[IGNORE] run bench for sort by rluvaton · Pull Request #23493 · apache/datafusion

rluvaton · 2026-07-12T14:39:55Z

No description provided.

# Conflicts: # datafusion/physical-plan/src/sorts/merge.rs

adriangbot · 2026-07-16T13:50:26Z

🤖 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                                                     HEAD                                   add-optimization-on-yield
-----                                                     ----                                   -------------------------
sort_axis/sort i64 1M High cardinality Sorted +100cols    1.03  1397.5±24.73ms        ? ?/sec    1.00  1357.5±19.27ms        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric	Value
Wall time	1025.2s
Peak memory	2.1 GiB
Avg memory	666.2 MiB
CPU user	565.5s
CPU sys	42.2s
Peak spill	0 B

sort — branch

Metric	Value
Wall time	1025.2s
Peak memory	2.1 GiB
Avg memory	655.2 MiB
CPU user	561.9s
CPU sys	43.2s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-16T14:04:32Z

run benchmark sort

env:
  BENCH_FILTER: 'i64 1M High cardinality Sorted \+100cols$'

rluvaton · 2026-07-16T14:04:54Z

run benchmark sort

env:
  BENCH_FILTER: 'mixed tuple 1M High cardinality Sorted \+100cols$'

adriangbot · 2026-07-16T14:06:36Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4992819664-1101-q8ssr 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (bcdbfe4) to a19a178 (merge-base) diff using: sort
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T14:07:00Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4992823341-1102-94m44 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (bcdbfe4) to a19a178 (merge-base) diff using: sort
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T14:42:48Z

🤖 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                                                     HEAD                                   add-optimization-on-yield
-----                                                     ----                                   -------------------------
sort_axis/sort i64 1M High cardinality Sorted +100cols    18.57 1379.5±20.83ms        ? ?/sec    1.00     74.3±1.54ms        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric	Value
Wall time	1025.2s
Peak memory	2.0 GiB
Avg memory	671.3 MiB
CPU user	564.1s
CPU sys	41.9s
Peak spill	0 B

sort — branch

Metric	Value
Wall time	1135.3s
Peak memory	2.2 GiB
Avg memory	359.2 MiB
CPU user	459.1s
CPU sys	46.8s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-16T14:46:47Z

🤖 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                                                             HEAD                                   add-optimization-on-yield
-----                                                             ----                                   -------------------------
sort_axis/sort mixed tuple 1M High cardinality Sorted +100cols    16.65 1449.7±33.92ms        ? ?/sec    1.00     87.1±1.06ms        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric	Value
Wall time	1195.3s
Peak memory	2.3 GiB
Avg memory	675.0 MiB
CPU user	709.5s
CPU sys	64.1s
Peak spill	0 B

sort — branch

Metric	Value
Wall time	1180.3s
Peak memory	2.4 GiB
Avg memory	461.2 MiB
CPU user	603.6s
CPU sys	66.7s
Peak spill	0 B

File an issue against this benchmark runner

This reverts commit bcdbfe4.

This reverts commit fa456a5.

rluvaton · 2026-07-16T15:03:13Z

run benchmark sort

env:
  BENCH_FILTER: 'mixed tuple 1M High cardinality Sorted \+100cols$'

rluvaton · 2026-07-16T15:03:18Z

run benchmark sort

env:
  BENCH_FILTER: 'i64 1M High cardinality Sorted \+100cols$'

adriangbot · 2026-07-16T15:05:26Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4993421198-1108-mwjmn 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: sort
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T15:06:12Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4993422026-1109-q9bxt 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: sort
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T15:40:39Z

🤖 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                                                     HEAD                                   add-optimization-on-yield
-----                                                     ----                                   -------------------------
sort_axis/sort i64 1M High cardinality Sorted +100cols    1.00  1399.5±30.24ms        ? ?/sec    1.00  1403.8±25.20ms        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric	Value
Wall time	1025.2s
Peak memory	2.0 GiB
Avg memory	661.3 MiB
CPU user	565.4s
CPU sys	41.7s
Peak spill	0 B

sort — branch

Metric	Value
Wall time	1030.2s
Peak memory	2.0 GiB
Avg memory	659.8 MiB
CPU user	565.7s
CPU sys	43.8s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-16T15:45:28Z

🤖 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                                                             HEAD                                   add-optimization-on-yield
-----                                                             ----                                   -------------------------
sort_axis/sort mixed tuple 1M High cardinality Sorted +100cols    1.03  1452.3±23.73ms        ? ?/sec    1.00  1404.2±25.52ms        ? ?/sec

Resource Usage

sort — base (merge-base)

Metric	Value
Wall time	1195.3s
Peak memory	2.3 GiB
Avg memory	669.2 MiB
CPU user	710.1s
CPU sys	64.6s
Peak spill	0 B

sort — branch

Metric	Value
Wall time	1195.3s
Peak memory	2.3 GiB
Avg memory	660.9 MiB
CPU user	702.0s
CPU sys	64.7s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-16T17:05:38Z

run benchmark tpch10_sort tpch10

adriangbot · 2026-07-16T17:08:23Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4994572538-1113-dnqn7 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: tpch10_sort
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T17:08:36Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4994572538-1114-2md54 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: tpch10
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T17:25:31Z

Benchmark for this request failed.

Last 20 lines of output:

Click to expand

# Micro-Benchmarks (specific operators and features)
cancellation:           How long cancelling a query takes
nlj:                    Benchmark for simple nested loop joins, testing various join scenarios
hj:                     Benchmark for simple hash joins, testing various join scenarios
smj:                    Benchmark for simple sort merge joins, testing various join scenarios
dict:                   Benchmark for dictionary-encoded group-by scenarios
compile_profile:        Compile and execute TPC-H across selected Cargo profiles, reporting timing and binary size


━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Supported Configuration (Environment Variables)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
DATA_DIR            directory to store datasets
CARGO_COMMAND       command that runs the benchmark binary
DATAFUSION_DIR      directory to use (default /workspace/datafusion-base)
RESULTS_NAME        folder where the benchmark files are stored
PREFER_HASH_JOIN    Prefer hash join algorithm (default true)
SIMULATE_LATENCY    Simulate object store latency to mimic S3 (default false)
DATAFUSION_*        Set the given datafusion configuration

File an issue against this benchmark runner

adriangbot · 2026-07-16T17:28:13Z

🤖 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 add-optimization-on-yield
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃          add-optimization-on-yield ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │  314.07 / 316.88 ±1.85 / 319.15 ms │  311.14 / 313.14 ±1.43 / 315.57 ms │    no change │
│ QQuery 2  │  103.64 / 105.87 ±2.41 / 110.46 ms │  100.16 / 107.74 ±4.60 / 113.96 ms │    no change │
│ QQuery 3  │  233.33 / 239.88 ±6.00 / 249.90 ms │  229.52 / 235.90 ±3.65 / 240.25 ms │    no change │
│ QQuery 4  │  112.40 / 112.97 ±0.39 / 113.58 ms │  112.00 / 115.29 ±2.82 / 120.28 ms │    no change │
│ QQuery 5  │  351.71 / 357.64 ±7.14 / 371.67 ms │  353.25 / 362.33 ±5.51 / 369.82 ms │    no change │
│ QQuery 6  │  126.88 / 128.75 ±2.86 / 134.35 ms │  127.00 / 129.44 ±3.51 / 136.08 ms │    no change │
│ QQuery 7  │  458.73 / 466.85 ±5.84 / 473.23 ms │  462.09 / 470.60 ±7.60 / 482.98 ms │    no change │
│ QQuery 8  │  379.35 / 386.01 ±4.21 / 392.16 ms │  386.50 / 391.59 ±3.41 / 396.81 ms │    no change │
│ QQuery 9  │  543.57 / 552.99 ±5.95 / 560.97 ms │  554.48 / 559.50 ±2.54 / 561.13 ms │    no change │
│ QQuery 10 │  299.89 / 315.52 ±9.17 / 325.61 ms │  301.58 / 308.77 ±7.05 / 321.12 ms │    no change │
│ QQuery 11 │     84.50 / 87.48 ±3.81 / 94.74 ms │    85.20 / 93.61 ±8.63 / 109.26 ms │ 1.07x slower │
│ QQuery 12 │  179.32 / 185.40 ±5.51 / 194.80 ms │  178.31 / 181.53 ±2.29 / 185.47 ms │    no change │
│ QQuery 13 │  285.11 / 292.14 ±6.00 / 300.17 ms │  291.95 / 296.35 ±4.57 / 305.01 ms │    no change │
│ QQuery 14 │  173.58 / 179.07 ±5.45 / 187.80 ms │  173.83 / 178.32 ±4.08 / 184.78 ms │    no change │
│ QQuery 15 │  308.91 / 311.43 ±2.50 / 316.22 ms │  307.99 / 311.37 ±3.40 / 317.46 ms │    no change │
│ QQuery 16 │     63.87 / 66.19 ±2.23 / 70.04 ms │     63.05 / 66.87 ±3.40 / 72.12 ms │    no change │
│ QQuery 17 │  628.51 / 631.30 ±1.82 / 633.96 ms │ 617.04 / 628.44 ±12.02 / 651.44 ms │    no change │
│ QQuery 18 │ 656.85 / 688.74 ±19.61 / 718.53 ms │ 677.41 / 697.48 ±18.85 / 732.90 ms │    no change │
│ QQuery 19 │ 251.92 / 266.49 ±11.79 / 281.75 ms │ 247.35 / 258.47 ±13.44 / 279.04 ms │    no change │
│ QQuery 20 │ 279.60 / 293.37 ±12.52 / 312.10 ms │ 276.44 / 294.84 ±10.71 / 305.57 ms │    no change │
│ QQuery 21 │  658.22 / 667.43 ±6.43 / 676.85 ms │  654.56 / 661.91 ±6.82 / 673.11 ms │    no change │
│ QQuery 22 │     61.68 / 64.79 ±4.50 / 73.71 ms │     61.29 / 63.38 ±1.65 / 66.00 ms │    no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                        ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                        │ 6717.19ms │
│ Total Time (add-optimization-on-yield)   │ 6726.87ms │
│ Average Time (HEAD)                      │  305.33ms │
│ Average Time (add-optimization-on-yield) │  305.77ms │
│ Queries Faster                           │         0 │
│ Queries Slower                           │         1 │
│ Queries with No Change                   │        21 │
│ Queries with Failure                     │         0 │
└──────────────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric	Value
Wall time	35.0s
Peak memory	4.5 GiB
Avg memory	1.4 GiB
CPU user	340.5s
CPU sys	20.6s
Peak spill	0 B

tpch10 — branch

Metric	Value
Wall time	35.0s
Peak memory	4.2 GiB
Avg memory	1.4 GiB
CPU user	339.7s
CPU sys	20.8s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-16T17:31:04Z

run benchmark sort_tpch10

adriangbot · 2026-07-16T17:33:51Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4994790843-1117-wlpdz 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: sort_tpch10
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T17:52:17Z

🤖 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 add-optimization-on-yield
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃                                   HEAD ┃                add-optimization-on-yield ┃        Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ Q1    │   1657.19 / 1659.80 ±2.48 / 1663.80 ms │     1728.73 / 1731.84 ±4.79 / 1741.27 ms │     no change │
│ Q2    │  1515.32 / 1569.98 ±31.83 / 1600.00 ms │       682.52 / 699.74 ±11.77 / 716.11 ms │ +2.24x faster │
│ Q3    │  8948.45 / 9017.30 ±54.12 / 9093.90 ms │    9298.99 / 9353.80 ±40.49 / 9401.81 ms │     no change │
│ Q4    │  2874.03 / 2901.55 ±18.77 / 2929.64 ms │    1252.58 / 1300.95 ±33.61 / 1351.22 ms │ +2.23x faster │
│ Q5    │   2669.74 / 2677.55 ±7.71 / 2688.29 ms │    2848.03 / 2864.25 ±16.17 / 2894.09 ms │  1.07x slower │
│ Q6    │  2799.62 / 2822.54 ±15.08 / 2838.74 ms │    2964.17 / 3004.10 ±29.08 / 3054.33 ms │  1.06x slower │
│ Q7    │  5642.32 / 5682.36 ±32.25 / 5715.66 ms │    5820.74 / 5876.66 ±41.57 / 5949.40 ms │     no change │
│ Q8    │  4062.45 / 4218.75 ±99.54 / 4336.90 ms │    3937.96 / 4009.53 ±52.18 / 4093.87 ms │     no change │
│ Q9    │  4570.33 / 4649.84 ±59.25 / 4726.17 ms │   4524.28 / 4647.23 ±120.78 / 4831.70 ms │     no change │
│ Q10   │ 7140.97 / 7420.77 ±169.36 / 7631.14 ms │ 10066.89 / 10207.35 ±86.33 / 10301.24 ms │  1.38x slower │
│ Q11   │ 3505.15 / 3589.59 ±113.99 / 3809.32 ms │   3758.14 / 3884.97 ±229.86 / 4344.16 ms │  1.08x slower │
└───────┴────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                        ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                        │ 46210.03ms │
│ Total Time (add-optimization-on-yield)   │ 47580.41ms │
│ Average Time (HEAD)                      │  4200.91ms │
│ Average Time (add-optimization-on-yield) │  4325.49ms │
│ Queries Faster                           │          2 │
│ Queries Slower                           │          4 │
│ Queries with No Change                   │          5 │
│ Queries with Failure                     │          0 │
└──────────────────────────────────────────┴────────────┘

Resource Usage

sort_tpch10 — base (merge-base)

Metric	Value
Wall time	235.1s
Peak memory	19.4 GiB
Avg memory	7.4 GiB
CPU user	913.7s
CPU sys	73.4s
Peak spill	0 B

sort_tpch10 — branch

Metric	Value
Wall time	240.1s
Peak memory	19.1 GiB
Avg memory	9.0 GiB
CPU user	911.1s
CPU sys	61.0s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-16T18:04:41Z

run benchmark sort_tpch10

baseline:
  env:
    DATAFUSION_EXECUTION_TARGET_PARTITIONS: 1

changed:
  env:
    DATAFUSION_EXECUTION_TARGET_PARTITIONS: 1

adriangbot · 2026-07-16T18:07:30Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4995081240-1118-vl9zm 6.12.85+ #1 SMP Mon May 11 08:17:35 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 add-optimization-on-yield (1277981) to a19a178 (merge-base) diff using: sort_tpch10
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-16T19:03:04Z

🤖 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 add-optimization-on-yield
--------------------
Benchmark sort_tpch10.json
--------------------
┏━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query ┃                                       HEAD ┃                  add-optimization-on-yield ┃        Change ┃
┡━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ Q1    │       3156.85 / 3163.93 ±7.59 / 3175.18 ms │      3220.78 / 3277.70 ±34.68 / 3319.28 ms │     no change │
│ Q2    │      3232.41 / 3416.51 ±96.54 / 3514.50 ms │      2224.07 / 2278.70 ±60.90 / 2391.16 ms │ +1.50x faster │
│ Q3    │ 38500.99 / 41840.91 ±2229.59 / 44043.18 ms │ 40526.01 / 42044.06 ±1198.16 / 43948.96 ms │     no change │
│ Q4    │     7009.79 / 7227.03 ±161.52 / 7417.00 ms │      4085.75 / 4166.71 ±64.28 / 4265.24 ms │ +1.73x faster │
│ Q5    │ 31318.71 / 33094.37 ±1637.23 / 35360.82 ms │ 31068.40 / 33387.42 ±1631.37 / 35992.15 ms │     no change │
│ Q6    │ 35463.22 / 37940.44 ±1745.29 / 40128.83 ms │ 35986.33 / 37169.46 ±1153.39 / 38580.53 ms │     no change │
│ Q7    │ 70422.79 / 73048.13 ±2112.52 / 76390.30 ms │ 69563.87 / 84447.76 ±7980.13 / 90908.05 ms │  1.16x slower │
│ Q8    │  12164.60 / 12514.63 ±341.45 / 13132.54 ms │  10726.04 / 11360.70 ±477.42 / 11958.81 ms │ +1.10x faster │
│ Q9    │  14194.23 / 14492.54 ±355.61 / 15153.63 ms │  13022.43 / 13439.36 ±356.62 / 14103.86 ms │ +1.08x faster │
│ Q10   │ 28761.77 / 31280.90 ±1569.86 / 33688.11 ms │ 29551.84 / 31661.72 ±1245.50 / 33072.28 ms │     no change │
│ Q11   │  10446.16 / 10717.37 ±186.30 / 10925.44 ms │  10459.06 / 10812.88 ±405.94 / 11584.26 ms │     no change │
└───────┴────────────────────────────────────────────┴────────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┓
┃ Benchmark Summary                        ┃             ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━┩
│ Total Time (HEAD)                        │ 268736.76ms │
│ Total Time (add-optimization-on-yield)   │ 274046.46ms │
│ Average Time (HEAD)                      │  24430.61ms │
│ Average Time (add-optimization-on-yield) │  24913.31ms │
│ Queries Faster                           │           4 │
│ Queries Slower                           │           1 │
│ Queries with No Change                   │           6 │
│ Queries with Failure                     │           0 │
└──────────────────────────────────────────┴─────────────┘

Resource Usage

sort_tpch10 — base (merge-base)

Metric	Value
Wall time	1345.3s
Peak memory	19.2 GiB
Avg memory	8.4 GiB
CPU user	1473.4s
CPU sys	70.5s
Peak spill	0 B

sort_tpch10 — branch

Metric	Value
Wall time	1375.3s
Peak memory	19.1 GiB
Avg memory	8.7 GiB
CPU user	1498.5s
CPU sys	72.3s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton added 23 commits July 9, 2026 09:44

use yield

4e64c65

prototype

4432b46

Merge branch 'main' into use-yield

81f6c59

fix compilation

079fc8a

try another way with different crate

18262df

fix - emit all and not wait before emitting next

d01f841

c

ffd603e

add comment

e6df5f5

add metrics support for adapter

847172e

align with main

0adc2e5

use observed stream

f438d7b

Merge branch 'main' into use-yield

fa51b4f

uncomment pr

6171c12

add into stream

e56c139

format

72f3513

revert test to main

03441d0

fix test

3e5ab00

Merge branch 'main' into use-yield

9251ed5

# Conflicts: # datafusion/physical-plan/src/sorts/merge.rs

update with main

3182a97

move comment

48aab96

format Cargo.toml

039d073

add passthrough last stream optimization

dfd1226

add batch wise skip

b9a5d05

github-actions Bot added the physical-plan Changes to the physical-plan crate label Jul 12, 2026

rluvaton added 3 commits July 12, 2026 17:46

fix condition

b744378

fix test

10c1133

Merge branch 'main' into add-optimization-on-yield

2e064a1

apache deleted a comment from adriangbot Jul 12, 2026

fix infinite loop

6063261

remove coalesce and also add batch emit right away

fa456a5

github-actions Bot added the auto detected api change Auto detected API change label Jul 16, 2026

fix build

bcdbfe4

github-actions Bot removed the auto detected api change Auto detected API change label Jul 16, 2026

rluvaton added 2 commits July 16, 2026 17:53

Revert "fix build"

c33dfca

This reverts commit bcdbfe4.

Revert "remove coalesce and also add batch emit right away"

1277981

This reverts commit fa456a5.

Uh oh!

Conversation

rluvaton commented Jul 12, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

rluvaton commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

adriangbot commented Jul 16, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants