CPU sys 27.3s
Disk read 0 B
Disk write 2.9 GiB

tpch10 — branch

Metric Value
Wall time 42.0s
Peak memory 9.8 GiB
Avg memory 7.7 GiB
CPU user 441.8s
CPU sys 27.3s
Disk read 0 B
Disk write 1.6 MiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃             worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           42.98 / 43.94 ±0.85 / 45.23 ms │           43.70 / 44.54 ±0.86 / 46.06 ms │     no change │
│ QQuery 2  │        149.16 / 149.59 ±0.44 / 150.33 ms │        146.31 / 147.70 ±0.93 / 148.90 ms │     no change │
│ QQuery 3  │        114.87 / 115.59 ±0.94 / 117.45 ms │        115.03 / 115.59 ±0.43 / 116.23 ms │     no change │
│ QQuery 4  │    1317.86 / 1388.70 ±51.95 / 1447.43 ms │    1340.19 / 1372.20 ±35.06 / 1423.77 ms │     no change │
│ QQuery 5  │        174.49 / 175.55 ±0.90 / 176.67 ms │        172.19 / 172.77 ±0.60 / 173.52 ms │     no change │
│ QQuery 6  │    1028.31 / 1045.10 ±10.98 / 1058.54 ms │     982.51 / 1017.67 ±23.38 / 1051.79 ms │     no change │
│ QQuery 7  │        351.21 / 354.55 ±2.35 / 358.18 ms │        356.07 / 360.24 ±2.86 / 363.47 ms │     no change │
│ QQuery 8  │        116.05 / 117.68 ±1.23 / 119.56 ms │        115.50 / 117.10 ±1.48 / 119.63 ms │     no change │
│ QQuery 9  │        100.93 / 103.92 ±2.64 / 107.18 ms │        100.85 / 111.83 ±9.66 / 128.15 ms │  1.08x slower │
│ QQuery 10 │        105.83 / 107.14 ±0.85 / 108.49 ms │        107.74 / 108.57 ±0.53 / 109.26 ms │     no change │
│ QQuery 11 │      929.24 / 990.15 ±49.19 / 1034.21 ms │        921.93 / 936.22 ±8.12 / 946.13 ms │ +1.06x faster │
│ QQuery 12 │           44.55 / 47.11 ±2.13 / 50.78 ms │           46.67 / 47.36 ±0.47 / 47.90 ms │     no change │
│ QQuery 13 │        402.85 / 405.07 ±1.50 / 406.98 ms │        394.40 / 398.66 ±2.52 / 401.67 ms │     no change │
│ QQuery 14 │     1019.31 / 1027.95 ±6.64 / 1037.13 ms │     1016.23 / 1021.46 ±4.36 / 1029.19 ms │     no change │
│ QQuery 15 │           16.50 / 17.88 ±1.04 / 19.07 ms │           15.89 / 17.64 ±1.26 / 19.42 ms │     no change │
│ QQuery 16 │           41.17 / 42.37 ±0.67 / 43.09 ms │           41.01 / 41.67 ±0.54 / 42.63 ms │     no change │
│ QQuery 17 │        238.88 / 240.69 ±1.56 / 243.40 ms │        242.56 / 246.57 ±2.69 / 250.38 ms │     no change │
│ QQuery 18 │        128.64 / 130.05 ±1.29 / 132.33 ms │        130.50 / 131.69 ±0.76 / 132.72 ms │     no change │
│ QQuery 19 │        155.54 / 156.42 ±0.79 / 157.65 ms │        155.60 / 157.44 ±1.01 / 158.48 ms │     no change │
│ QQuery 20 │           13.54 / 14.39 ±0.49 / 14.94 ms │           13.92 / 14.52 ±0.34 / 14.91 ms │     no change │
│ QQuery 21 │           19.68 / 20.44 ±0.50 / 21.01 ms │           19.73 / 20.14 ±0.40 / 20.88 ms │     no change │
│ QQuery 22 │        493.22 / 499.96 ±9.56 / 518.50 ms │        493.54 / 498.47 ±4.13 / 503.45 ms │     no change │
│ QQuery 23 │       894.31 / 936.85 ±30.81 / 985.27 ms │       886.99 / 901.48 ±10.35 / 914.96 ms │     no change │
│ QQuery 24 │        412.75 / 416.86 ±3.52 / 421.75 ms │        412.14 / 415.00 ±3.25 / 421.01 ms │     no change │
│ QQuery 25 │        351.79 / 353.38 ±1.09 / 354.82 ms │        353.82 / 356.92 ±1.90 / 359.41 ms │     no change │
│ QQuery 26 │           82.37 / 83.97 ±1.54 / 86.43 ms │           82.06 / 84.75 ±2.22 / 88.44 ms │     no change │
│ QQuery 27 │        341.39 / 345.55 ±3.24 / 349.73 ms │        345.69 / 348.79 ±1.99 / 350.82 ms │     no change │
│ QQuery 28 │        149.76 / 150.73 ±1.03 / 152.24 ms │        148.98 / 149.93 ±0.81 / 151.17 ms │     no change │
│ QQuery 29 │        296.16 / 299.32 ±2.15 / 301.80 ms │        307.73 / 310.21 ±1.50 / 311.87 ms │     no change │
│ QQuery 30 │           42.56 / 44.78 ±2.49 / 49.44 ms │           44.33 / 46.70 ±2.11 / 49.99 ms │     no change │
│ QQuery 31 │        173.54 / 174.28 ±0.50 / 175.04 ms │        171.70 / 172.65 ±0.94 / 174.19 ms │     no change │
│ QQuery 32 │           56.33 / 58.22 ±1.24 / 59.51 ms │           57.04 / 57.46 ±0.40 / 58.03 ms │     no change │
│ QQuery 33 │        144.57 / 145.55 ±0.63 / 146.56 ms │        140.76 / 142.46 ±1.80 / 145.57 ms │     no change │
│ QQuery 34 │        109.24 / 110.47 ±0.62 / 110.87 ms │        105.98 / 107.65 ±1.17 / 109.43 ms │     no change │
│ QQuery 35 │        110.93 / 112.07 ±1.05 / 113.93 ms │        107.66 / 108.91 ±0.96 / 110.48 ms │     no change │
│ QQuery 36 │        218.80 / 226.24 ±4.40 / 232.01 ms │        216.81 / 218.53 ±1.47 / 220.10 ms │     no change │
│ QQuery 37 │        179.05 / 181.10 ±2.24 / 184.14 ms │        175.24 / 179.25 ±2.20 / 181.12 ms │     no change │
│ QQuery 38 │           86.64 / 89.76 ±3.58 / 96.51 ms │           84.62 / 87.21 ±2.49 / 91.57 ms │     no change │
│ QQuery 39 │        125.17 / 130.38 ±4.13 / 136.23 ms │        121.19 / 125.41 ±4.25 / 132.56 ms │     no change │
│ QQuery 40 │        112.06 / 118.62 ±6.04 / 128.19 ms │        110.82 / 118.34 ±5.72 / 127.03 ms │     no change │
│ QQuery 41 │           14.49 / 15.63 ±1.16 / 17.47 ms │           14.16 / 15.19 ±0.95 / 16.54 ms │     no change │
│ QQuery 42 │        107.46 / 109.49 ±1.36 / 111.18 ms │        106.56 / 109.02 ±1.71 / 111.81 ms │     no change │
│ QQuery 43 │           83.39 / 84.54 ±1.05 / 86.03 ms │           82.42 / 84.09 ±1.21 / 85.91 ms │     no change │
│ QQuery 44 │           11.43 / 12.37 ±0.79 / 13.36 ms │           11.88 / 12.39 ±0.61 / 13.57 ms │     no change │
│ QQuery 45 │           52.54 / 53.61 ±0.76 / 54.78 ms │           51.99 / 53.64 ±1.16 / 55.12 ms │     no change │
│ QQuery 46 │        229.50 / 232.42 ±1.56 / 234.16 ms │        227.06 / 230.21 ±2.28 / 233.13 ms │     no change │
│ QQuery 47 │        701.26 / 702.95 ±0.88 / 703.60 ms │       695.61 / 772.11 ±44.68 / 818.41 ms │  1.10x slower │
│ QQuery 48 │        284.35 / 288.43 ±2.91 / 291.97 ms │        293.76 / 294.94 ±1.26 / 297.20 ms │     no change │
│ QQuery 49 │        253.76 / 255.51 ±2.00 / 259.40 ms │        256.59 / 258.97 ±1.39 / 260.39 ms │     no change │
│ QQuery 50 │        233.39 / 236.03 ±2.03 / 238.92 ms │        224.88 / 234.81 ±8.24 / 249.41 ms │     no change │
│ QQuery 51 │        186.84 / 189.28 ±2.80 / 194.73 ms │        180.17 / 182.78 ±2.46 / 186.55 ms │     no change │
│ QQuery 52 │        109.97 / 110.99 ±1.31 / 113.56 ms │        107.11 / 108.76 ±1.69 / 111.63 ms │     no change │
│ QQuery 53 │        104.00 / 104.90 ±0.85 / 106.17 ms │        102.40 / 103.38 ±0.92 / 104.80 ms │     no change │
│ QQuery 54 │        149.10 / 150.52 ±1.50 / 153.16 ms │        146.12 / 147.89 ±0.97 / 148.87 ms │     no change │
│ QQuery 55 │        107.42 / 108.15 ±0.57 / 108.90 ms │        106.68 / 107.23 ±0.36 / 107.70 ms │     no change │
│ QQuery 56 │        142.89 / 143.95 ±0.84 / 145.23 ms │        138.52 / 141.26 ±1.87 / 144.29 ms │     no change │
│ QQuery 57 │        170.91 / 174.97 ±2.48 / 177.72 ms │        173.91 / 174.95 ±0.96 / 176.14 ms │     no change │
│ QQuery 58 │        288.66 / 295.94 ±6.06 / 305.80 ms │        289.08 / 297.96 ±5.10 / 304.79 ms │     no change │
│ QQuery 59 │        199.51 / 201.22 ±1.66 / 203.90 ms │        198.28 / 199.43 ±1.22 / 201.57 ms │     no change │
│ QQuery 60 │        143.80 / 145.14 ±0.87 / 146.50 ms │        142.02 / 143.34 ±1.14 / 144.93 ms │     no change │
│ QQuery 61 │        171.85 / 172.45 ±0.39 / 172.92 ms │        171.39 / 174.66 ±2.89 / 178.51 ms │     no change │
│ QQuery 62 │       886.52 / 926.93 ±28.08 / 961.76 ms │      913.12 / 946.68 ±35.15 / 1014.60 ms │     no change │
│ QQuery 63 │        103.94 / 109.33 ±5.67 / 119.86 ms │        103.80 / 105.50 ±1.44 / 108.13 ms │     no change │
│ QQuery 64 │       706.41 / 728.95 ±13.62 / 744.85 ms │        693.85 / 697.07 ±1.95 / 699.75 ms │     no change │
│ QQuery 65 │        256.75 / 262.97 ±3.69 / 267.35 ms │        250.35 / 253.77 ±5.04 / 263.71 ms │     no change │
│ QQuery 66 │       254.93 / 268.25 ±11.49 / 288.01 ms │        244.83 / 253.12 ±6.47 / 264.51 ms │ +1.06x faster │
│ QQuery 67 │        324.63 / 329.20 ±3.65 / 335.18 ms │        312.07 / 315.60 ±3.38 / 321.35 ms │     no change │
│ QQuery 68 │        288.09 / 291.17 ±1.93 / 293.42 ms │        278.82 / 285.79 ±4.05 / 290.92 ms │     no change │
│ QQuery 69 │        103.16 / 104.51 ±0.85 / 105.32 ms │        103.68 / 104.79 ±0.84 / 105.93 ms │     no change │
│ QQuery 70 │        354.47 / 361.88 ±9.34 / 378.45 ms │       324.04 / 341.71 ±17.02 / 371.16 ms │ +1.06x faster │
│ QQuery 71 │        134.06 / 136.63 ±1.77 / 138.62 ms │        136.51 / 140.11 ±2.43 / 143.21 ms │     no change │
│ QQuery 72 │       704.07 / 719.59 ±12.37 / 737.72 ms │        710.33 / 721.54 ±8.43 / 733.60 ms │     no change │
│ QQuery 73 │        102.09 / 105.87 ±2.94 / 110.69 ms │        102.02 / 104.28 ±1.83 / 107.18 ms │     no change │
│ QQuery 74 │       624.95 / 645.40 ±16.03 / 671.24 ms │        583.91 / 595.43 ±7.53 / 603.72 ms │ +1.08x faster │
│ QQuery 75 │        279.42 / 283.48 ±2.77 / 287.18 ms │        276.61 / 279.12 ±2.38 / 282.94 ms │     no change │
│ QQuery 76 │        132.01 / 134.24 ±1.75 / 137.36 ms │        131.86 / 132.99 ±0.87 / 134.29 ms │     no change │
│ QQuery 77 │        188.37 / 190.72 ±2.04 / 193.26 ms │        185.49 / 188.88 ±2.37 / 191.59 ms │     no change │
│ QQuery 78 │        348.72 / 353.01 ±4.86 / 362.11 ms │        351.26 / 354.08 ±1.58 / 355.52 ms │     no change │
│ QQuery 79 │        232.29 / 236.41 ±3.38 / 241.53 ms │        232.32 / 235.58 ±2.19 / 239.14 ms │     no change │
│ QQuery 80 │        329.31 / 333.45 ±2.31 / 335.94 ms │        325.20 / 327.97 ±2.73 / 332.83 ms │     no change │
│ QQuery 81 │           26.59 / 27.36 ±0.93 / 29.12 ms │           27.92 / 28.89 ±0.84 / 30.40 ms │  1.06x slower │
│ QQuery 82 │        198.54 / 200.07 ±1.72 / 203.42 ms │        200.84 / 203.22 ±2.63 / 206.64 ms │     no change │
│ QQuery 83 │           39.12 / 41.56 ±1.46 / 43.28 ms │           40.66 / 41.54 ±1.01 / 43.43 ms │     no change │
│ QQuery 84 │           48.49 / 49.57 ±1.30 / 52.11 ms │           49.81 / 50.45 ±0.83 / 51.92 ms │     no change │
│ QQuery 85 │        148.07 / 149.78 ±1.66 / 152.53 ms │        149.53 / 151.19 ±1.64 / 153.84 ms │     no change │
│ QQuery 86 │           39.84 / 41.36 ±1.81 / 44.69 ms │           38.64 / 39.95 ±0.90 / 41.28 ms │     no change │
│ QQuery 87 │           87.80 / 91.67 ±2.97 / 96.43 ms │           87.27 / 91.38 ±2.42 / 93.81 ms │     no change │
│ QQuery 88 │        102.27 / 103.04 ±0.61 / 103.71 ms │         99.91 / 101.16 ±1.17 / 103.11 ms │     no change │
│ QQuery 89 │        120.71 / 121.50 ±0.47 / 122.08 ms │        118.64 / 119.46 ±0.87 / 120.67 ms │     no change │
│ QQuery 90 │           23.91 / 24.93 ±0.63 / 25.87 ms │           23.50 / 23.79 ±0.17 / 24.02 ms │     no change │
│ QQuery 91 │           64.09 / 66.49 ±1.50 / 68.26 ms │           62.84 / 64.56 ±1.52 / 67.36 ms │     no change │
│ QQuery 92 │           59.12 / 60.23 ±0.65 / 60.95 ms │           56.52 / 57.02 ±0.37 / 57.61 ms │ +1.06x faster │
│ QQuery 93 │        199.52 / 203.04 ±3.02 / 207.76 ms │        191.90 / 193.08 ±1.22 / 195.10 ms │     no change │
│ QQuery 94 │           61.91 / 63.56 ±1.29 / 65.33 ms │           59.32 / 60.73 ±1.07 / 62.54 ms │     no change │
│ QQuery 95 │        135.16 / 137.98 ±1.64 / 140.03 ms │        132.26 / 134.58 ±1.47 / 136.50 ms │     no change │
│ QQuery 96 │           74.26 / 75.05 ±0.71 / 76.21 ms │           72.57 / 73.47 ±0.61 / 74.10 ms │     no change │
│ QQuery 97 │        131.22 / 132.80 ±0.87 / 133.76 ms │        126.73 / 131.00 ±3.15 / 134.70 ms │     no change │
│ QQuery 98 │        151.80 / 154.23 ±1.48 / 155.69 ms │        153.95 / 155.33 ±1.18 / 157.38 ms │     no change │
│ QQuery 99 │ 10712.41 / 10761.77 ±37.12 / 10814.96 ms │ 10733.24 / 10760.75 ±17.97 / 10779.76 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 33988.86ms │
│ Total Time (worktree-mutable-binary-eval)   │ 33744.26ms │
│ Average Time (HEAD)                         │   343.32ms │
│ Average Time (worktree-mutable-binary-eval) │   340.85ms │
│ Queries Faster                              │          5 │
│ Queries Slower                              │          3 │
│ Queries with No Change                      │         91 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 170.3s
Peak memory 4.7 GiB
Avg memory 4.0 GiB
CPU user 274.3s
CPU sys 18.2s
Disk read 0 B
Disk write 639.4 MiB

tpcds — branch

Metric Value
Wall time 169.0s
Peak memory 5.2 GiB
Avg memory 4.4 GiB
CPU user 272.1s
CPU sys 17.3s
Disk read 0 B
Disk write 784.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.29 / 4.55 ±6.40 / 17.34 ms │          1.27 / 4.51 ±6.37 / 17.25 ms │     no change │
│ QQuery 1  │        14.32 / 14.78 ±0.26 / 15.04 ms │        14.38 / 14.63 ±0.15 / 14.80 ms │     no change │
│ QQuery 2  │        44.15 / 44.53 ±0.24 / 44.78 ms │        44.44 / 44.58 ±0.15 / 44.85 ms │     no change │
│ QQuery 3  │        44.20 / 46.19 ±1.11 / 47.33 ms │        44.81 / 46.17 ±0.77 / 47.20 ms │     no change │
│ QQuery 4  │    287.95 / 300.58 ±11.04 / 314.56 ms │     290.81 / 301.54 ±8.96 / 315.33 ms │     no change │
│ QQuery 5  │     337.94 / 346.26 ±6.06 / 354.46 ms │     350.30 / 351.84 ±1.13 / 353.57 ms │     no change │
│ QQuery 6  │          6.34 / 8.50 ±1.80 / 11.33 ms │           5.67 / 6.43 ±0.76 / 7.51 ms │ +1.32x faster │
│ QQuery 7  │        16.56 / 17.11 ±0.37 / 17.62 ms │        17.02 / 17.57 ±0.82 / 19.20 ms │     no change │
│ QQuery 8  │     413.98 / 421.89 ±8.42 / 438.09 ms │    437.82 / 447.21 ±12.33 / 471.54 ms │  1.06x slower │
│ QQuery 9  │     627.21 / 645.85 ±9.89 / 654.42 ms │     648.46 / 653.32 ±3.69 / 658.56 ms │     no change │
│ QQuery 10 │       91.50 / 94.34 ±3.62 / 101.38 ms │      96.57 / 100.84 ±5.50 / 111.12 ms │  1.07x slower │
│ QQuery 11 │     105.57 / 107.44 ±1.31 / 108.96 ms │     107.98 / 109.58 ±0.96 / 110.84 ms │     no change │
│ QQuery 12 │     347.63 / 353.85 ±6.60 / 362.78 ms │     352.66 / 353.62 ±1.20 / 355.97 ms │     no change │
│ QQuery 13 │     457.74 / 462.94 ±7.07 / 476.82 ms │    455.51 / 477.47 ±17.78 / 500.47 ms │     no change │
│ QQuery 14 │     356.96 / 361.39 ±3.09 / 365.74 ms │     354.63 / 357.03 ±1.90 / 359.19 ms │     no change │
│ QQuery 15 │     367.74 / 370.40 ±3.63 / 377.58 ms │    362.61 / 384.70 ±20.57 / 422.37 ms │     no change │
│ QQuery 16 │    729.91 / 754.63 ±23.33 / 792.53 ms │    743.18 / 784.01 ±42.41 / 848.81 ms │     no change │
│ QQuery 17 │     718.69 / 735.57 ±9.62 / 746.35 ms │    751.22 / 766.74 ±16.08 / 792.35 ms │     no change │
│ QQuery 18 │ 1373.96 / 1408.53 ±29.46 / 1447.54 ms │ 1484.42 / 1552.36 ±38.03 / 1598.55 ms │  1.10x slower │
│ QQuery 19 │        35.39 / 36.99 ±1.11 / 38.03 ms │       38.61 / 59.50 ±24.24 / 91.02 ms │  1.61x slower │
│ QQuery 20 │    714.90 / 726.82 ±12.18 / 741.82 ms │    726.44 / 746.15 ±27.04 / 797.92 ms │     no change │
│ QQuery 21 │     765.06 / 766.56 ±1.56 / 769.52 ms │     768.19 / 773.23 ±4.08 / 778.93 ms │     no change │
│ QQuery 22 │ 1136.79 / 1148.38 ±10.47 / 1167.90 ms │  1144.31 / 1146.98 ±2.09 / 1150.56 ms │     no change │
│ QQuery 23 │ 3076.68 / 3108.83 ±21.23 / 3141.66 ms │ 3129.61 / 3145.89 ±10.88 / 3163.00 ms │     no change │
│ QQuery 24 │      99.21 / 102.91 ±3.71 / 108.88 ms │     102.61 / 106.26 ±3.21 / 111.67 ms │     no change │
│ QQuery 25 │     140.37 / 142.47 ±2.05 / 146.17 ms │     143.62 / 144.27 ±0.69 / 145.48 ms │     no change │
│ QQuery 26 │      99.85 / 102.35 ±2.04 / 105.09 ms │     103.15 / 104.92 ±1.40 / 106.76 ms │     no change │
│ QQuery 27 │     852.38 / 858.03 ±4.79 / 864.30 ms │     854.99 / 862.18 ±4.56 / 868.38 ms │     no change │
│ QQuery 28 │ 7695.39 / 7773.63 ±40.84 / 7814.22 ms │ 7738.48 / 7792.91 ±30.11 / 7821.89 ms │     no change │
│ QQuery 29 │        52.00 / 54.46 ±2.50 / 59.16 ms │        52.11 / 56.06 ±3.21 / 60.16 ms │     no change │
│ QQuery 30 │     367.40 / 373.93 ±4.37 / 380.78 ms │     362.33 / 371.39 ±6.60 / 381.34 ms │     no change │
│ QQuery 31 │     376.73 / 382.91 ±4.82 / 390.44 ms │     377.17 / 389.31 ±8.01 / 396.93 ms │     no change │
│ QQuery 32 │  1049.02 / 1060.30 ±9.07 / 1071.83 ms │ 1046.76 / 1062.22 ±14.92 / 1086.54 ms │     no change │
│ QQuery 33 │ 1466.36 / 1506.85 ±54.13 / 1613.81 ms │ 1475.61 / 1495.88 ±22.06 / 1537.81 ms │     no change │
│ QQuery 34 │ 1469.75 / 1516.25 ±29.52 / 1550.34 ms │ 1482.64 / 1494.83 ±11.92 / 1515.36 ms │     no change │
│ QQuery 35 │     388.03 / 392.55 ±2.68 / 395.11 ms │     402.05 / 404.13 ±1.68 / 406.61 ms │     no change │
│ QQuery 36 │     117.61 / 123.25 ±4.47 / 128.85 ms │     126.00 / 128.04 ±1.72 / 130.39 ms │     no change │
│ QQuery 37 │        47.66 / 49.49 ±1.50 / 51.53 ms │        52.02 / 52.77 ±0.52 / 53.59 ms │  1.07x slower │
│ QQuery 38 │        75.62 / 77.92 ±1.71 / 80.49 ms │        77.88 / 78.64 ±0.57 / 79.55 ms │     no change │
│ QQuery 39 │     215.48 / 223.77 ±6.71 / 231.71 ms │     215.80 / 226.85 ±7.46 / 237.37 ms │     no change │
│ QQuery 40 │        24.52 / 25.85 ±0.74 / 26.66 ms │        24.80 / 27.79 ±2.06 / 31.13 ms │  1.08x slower │
│ QQuery 41 │        20.00 / 20.79 ±0.59 / 21.61 ms │        21.36 / 22.87 ±0.97 / 23.97 ms │  1.10x slower │
│ QQuery 42 │        18.99 / 19.56 ±0.37 / 20.02 ms │        19.96 / 20.49 ±0.38 / 21.10 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 27094.18ms │
│ Total Time (worktree-mutable-binary-eval)   │ 27487.72ms │
│ Average Time (HEAD)                         │   630.10ms │
│ Average Time (worktree-mutable-binary-eval) │   639.25ms │
│ Queries Faster                              │          1 │
│ Queries Slower                              │          7 │
│ Queries with No Change                      │         35 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 136.7s
Peak memory 41.1 GiB
Avg memory 33.5 GiB
CPU user 1289.9s
CPU sys 87.3s
Disk read 0 B
Disk write 5.4 GiB

clickbench_partitioned — branch

Metric Value
Wall time 138.6s
Peak memory 41.6 GiB
Avg memory 31.5 GiB
CPU user 1297.8s
CPU sys 95.4s
Disk read 0 B
Disk write 120.0 KiB

File an issue against this benchmark runner

@Dandandan
Copy link
Copy Markdown
Contributor Author

run benchmarks

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183404626-738-5r8k2 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (f70c9e7) to 1e93a67 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183404626-739-p2w7s 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (f70c9e7) to 1e93a67 (merge-base) diff using: tpcds
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183404626-740-ch72f 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (f70c9e7) to 1e93a67 (merge-base) diff using: tpch
Results will be posted here when complete


File an issue against this benchmark runner

@alamb
Copy link
Copy Markdown
Contributor

alamb commented Apr 3, 2026

run benchmarks sql_planner

@adriangbot
Copy link
Copy Markdown

🤖 Criterion benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4183443930-741-59cbc 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected

Comparing worktree-mutable-binary-eval (f70c9e7) to 1e93a67 (merge-base) diff
BENCH_NAME=sql_planner
BENCH_COMMAND=cargo bench --features=parquet --bench sql_planner
BENCH_FILTER=
Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃   worktree-mutable-binary-eval ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │ 45.21 / 45.75 ±0.64 / 46.94 ms │ 45.59 / 46.45 ±1.00 / 48.10 ms │    no change │
│ QQuery 2  │ 21.07 / 21.52 ±0.49 / 22.41 ms │ 21.17 / 21.42 ±0.22 / 21.82 ms │    no change │
│ QQuery 3  │ 31.35 / 31.70 ±0.24 / 32.10 ms │ 31.83 / 32.06 ±0.14 / 32.26 ms │    no change │
│ QQuery 4  │ 20.18 / 20.75 ±0.68 / 22.04 ms │ 20.33 / 20.71 ±0.25 / 21.07 ms │    no change │
│ QQuery 5  │ 48.56 / 49.88 ±1.36 / 51.92 ms │ 48.14 / 51.14 ±2.00 / 54.06 ms │    no change │
│ QQuery 6  │ 17.01 / 17.16 ±0.13 / 17.38 ms │ 17.03 / 17.35 ±0.32 / 17.92 ms │    no change │
│ QQuery 7  │ 54.74 / 55.48 ±0.82 / 57.00 ms │ 55.22 / 58.82 ±2.33 / 61.72 ms │ 1.06x slower │
│ QQuery 8  │ 47.83 / 48.02 ±0.13 / 48.21 ms │ 47.78 / 48.35 ±0.44 / 49.06 ms │    no change │
│ QQuery 9  │ 53.52 / 55.14 ±0.90 / 56.03 ms │ 54.53 / 55.47 ±0.70 / 56.65 ms │    no change │
│ QQuery 10 │ 69.67 / 71.95 ±1.50 / 74.20 ms │ 71.24 / 72.42 ±0.95 / 73.99 ms │    no change │
│ QQuery 11 │ 13.89 / 14.19 ±0.33 / 14.84 ms │ 14.27 / 14.44 ±0.18 / 14.74 ms │    no change │
│ QQuery 12 │ 27.69 / 27.93 ±0.18 / 28.19 ms │ 27.50 / 27.96 ±0.32 / 28.36 ms │    no change │
│ QQuery 13 │ 38.26 / 39.35 ±0.64 / 40.08 ms │ 38.61 / 39.62 ±1.26 / 41.75 ms │    no change │
│ QQuery 14 │ 28.34 / 28.56 ±0.27 / 29.04 ms │ 28.51 / 28.78 ±0.25 / 29.24 ms │    no change │
│ QQuery 15 │ 33.11 / 33.79 ±0.48 / 34.30 ms │ 33.48 / 34.26 ±0.98 / 36.14 ms │    no change │
│ QQuery 16 │ 15.96 / 16.33 ±0.29 / 16.73 ms │ 15.88 / 16.52 ±0.40 / 16.87 ms │    no change │
│ QQuery 17 │ 73.35 / 74.21 ±1.18 / 76.48 ms │ 74.02 / 74.53 ±0.41 / 74.94 ms │    no change │
│ QQuery 18 │ 77.94 / 78.83 ±0.45 / 79.19 ms │ 78.34 / 79.28 ±0.99 / 80.60 ms │    no change │
│ QQuery 19 │ 36.93 / 37.34 ±0.44 / 38.03 ms │ 37.44 / 38.46 ±1.04 / 40.36 ms │    no change │
│ QQuery 20 │ 39.92 / 40.76 ±0.72 / 41.90 ms │ 41.13 / 41.65 ±0.48 / 42.51 ms │    no change │
│ QQuery 21 │ 64.27 / 65.03 ±0.76 / 66.46 ms │ 63.67 / 65.70 ±1.18 / 67.26 ms │    no change │
│ QQuery 22 │ 18.07 / 18.39 ±0.34 / 19.02 ms │ 18.00 / 18.26 ±0.28 / 18.75 ms │    no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                           ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 892.09ms │
│ Total Time (worktree-mutable-binary-eval)   │ 903.64ms │
│ Average Time (HEAD)                         │  40.55ms │
│ Average Time (worktree-mutable-binary-eval) │  41.07ms │
│ Queries Faster                              │        0 │
│ Queries Slower                              │        1 │
│ Queries with No Change                      │       21 │
│ Queries with Failure                        │        0 │
└─────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 4.7s
Peak memory 4.1 GiB
Avg memory 3.6 GiB
CPU user 33.6s
CPU sys 2.6s
Disk read 0 B
Disk write 136.0 KiB

tpch — branch

Metric Value
Wall time 4.8s
Peak memory 4.0 GiB
Avg memory 3.6 GiB
CPU user 33.8s
CPU sys 2.7s
Disk read 0 B
Disk write 60.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃             worktree-mutable-binary-eval ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │           43.21 / 43.67 ±0.66 / 44.96 ms │           43.10 / 44.08 ±0.92 / 45.82 ms │    no change │
│ QQuery 2  │        145.61 / 146.82 ±0.92 / 148.47 ms │        147.41 / 149.40 ±1.27 / 151.10 ms │    no change │
│ QQuery 3  │        113.53 / 114.36 ±0.69 / 115.34 ms │        113.79 / 114.59 ±0.74 / 115.51 ms │    no change │
│ QQuery 4  │    1352.36 / 1370.64 ±11.13 / 1383.23 ms │    1349.91 / 1370.38 ±14.87 / 1395.83 ms │    no change │
│ QQuery 5  │        173.07 / 174.03 ±0.68 / 175.14 ms │        170.47 / 172.59 ±1.60 / 174.11 ms │    no change │
│ QQuery 6  │    1001.69 / 1029.88 ±25.25 / 1071.42 ms │     997.36 / 1017.95 ±15.71 / 1032.64 ms │    no change │
│ QQuery 7  │        349.76 / 353.73 ±2.56 / 357.61 ms │        350.69 / 354.54 ±2.80 / 357.85 ms │    no change │
│ QQuery 8  │        116.08 / 116.99 ±0.61 / 117.83 ms │        116.88 / 117.90 ±0.85 / 119.44 ms │    no change │
│ QQuery 9  │        101.05 / 106.53 ±5.12 / 115.61 ms │        100.99 / 108.71 ±9.88 / 127.79 ms │    no change │
│ QQuery 10 │        106.21 / 107.35 ±1.12 / 109.47 ms │        106.92 / 107.66 ±0.51 / 108.43 ms │    no change │
│ QQuery 11 │        939.41 / 946.29 ±6.04 / 956.86 ms │        938.23 / 946.58 ±8.81 / 958.19 ms │    no change │
│ QQuery 12 │           46.83 / 47.76 ±0.72 / 48.82 ms │           44.68 / 46.41 ±1.92 / 50.01 ms │    no change │
│ QQuery 13 │        399.31 / 402.11 ±2.52 / 406.52 ms │        394.31 / 400.07 ±4.56 / 406.37 ms │    no change │
│ QQuery 14 │     1015.04 / 1025.85 ±6.93 / 1035.98 ms │     1022.11 / 1033.02 ±7.26 / 1043.17 ms │    no change │
│ QQuery 15 │           15.98 / 16.67 ±0.47 / 17.40 ms │           16.42 / 17.54 ±0.84 / 18.83 ms │ 1.05x slower │
│ QQuery 16 │           40.29 / 41.18 ±0.63 / 42.10 ms │           40.32 / 41.90 ±1.20 / 43.84 ms │    no change │
│ QQuery 17 │        241.95 / 242.56 ±0.87 / 244.26 ms │        240.32 / 242.19 ±0.96 / 242.92 ms │    no change │
│ QQuery 18 │        129.33 / 130.63 ±1.32 / 133.15 ms │        129.07 / 130.90 ±1.95 / 133.39 ms │    no change │
│ QQuery 19 │        155.56 / 156.24 ±0.43 / 156.75 ms │        156.22 / 156.91 ±0.92 / 158.62 ms │    no change │
│ QQuery 20 │           13.77 / 14.08 ±0.30 / 14.60 ms │           13.72 / 14.16 ±0.40 / 14.87 ms │    no change │
│ QQuery 21 │           19.27 / 19.93 ±0.44 / 20.57 ms │           19.17 / 19.70 ±0.44 / 20.41 ms │    no change │
│ QQuery 22 │        494.30 / 497.17 ±2.72 / 502.17 ms │        484.33 / 488.44 ±3.63 / 493.89 ms │    no change │
│ QQuery 23 │        889.51 / 903.05 ±7.58 / 911.63 ms │        902.12 / 908.48 ±6.68 / 920.58 ms │    no change │
│ QQuery 24 │        413.99 / 418.12 ±3.11 / 423.51 ms │        414.30 / 417.44 ±3.47 / 423.91 ms │    no change │
│ QQuery 25 │        353.78 / 356.73 ±1.67 / 358.80 ms │        353.37 / 355.84 ±1.90 / 359.02 ms │    no change │
│ QQuery 26 │           81.93 / 83.16 ±1.30 / 85.10 ms │           81.61 / 82.44 ±0.61 / 83.48 ms │    no change │
│ QQuery 27 │        351.10 / 352.19 ±1.14 / 354.37 ms │        345.86 / 349.11 ±2.24 / 352.30 ms │    no change │
│ QQuery 28 │        147.49 / 148.41 ±0.51 / 148.97 ms │        148.34 / 149.76 ±1.48 / 151.80 ms │    no change │
│ QQuery 29 │        299.05 / 301.24 ±1.42 / 302.81 ms │        298.38 / 301.42 ±1.95 / 304.25 ms │    no change │
│ QQuery 30 │           43.44 / 45.60 ±1.83 / 48.04 ms │           43.21 / 45.21 ±1.51 / 47.72 ms │    no change │
│ QQuery 31 │        172.15 / 173.38 ±1.35 / 175.79 ms │        169.54 / 172.00 ±1.28 / 173.25 ms │    no change │
│ QQuery 32 │           57.00 / 58.54 ±0.88 / 59.50 ms │           56.98 / 57.54 ±0.78 / 59.07 ms │    no change │
│ QQuery 33 │        141.22 / 143.00 ±1.29 / 144.58 ms │        139.25 / 141.58 ±1.75 / 144.49 ms │    no change │
│ QQuery 34 │        107.12 / 107.95 ±0.70 / 108.92 ms │        106.81 / 107.44 ±0.43 / 108.07 ms │    no change │
│ QQuery 35 │        105.83 / 108.39 ±1.77 / 111.22 ms │        107.62 / 109.63 ±1.40 / 112.02 ms │    no change │
│ QQuery 36 │        217.36 / 219.84 ±2.74 / 224.54 ms │        217.49 / 219.95 ±1.29 / 221.00 ms │    no change │
│ QQuery 37 │        174.67 / 176.24 ±1.50 / 178.81 ms │        176.07 / 177.67 ±1.29 / 179.59 ms │    no change │
│ QQuery 38 │           83.75 / 88.67 ±2.94 / 92.05 ms │           83.88 / 89.62 ±4.72 / 95.61 ms │    no change │
│ QQuery 39 │        127.42 / 129.79 ±1.87 / 132.04 ms │        126.09 / 130.35 ±2.29 / 132.69 ms │    no change │
│ QQuery 40 │        112.13 / 118.27 ±5.20 / 127.80 ms │        113.27 / 117.83 ±5.78 / 127.62 ms │    no change │
│ QQuery 41 │           14.53 / 15.87 ±0.89 / 17.25 ms │           14.15 / 15.82 ±0.94 / 16.81 ms │    no change │
│ QQuery 42 │        106.42 / 109.36 ±1.92 / 112.12 ms │        106.18 / 108.17 ±2.38 / 111.77 ms │    no change │
│ QQuery 43 │           82.69 / 83.87 ±0.78 / 84.82 ms │           83.59 / 84.03 ±0.49 / 84.98 ms │    no change │
│ QQuery 44 │           11.82 / 12.33 ±0.52 / 13.14 ms │           11.79 / 12.56 ±1.06 / 14.49 ms │    no change │
│ QQuery 45 │           52.10 / 53.31 ±0.97 / 54.47 ms │           52.50 / 53.62 ±0.84 / 54.93 ms │    no change │
│ QQuery 46 │        234.25 / 235.39 ±1.08 / 237.20 ms │        231.97 / 234.50 ±1.97 / 237.87 ms │    no change │
│ QQuery 47 │        700.34 / 705.23 ±4.02 / 710.12 ms │        715.06 / 720.70 ±5.15 / 728.79 ms │    no change │
│ QQuery 48 │        280.60 / 290.96 ±7.12 / 299.14 ms │        291.35 / 292.59 ±0.94 / 293.91 ms │    no change │
│ QQuery 49 │        254.81 / 256.22 ±1.46 / 258.13 ms │        251.93 / 255.15 ±3.27 / 260.64 ms │    no change │
│ QQuery 50 │        231.12 / 238.04 ±3.60 / 240.88 ms │        234.23 / 236.45 ±2.37 / 240.60 ms │    no change │
│ QQuery 51 │        182.94 / 185.39 ±2.39 / 189.71 ms │        184.73 / 187.31 ±1.74 / 189.39 ms │    no change │
│ QQuery 52 │        106.35 / 108.40 ±1.66 / 111.15 ms │        107.13 / 108.15 ±1.14 / 110.22 ms │    no change │
│ QQuery 53 │        102.00 / 103.84 ±1.88 / 106.87 ms │        101.98 / 103.31 ±1.49 / 105.93 ms │    no change │
│ QQuery 54 │        147.34 / 149.21 ±1.66 / 152.11 ms │        146.61 / 148.48 ±0.98 / 149.26 ms │    no change │
│ QQuery 55 │        106.88 / 107.72 ±0.54 / 108.18 ms │        106.41 / 107.82 ±1.18 / 109.80 ms │    no change │
│ QQuery 56 │        140.12 / 142.01 ±1.51 / 144.03 ms │        139.79 / 140.85 ±1.08 / 142.86 ms │    no change │
│ QQuery 57 │        174.30 / 175.69 ±1.43 / 177.93 ms │        174.99 / 177.37 ±1.95 / 179.62 ms │    no change │
│ QQuery 58 │        292.25 / 297.30 ±3.96 / 303.76 ms │        297.66 / 306.34 ±5.68 / 314.43 ms │    no change │
│ QQuery 59 │        198.82 / 200.28 ±1.01 / 201.67 ms │        202.52 / 203.17 ±0.53 / 204.13 ms │    no change │
│ QQuery 60 │        143.37 / 144.20 ±1.10 / 146.11 ms │        142.77 / 144.00 ±1.09 / 145.36 ms │    no change │
│ QQuery 61 │        170.96 / 172.52 ±1.38 / 174.83 ms │        168.86 / 170.68 ±1.78 / 173.92 ms │    no change │
│ QQuery 62 │       898.70 / 927.02 ±33.93 / 993.64 ms │       892.77 / 922.27 ±20.32 / 943.52 ms │    no change │
│ QQuery 63 │        103.22 / 107.19 ±3.36 / 112.74 ms │        104.81 / 105.70 ±1.02 / 107.65 ms │    no change │
│ QQuery 64 │        701.44 / 705.96 ±4.15 / 712.94 ms │        703.19 / 707.13 ±3.65 / 712.41 ms │    no change │
│ QQuery 65 │        250.67 / 255.90 ±4.15 / 263.34 ms │        249.09 / 252.21 ±2.19 / 254.56 ms │    no change │
│ QQuery 66 │        248.46 / 263.47 ±8.70 / 275.54 ms │        255.52 / 260.36 ±3.97 / 265.27 ms │    no change │
│ QQuery 67 │        305.42 / 316.44 ±7.44 / 324.52 ms │        309.78 / 319.93 ±8.36 / 334.32 ms │    no change │
│ QQuery 68 │        275.53 / 283.67 ±5.30 / 291.77 ms │        280.09 / 284.17 ±3.67 / 290.92 ms │    no change │
│ QQuery 69 │        101.51 / 102.46 ±0.87 / 103.59 ms │        102.06 / 103.09 ±0.95 / 104.20 ms │    no change │
│ QQuery 70 │        337.95 / 347.76 ±9.58 / 365.64 ms │       328.58 / 343.73 ±10.14 / 356.80 ms │    no change │
│ QQuery 71 │        134.36 / 136.02 ±1.41 / 137.82 ms │        135.64 / 136.86 ±1.27 / 138.77 ms │    no change │
│ QQuery 72 │        714.95 / 726.72 ±7.83 / 737.71 ms │       699.99 / 720.47 ±12.66 / 732.72 ms │    no change │
│ QQuery 73 │        102.07 / 104.20 ±2.27 / 107.63 ms │        100.82 / 104.95 ±2.94 / 109.68 ms │    no change │
│ QQuery 74 │        580.99 / 586.50 ±2.96 / 589.66 ms │        582.81 / 591.11 ±6.72 / 603.21 ms │    no change │
│ QQuery 75 │        275.47 / 278.58 ±3.16 / 284.53 ms │        277.34 / 280.46 ±2.42 / 284.71 ms │    no change │
│ QQuery 76 │        131.76 / 133.19 ±1.91 / 136.97 ms │        131.49 / 133.25 ±1.42 / 135.49 ms │    no change │
│ QQuery 77 │        187.21 / 190.31 ±1.85 / 192.54 ms │        186.83 / 189.74 ±1.69 / 191.72 ms │    no change │
│ QQuery 78 │        345.40 / 352.41 ±5.91 / 362.04 ms │        342.37 / 352.40 ±5.32 / 357.44 ms │    no change │
│ QQuery 79 │        234.99 / 236.14 ±0.92 / 237.74 ms │        230.85 / 233.91 ±2.33 / 236.62 ms │    no change │
│ QQuery 80 │        328.52 / 331.62 ±3.37 / 336.50 ms │        327.63 / 331.73 ±3.30 / 335.88 ms │    no change │
│ QQuery 81 │           26.27 / 27.57 ±0.83 / 28.75 ms │           26.99 / 28.22 ±1.33 / 30.67 ms │    no change │
│ QQuery 82 │        197.76 / 200.00 ±1.95 / 202.26 ms │        195.61 / 200.13 ±3.11 / 203.34 ms │    no change │
│ QQuery 83 │           40.32 / 41.12 ±0.73 / 42.46 ms │           40.17 / 40.65 ±0.57 / 41.73 ms │    no change │
│ QQuery 84 │           48.47 / 50.06 ±1.23 / 51.38 ms │           48.64 / 50.34 ±1.16 / 51.79 ms │    no change │
│ QQuery 85 │        148.97 / 149.87 ±0.67 / 150.93 ms │        147.95 / 150.56 ±1.67 / 153.04 ms │    no change │
│ QQuery 86 │           39.01 / 40.72 ±1.76 / 43.38 ms │           38.04 / 39.49 ±0.92 / 40.60 ms │    no change │
│ QQuery 87 │           85.40 / 88.99 ±3.32 / 94.93 ms │           85.31 / 88.08 ±3.64 / 95.23 ms │    no change │
│ QQuery 88 │          98.85 / 99.67 ±0.77 / 100.85 ms │         99.95 / 100.54 ±0.61 / 101.67 ms │    no change │
│ QQuery 89 │        119.19 / 120.05 ±0.58 / 120.82 ms │        117.25 / 119.12 ±1.38 / 120.69 ms │    no change │
│ QQuery 90 │           22.93 / 23.71 ±0.55 / 24.36 ms │           23.79 / 24.33 ±0.31 / 24.71 ms │    no change │
│ QQuery 91 │           63.87 / 64.73 ±0.81 / 66.08 ms │           63.33 / 65.11 ±1.25 / 66.92 ms │    no change │
│ QQuery 92 │           56.69 / 58.62 ±1.36 / 60.40 ms │           57.39 / 58.30 ±0.84 / 59.37 ms │    no change │
│ QQuery 93 │        192.78 / 195.11 ±3.01 / 200.85 ms │        193.29 / 195.46 ±1.93 / 198.05 ms │    no change │
│ QQuery 94 │           60.84 / 61.78 ±0.76 / 62.94 ms │           60.57 / 61.53 ±0.54 / 62.24 ms │    no change │
│ QQuery 95 │        135.25 / 136.59 ±1.57 / 139.44 ms │        134.46 / 137.20 ±1.49 / 138.47 ms │    no change │
│ QQuery 96 │           72.90 / 73.91 ±0.59 / 74.54 ms │           70.05 / 73.25 ±1.65 / 74.61 ms │    no change │
│ QQuery 97 │        129.24 / 131.25 ±1.65 / 133.34 ms │        129.95 / 132.43 ±1.68 / 134.48 ms │    no change │
│ QQuery 98 │        153.89 / 155.19 ±0.98 / 156.48 ms │        151.20 / 154.24 ±2.00 / 156.43 ms │    no change │
│ QQuery 99 │ 10749.03 / 10816.22 ±42.13 / 10864.13 ms │ 10741.82 / 10783.74 ±27.72 / 10815.72 ms │    no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 33744.85ms │
│ Total Time (worktree-mutable-binary-eval)   │ 33716.21ms │
│ Average Time (HEAD)                         │   340.86ms │
│ Average Time (worktree-mutable-binary-eval) │   340.57ms │
│ Queries Faster                              │          0 │
│ Queries Slower                              │          1 │
│ Queries with No Change                      │         98 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 169.1s
Peak memory 5.3 GiB
Avg memory 4.4 GiB
CPU user 270.6s
CPU sys 18.3s
Disk read 0 B
Disk write 638.2 MiB

tpcds — branch

Metric Value
Wall time 168.9s
Peak memory 5.7 GiB
Avg memory 4.7 GiB
CPU user 271.8s
CPU sys 17.8s
Disk read 0 B
Disk write 164.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

Comparing HEAD and worktree-mutable-binary-eval
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃          worktree-mutable-binary-eval ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.35 / 4.73 ±6.58 / 17.90 ms │          1.28 / 4.51 ±6.37 / 17.25 ms │     no change │
│ QQuery 1  │        14.62 / 15.31 ±0.36 / 15.60 ms │        14.49 / 14.69 ±0.11 / 14.84 ms │     no change │
│ QQuery 2  │        44.35 / 44.78 ±0.30 / 45.05 ms │        43.32 / 43.93 ±0.36 / 44.32 ms │     no change │
│ QQuery 3  │        45.16 / 46.70 ±1.25 / 48.88 ms │        42.91 / 45.33 ±1.37 / 46.74 ms │     no change │
│ QQuery 4  │    330.27 / 342.98 ±10.62 / 357.17 ms │    301.54 / 313.26 ±11.50 / 332.14 ms │ +1.09x faster │
│ QQuery 5  │    345.11 / 364.08 ±18.78 / 393.31 ms │    346.57 / 361.82 ±16.36 / 387.56 ms │     no change │
│ QQuery 6  │           6.03 / 6.74 ±0.63 / 7.65 ms │           6.64 / 7.62 ±1.02 / 9.55 ms │  1.13x slower │
│ QQuery 7  │        16.89 / 18.07 ±1.77 / 21.57 ms │        16.48 / 17.09 ±0.78 / 18.62 ms │ +1.06x faster │
│ QQuery 8  │     459.95 / 465.79 ±5.67 / 475.07 ms │    429.07 / 448.13 ±16.78 / 468.54 ms │     no change │
│ QQuery 9  │    656.22 / 690.31 ±26.44 / 724.44 ms │    641.11 / 673.32 ±27.79 / 713.48 ms │     no change │
│ QQuery 10 │       91.70 / 97.02 ±4.25 / 103.83 ms │       91.86 / 97.81 ±6.05 / 108.64 ms │     no change │
│ QQuery 11 │     103.98 / 109.59 ±4.03 / 114.38 ms │     105.49 / 107.61 ±2.77 / 112.99 ms │     no change │
│ QQuery 12 │    348.86 / 371.78 ±27.08 / 424.16 ms │    342.89 / 360.17 ±14.95 / 381.03 ms │     no change │
│ QQuery 13 │     491.95 / 502.15 ±6.07 / 509.11 ms │    452.34 / 484.31 ±20.11 / 508.94 ms │     no change │
│ QQuery 14 │    351.06 / 367.66 ±15.24 / 396.40 ms │    351.35 / 363.95 ±13.97 / 387.73 ms │     no change │
│ QQuery 15 │    358.43 / 395.27 ±29.00 / 430.24 ms │    373.15 / 403.14 ±16.10 / 420.29 ms │     no change │
│ QQuery 16 │     759.00 / 770.55 ±8.82 / 784.05 ms │    753.81 / 771.97 ±21.51 / 813.38 ms │     no change │
│ QQuery 17 │    714.74 / 782.61 ±45.45 / 842.07 ms │    716.72 / 737.87 ±15.44 / 764.05 ms │ +1.06x faster │
│ QQuery 18 │ 1492.53 / 1530.59 ±40.15 / 1598.99 ms │ 1483.99 / 1550.28 ±45.86 / 1613.06 ms │     no change │
│ QQuery 19 │       36.71 / 57.93 ±25.47 / 94.55 ms │       40.01 / 56.82 ±21.05 / 91.29 ms │     no change │
│ QQuery 20 │    718.63 / 753.84 ±32.17 / 806.63 ms │    731.32 / 747.89 ±18.09 / 781.98 ms │     no change │
│ QQuery 21 │     763.56 / 774.12 ±7.77 / 787.75 ms │    761.18 / 780.97 ±13.01 / 794.23 ms │     no change │
│ QQuery 22 │ 1141.67 / 1164.96 ±12.72 / 1177.99 ms │ 1154.35 / 1166.88 ±13.30 / 1192.64 ms │     no change │
│ QQuery 23 │ 3135.67 / 3212.62 ±53.20 / 3280.13 ms │ 3141.32 / 3167.74 ±27.90 / 3221.24 ms │     no change │
│ QQuery 24 │      96.73 / 101.93 ±3.19 / 105.45 ms │      98.50 / 105.28 ±6.90 / 118.16 ms │     no change │
│ QQuery 25 │     138.20 / 140.74 ±1.82 / 142.68 ms │     137.41 / 141.54 ±3.03 / 145.65 ms │     no change │
│ QQuery 26 │     104.10 / 106.07 ±1.24 / 107.75 ms │     101.14 / 105.71 ±3.51 / 110.42 ms │     no change │
│ QQuery 27 │     861.83 / 867.33 ±4.33 / 873.63 ms │     843.63 / 859.43 ±8.26 / 867.87 ms │     no change │
│ QQuery 28 │ 7682.52 / 7801.38 ±63.15 / 7869.80 ms │ 7744.88 / 7828.80 ±45.04 / 7867.99 ms │     no change │
│ QQuery 29 │      51.89 / 99.82 ±71.23 / 241.47 ms │        50.98 / 54.69 ±5.06 / 64.65 ms │ +1.83x faster │
│ QQuery 30 │    364.59 / 383.04 ±10.38 / 396.27 ms │    370.24 / 388.32 ±15.90 / 411.27 ms │     no change │
│ QQuery 31 │    394.54 / 414.86 ±11.17 / 426.33 ms │    364.12 / 380.93 ±11.70 / 398.35 ms │ +1.09x faster │
│ QQuery 32 │ 1395.31 / 1439.06 ±32.48 / 1478.98 ms │ 1068.56 / 1095.86 ±28.76 / 1137.71 ms │ +1.31x faster │
│ QQuery 33 │ 1522.57 / 1570.83 ±39.77 / 1637.71 ms │ 1518.09 / 1540.96 ±14.76 / 1556.89 ms │     no change │
│ QQuery 34 │ 1501.15 / 1531.04 ±26.50 / 1570.54 ms │ 1516.50 / 1530.57 ±15.92 / 1560.49 ms │     no change │
│ QQuery 35 │    396.87 / 447.91 ±32.42 / 496.46 ms │    386.65 / 411.08 ±23.26 / 447.62 ms │ +1.09x faster │
│ QQuery 36 │     120.21 / 122.52 ±1.85 / 125.62 ms │     116.72 / 126.16 ±7.19 / 138.42 ms │     no change │
│ QQuery 37 │        51.30 / 53.46 ±2.10 / 56.45 ms │        48.70 / 49.69 ±1.01 / 51.34 ms │ +1.08x faster │
│ QQuery 38 │        76.88 / 78.59 ±1.20 / 79.87 ms │        77.44 / 79.26 ±1.23 / 80.40 ms │     no change │
│ QQuery 39 │    215.60 / 229.15 ±13.23 / 247.19 ms │    218.73 / 230.28 ±12.85 / 254.20 ms │     no change │
│ QQuery 40 │        21.60 / 25.43 ±2.27 / 27.94 ms │        25.86 / 26.80 ±0.97 / 28.39 ms │  1.05x slower │
│ QQuery 41 │        19.72 / 20.91 ±0.66 / 21.46 ms │        19.21 / 21.70 ±2.29 / 26.02 ms │     no change │
│ QQuery 42 │        20.22 / 20.92 ±0.78 / 22.20 ms │        19.24 / 19.75 ±0.34 / 20.22 ms │ +1.06x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                           ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                           │ 28345.17ms │
│ Total Time (worktree-mutable-binary-eval)   │ 27723.91ms │
│ Average Time (HEAD)                         │   659.19ms │
│ Average Time (worktree-mutable-binary-eval) │   644.74ms │
│ Queries Faster                              │          9 │
│ Queries Slower                              │          2 │
│ Queries with No Change                      │         32 │
│ Queries with Failure                        │          0 │
└─────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 143.0s
Peak memory 40.5 GiB
Avg memory 29.5 GiB
CPU user 1324.1s
CPU sys 109.0s
Disk read 0 B
Disk write 3.7 GiB

clickbench_partitioned — branch

Metric Value
Wall time 139.8s
Peak memory 42.5 GiB
Avg memory 33.3 GiB
CPU user 1314.5s
CPU sys 92.7s
Disk read 0 B
Disk write 120.0 KiB

File an issue against this benchmark runner

@adriangbot
Copy link
Copy Markdown

🤖 Criterion benchmark completed (GKE) | trigger

Instance: c4a-highmem-16 (12 vCPU / 65 GiB)

CPU Details (lscpu)
Architecture:                            aarch64
CPU op-mode(s):                          64-bit
Byte Order:                              Little Endian
CPU(s):                                  16
On-line CPU(s) list:                     0-15
Vendor ID:                               ARM
Model name:                              Neoverse-V2
Model:                                   1
Thread(s) per core:                      1
Core(s) per cluster:                     16
Socket(s):                               -
Cluster(s):                              1
Stepping:                                r0p1
BogoMIPS:                                2000.00
Flags:                                   fp asimd evtstrm aes pmull sha1 sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm bf16 dgh rng bti
L1d cache:                               1 MiB (16 instances)
L1i cache:                               1 MiB (16 instances)
L2 cache:                                32 MiB (16 instances)
L3 cache:                                80 MiB (1 instance)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-15
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit:             Not affected
Vulnerability L1tf:                      Not affected
Vulnerability Mds:                       Not affected
Vulnerability Meltdown:                  Not affected
Vulnerability Mmio stale data:           Not affected
Vulnerability Reg file data sampling:    Not affected
Vulnerability Retbleed:                  Not affected
Vulnerability Spec rstack overflow:      Not affected
Vulnerability Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1:                Mitigation; __user pointer sanitization
Vulnerability Spectre v2:                Mitigation; CSV2, BHB
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Not affected
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Not affected
Details

group                                                 main                                    worktree-mutable-binary-eval
-----                                                 ----                                    ----------------------------
logical_aggregate_with_join                           1.00    439.4±1.73µs        ? ?/sec     1.00    439.1±0.83µs        ? ?/sec
logical_plan_struct_join_agg_sort                     1.00    168.1±1.57µs        ? ?/sec     1.03    172.9±0.76µs        ? ?/sec
logical_select_all_from_1000                          1.01      7.9±0.04ms        ? ?/sec     1.00      7.9±0.03ms        ? ?/sec
logical_select_one_from_700                           1.01    322.0±1.30µs        ? ?/sec     1.00    318.8±1.12µs        ? ?/sec
logical_trivial_join_high_numbered_columns            1.01    266.9±1.48µs        ? ?/sec     1.00    264.4±0.77µs        ? ?/sec
logical_trivial_join_low_numbered_columns             1.01    254.4±0.66µs        ? ?/sec     1.00    251.9±0.71µs        ? ?/sec
physical_intersection                                 1.03    596.0±3.33µs        ? ?/sec     1.00    580.7±1.28µs        ? ?/sec
physical_join_consider_sort                           1.03  1046.3±15.07µs        ? ?/sec     1.00   1018.3±6.58µs        ? ?/sec
physical_join_distinct                                1.01    249.3±0.72µs        ? ?/sec     1.00    247.3±1.37µs        ? ?/sec
physical_many_self_joins                              1.02      7.7±0.10ms        ? ?/sec     1.00      7.5±0.08ms        ? ?/sec
physical_plan_clickbench_all                          1.00    113.4±6.11ms        ? ?/sec     1.01    114.7±7.99ms        ? ?/sec
physical_plan_clickbench_q1                           1.00  1123.0±38.90µs        ? ?/sec     1.01  1130.8±51.39µs        ? ?/sec
physical_plan_clickbench_q10                          1.00  1815.6±119.66µs        ? ?/sec    1.01  1838.1±122.18µs        ? ?/sec
physical_plan_clickbench_q11                          1.00  1935.1±101.76µs        ? ?/sec    1.02  1966.4±143.04µs        ? ?/sec
physical_plan_clickbench_q12                          1.00      2.0±0.14ms        ? ?/sec     1.00      2.0±0.14ms        ? ?/sec
physical_plan_clickbench_q13                          1.05  1845.6±133.67µs        ? ?/sec    1.00  1762.9±97.35µs        ? ?/sec
physical_plan_clickbench_q14                          1.02  1968.2±128.02µs        ? ?/sec    1.00  1936.8±128.85µs        ? ?/sec
physical_plan_clickbench_q15                          1.01  1850.0±112.42µs        ? ?/sec    1.00  1824.1±90.61µs        ? ?/sec
physical_plan_clickbench_q16                          1.04  1559.5±108.63µs        ? ?/sec    1.00  1496.7±55.87µs        ? ?/sec
physical_plan_clickbench_q17                          1.01  1580.0±77.43µs        ? ?/sec     1.00  1566.4±71.30µs        ? ?/sec
physical_plan_clickbench_q18                          1.00  1393.6±58.57µs        ? ?/sec     1.02  1424.9±83.13µs        ? ?/sec
physical_plan_clickbench_q19                          1.00  1784.0±91.60µs        ? ?/sec     1.02  1824.1±114.51µs        ? ?/sec
physical_plan_clickbench_q2                           1.00  1516.4±120.79µs        ? ?/sec    1.00  1517.1±103.26µs        ? ?/sec
physical_plan_clickbench_q20                          1.00  1216.6±52.53µs        ? ?/sec     1.01  1224.0±51.50µs        ? ?/sec
physical_plan_clickbench_q21                          1.00  1509.1±100.68µs        ? ?/sec    1.00  1511.7±93.30µs        ? ?/sec
physical_plan_clickbench_q22                          1.01  1968.1±144.45µs        ? ?/sec    1.00  1949.8±123.53µs        ? ?/sec
physical_plan_clickbench_q23                          1.00      2.1±0.13ms        ? ?/sec     1.02      2.2±0.15ms        ? ?/sec
physical_plan_clickbench_q24                          1.00      3.1±0.15ms        ? ?/sec     1.00      3.1±0.13ms        ? ?/sec
physical_plan_clickbench_q25                          1.00  1600.5±89.44µs        ? ?/sec     1.00  1598.1±72.73µs        ? ?/sec
physical_plan_clickbench_q26                          1.00  1436.8±80.45µs        ? ?/sec     1.00  1430.4±66.78µs        ? ?/sec
physical_plan_clickbench_q27                          1.02  1639.1±102.18µs        ? ?/sec    1.00  1600.5±69.17µs        ? ?/sec
physical_plan_clickbench_q28                          1.03      2.1±0.16ms        ? ?/sec     1.00      2.1±0.11ms        ? ?/sec
physical_plan_clickbench_q29                          1.02      2.3±0.15ms        ? ?/sec     1.00      2.3±0.12ms        ? ?/sec
physical_plan_clickbench_q3                           1.02  1430.7±89.66µs        ? ?/sec     1.00  1404.8±78.64µs        ? ?/sec
physical_plan_clickbench_q30                          1.00     16.4±0.23ms        ? ?/sec     1.00     16.4±0.22ms        ? ?/sec
physical_plan_clickbench_q31                          1.00      2.2±0.15ms        ? ?/sec     1.02      2.2±0.15ms        ? ?/sec
physical_plan_clickbench_q32                          1.00      2.2±0.16ms        ? ?/sec     1.01      2.3±0.15ms        ? ?/sec
physical_plan_clickbench_q33                          1.01  1812.5±108.86µs        ? ?/sec    1.00  1802.5±82.56µs        ? ?/sec
physical_plan_clickbench_q34                          1.00  1550.3±82.74µs        ? ?/sec     1.00  1543.9±81.20µs        ? ?/sec
physical_plan_clickbench_q35                          1.02  1657.5±131.37µs        ? ?/sec    1.00  1625.7±93.85µs        ? ?/sec
physical_plan_clickbench_q36                          1.00  1925.7±104.86µs        ? ?/sec    1.02  1964.7±96.74µs        ? ?/sec
physical_plan_clickbench_q37                          1.00      2.4±0.18ms        ? ?/sec     1.00      2.4±0.14ms        ? ?/sec
physical_plan_clickbench_q38                          1.00      2.3±0.15ms        ? ?/sec     1.02      2.4±0.17ms        ? ?/sec
physical_plan_clickbench_q39                          1.01      2.3±0.15ms        ? ?/sec     1.00      2.3±0.16ms        ? ?/sec
physical_plan_clickbench_q4                           1.01  1223.2±60.64µs        ? ?/sec     1.00  1213.1±58.36µs        ? ?/sec
physical_plan_clickbench_q40                          1.00      2.8±0.20ms        ? ?/sec     1.01      2.8±0.19ms        ? ?/sec
physical_plan_clickbench_q41                          1.01      2.4±0.17ms        ? ?/sec     1.00      2.4±0.15ms        ? ?/sec
physical_plan_clickbench_q42                          1.00      2.3±0.14ms        ? ?/sec     1.03      2.4±0.18ms        ? ?/sec
physical_plan_clickbench_q43                          1.00      2.5±0.15ms        ? ?/sec     1.02      2.6±0.16ms        ? ?/sec
physical_plan_clickbench_q44                          1.03  1358.8±91.69µs        ? ?/sec     1.00  1317.4±49.33µs        ? ?/sec
physical_plan_clickbench_q45                          1.00  1330.5±72.65µs        ? ?/sec     1.18  1572.2±165.04µs        ? ?/sec
physical_plan_clickbench_q46                          1.00  1649.9±81.64µs        ? ?/sec     1.02  1677.4±107.69µs        ? ?/sec
physical_plan_clickbench_q47                          1.00      2.4±0.15ms        ? ?/sec     1.04      2.5±0.23ms        ? ?/sec
physical_plan_clickbench_q48                          1.00      2.5±0.15ms        ? ?/sec     1.00      2.5±0.14ms        ? ?/sec
physical_plan_clickbench_q49                          1.01      2.7±0.17ms        ? ?/sec     1.00      2.7±0.15ms        ? ?/sec
physical_plan_clickbench_q5                           1.00  1339.0±58.40µs        ? ?/sec     1.02  1367.2±82.34µs        ? ?/sec
physical_plan_clickbench_q50                          1.02      2.6±0.18ms        ? ?/sec     1.00      2.6±0.14ms        ? ?/sec
physical_plan_clickbench_q51                          1.01  1789.1±101.29µs        ? ?/sec    1.00  1766.7±101.17µs        ? ?/sec
physical_plan_clickbench_q6                           1.02  1373.4±76.07µs        ? ?/sec     1.00  1351.3±75.74µs        ? ?/sec
physical_plan_clickbench_q7                           1.00  1137.1±46.52µs        ? ?/sec     1.04  1184.8±74.97µs        ? ?/sec
physical_plan_clickbench_q8                           1.00  1663.2±118.83µs        ? ?/sec    1.00  1655.5±112.40µs        ? ?/sec
physical_plan_clickbench_q9                           1.00  1667.4±94.63µs        ? ?/sec     1.01  1684.3±98.64µs        ? ?/sec
physical_plan_struct_join_agg_sort                    1.00  1374.9±22.52µs        ? ?/sec     1.00  1372.4±22.07µs        ? ?/sec
physical_plan_tpcds_all                               1.00    770.2±6.84ms        ? ?/sec     1.02   784.4±10.12ms        ? ?/sec
physical_plan_tpch_all                                1.00     49.1±1.65ms        ? ?/sec     1.05     51.5±1.62ms        ? ?/sec
physical_plan_tpch_q1                                 1.00  1527.4±21.00µs        ? ?/sec     1.02  1551.0±21.08µs        ? ?/sec
physical_plan_tpch_q10                                1.00      2.9±0.08ms        ? ?/sec     1.03      3.0±0.08ms        ? ?/sec
physical_plan_tpch_q11                                1.00      2.6±0.06ms        ? ?/sec     1.09      2.8±0.10ms        ? ?/sec
physical_plan_tpch_q12                                1.00  1297.0±36.85µs        ? ?/sec     1.03  1340.8±38.30µs        ? ?/sec
physical_plan_tpch_q13                                1.00    996.2±9.42µs        ? ?/sec     1.02  1019.5±11.56µs        ? ?/sec
physical_plan_tpch_q14                                1.00  1316.1±29.42µs        ? ?/sec     1.04  1371.9±29.32µs        ? ?/sec
physical_plan_tpch_q16                                1.00  1713.4±66.96µs        ? ?/sec     1.00  1712.3±43.93µs        ? ?/sec
physical_plan_tpch_q17                                1.00  1835.8±62.78µs        ? ?/sec     1.02  1878.9±32.79µs        ? ?/sec
physical_plan_tpch_q18                                1.00  1987.6±64.34µs        ? ?/sec     1.03      2.0±0.07ms        ? ?/sec
physical_plan_tpch_q19                                1.00      2.5±0.07ms        ? ?/sec     1.09      2.7±0.12ms        ? ?/sec
physical_plan_tpch_q2                                 1.00      4.5±0.12ms        ? ?/sec     1.01      4.6±0.12ms        ? ?/sec
physical_plan_tpch_q20                                1.00      2.4±0.08ms        ? ?/sec     1.05      2.5±0.13ms        ? ?/sec
physical_plan_tpch_q21                                1.00      3.1±0.12ms        ? ?/sec     1.02      3.2±0.09ms        ? ?/sec
physical_plan_tpch_q22                                1.00      2.1±0.09ms        ? ?/sec     1.04      2.2±0.09ms        ? ?/sec
physical_plan_tpch_q3                                 1.00  1933.4±46.91µs        ? ?/sec     1.01  1953.7±25.47µs        ? ?/sec
physical_plan_tpch_q4                                 1.00  1032.8±12.47µs        ? ?/sec     1.02  1052.1±15.96µs        ? ?/sec
physical_plan_tpch_q5                                 1.00      2.5±0.06ms        ? ?/sec     1.04      2.6±0.07ms        ? ?/sec
physical_plan_tpch_q6                                 1.00    634.8±2.77µs        ? ?/sec     1.05    665.5±2.92µs        ? ?/sec
physical_plan_tpch_q7                                 1.00      3.1±0.09ms        ? ?/sec     1.03      3.2±0.09ms        ? ?/sec
physical_plan_tpch_q8                                 1.00      4.2±0.13ms        ? ?/sec     1.04      4.3±0.14ms        ? ?/sec
physical_plan_tpch_q9                                 1.00      3.1±0.09ms        ? ?/sec     1.02      3.1±0.09ms        ? ?/sec
physical_select_aggregates_from_200                   1.00     14.8±0.10ms        ? ?/sec     1.00     14.8±0.10ms        ? ?/sec
physical_select_all_from_1000                         1.00     17.5±0.08ms        ? ?/sec     1.00     17.4±0.06ms        ? ?/sec
physical_select_one_from_700                          1.00    781.8±3.44µs        ? ?/sec     1.00    781.6±2.85µs        ? ?/sec
physical_sorted_union_order_by_10_int64               1.00      4.9±0.10ms        ? ?/sec     1.00      4.9±0.09ms        ? ?/sec
physical_sorted_union_order_by_10_uint64              1.00     11.8±0.12ms        ? ?/sec     1.00     11.8±0.12ms        ? ?/sec
physical_sorted_union_order_by_50_int64               1.01    119.6±1.76ms        ? ?/sec     1.00    118.7±1.53ms        ? ?/sec
physical_sorted_union_order_by_50_uint64              1.00    629.8±8.45ms        ? ?/sec     1.00   628.5±10.20ms        ? ?/sec
physical_theta_join_consider_sort                     1.02   1076.4±6.97µs        ? ?/sec     1.00   1053.7±6.37µs        ? ?/sec
physical_unnest_to_join                               1.02  1242.1±14.19µs        ? ?/sec     1.00   1216.0±7.78µs        ? ?/sec
physical_window_function_partition_by_12_on_values    1.00    743.9±1.98µs        ? ?/sec     1.01    749.1±2.96µs        ? ?/sec
physical_window_function_partition_by_30_on_values    1.00  1510.5±13.31µs        ? ?/sec     1.00  1516.0±11.56µs        ? ?/sec
physical_window_function_partition_by_4_on_values     1.01    445.6±3.63µs        ? ?/sec     1.00    441.9±1.52µs        ? ?/sec
physical_window_function_partition_by_7_on_values     1.01    557.5±3.11µs        ? ?/sec     1.00    552.8±1.53µs        ? ?/sec
physical_window_function_partition_by_8_on_values     1.00    600.1±2.48µs        ? ?/sec     1.00    601.9±1.85µs        ? ?/sec
with_param_values_many_columns                        1.00    460.9±2.74µs        ? ?/sec     1.00    460.3±2.76µs        ? ?/sec

Resource Usage

base (merge-base)

Metric Value
Wall time 1256.0s
Peak memory 18.5 GiB
Avg memory 18.5 GiB
CPU user 1499.3s
CPU sys 1.8s
Disk read 0 B
Disk write 572.7 MiB

branch

Metric Value
Wall time 1251.2s
Peak memory 18.5 GiB
Avg memory 18.5 GiB
CPU user 1494.5s
CPU sys 1.3s
Disk read 0 B
Disk write 23.7 MiB

File an issue against this benchmark runner

@Dandandan Dandandan closed this Apr 4, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

physical-expr Changes to the physical-expr crates

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants