refactor: Split Parquet BloomFilter CPU and IO into separate states by alamb · Pull Request #21285 · apache/datafusion

alamb · 2026-03-31T11:46:29Z

Which issue does this PR close?

part of Introduce morsel-driven Parquet scan #20529
Broken out of Sketch out a Morselize API #20820

Rationale for this change

We are try to be explicit about CPU and IO in the Parquet reader, however, the code for applying bloom filters still did both CPU and IO in the same future

What changes are included in this PR?

Split the states for IO and CPU work when applying bloom filters

Are these changes tested?

Functionally by existing tests

I also ran performance tests and they didn't show any substantial performance change

Are there any user-facing changes?

No, this is internal code reorganization

alamb · 2026-03-31T13:00:52Z

run benchmarks

adriangbot · 2026-03-31T13:03:29Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4162481196-627-f46qb 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 alamb/split_bloom (ac46396) to 5ff80e4 (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T13:03:29Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4162481196-625-kbvsm 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 alamb/split_bloom (ac46396) to 5ff80e4 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T13:03:29Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4162481196-626-vcq8m 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 alamb/split_bloom (ac46396) to 5ff80e4 (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T13:16:23Z

🤖 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 alamb_split_bloom
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃              alamb_split_bloom ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 46.16 / 47.17 ±0.97 / 48.62 ms │ 45.62 / 46.28 ±0.72 / 47.65 ms │ no change │
│ QQuery 2  │ 21.71 / 22.11 ±0.26 / 22.53 ms │ 21.77 / 22.92 ±1.37 / 25.49 ms │ no change │
│ QQuery 3  │ 32.44 / 33.38 ±1.25 / 35.85 ms │ 32.46 / 32.63 ±0.12 / 32.77 ms │ no change │
│ QQuery 4  │ 20.92 / 21.98 ±0.66 / 22.72 ms │ 22.03 / 22.58 ±0.35 / 22.97 ms │ no change │
│ QQuery 5  │ 49.34 / 51.24 ±1.79 / 53.66 ms │ 49.10 / 49.84 ±0.90 / 51.58 ms │ no change │
│ QQuery 6  │ 17.30 / 17.51 ±0.18 / 17.76 ms │ 17.40 / 17.55 ±0.16 / 17.80 ms │ no change │
│ QQuery 7  │ 56.11 / 57.49 ±0.97 / 58.92 ms │ 55.35 / 57.25 ±1.74 / 59.70 ms │ no change │
│ QQuery 8  │ 49.60 / 49.83 ±0.32 / 50.47 ms │ 49.19 / 49.46 ±0.19 / 49.64 ms │ no change │
│ QQuery 9  │ 55.93 / 56.76 ±0.97 / 58.52 ms │ 55.17 / 55.78 ±0.54 / 56.51 ms │ no change │
│ QQuery 10 │ 72.45 / 73.48 ±0.71 / 74.38 ms │ 71.96 / 72.67 ±0.47 / 73.21 ms │ no change │
│ QQuery 11 │ 14.54 / 14.86 ±0.36 / 15.55 ms │ 14.52 / 14.73 ±0.29 / 15.29 ms │ no change │
│ QQuery 12 │ 28.52 / 28.80 ±0.29 / 29.34 ms │ 28.30 / 28.51 ±0.25 / 28.93 ms │ no change │
│ QQuery 13 │ 38.87 / 39.53 ±0.46 / 40.02 ms │ 39.55 / 40.12 ±0.40 / 40.64 ms │ no change │
│ QQuery 14 │ 28.77 / 29.05 ±0.26 / 29.47 ms │ 29.10 / 29.41 ±0.25 / 29.71 ms │ no change │
│ QQuery 15 │ 33.84 / 34.39 ±0.48 / 35.27 ms │ 34.24 / 34.43 ±0.15 / 34.65 ms │ no change │
│ QQuery 16 │ 16.24 / 16.56 ±0.21 / 16.80 ms │ 16.06 / 16.62 ±0.39 / 17.16 ms │ no change │
│ QQuery 17 │ 74.52 / 75.23 ±0.86 / 76.85 ms │ 74.16 / 76.14 ±1.86 / 79.21 ms │ no change │
│ QQuery 18 │ 78.68 / 80.06 ±1.17 / 82.11 ms │ 78.47 / 79.68 ±1.17 / 81.49 ms │ no change │
│ QQuery 19 │ 38.07 / 38.35 ±0.25 / 38.77 ms │ 37.69 / 38.24 ±0.40 / 38.92 ms │ no change │
│ QQuery 20 │ 40.50 / 41.40 ±1.02 / 43.34 ms │ 40.51 / 41.48 ±0.77 / 42.37 ms │ no change │
│ QQuery 21 │ 66.42 / 67.12 ±0.73 / 68.51 ms │ 65.82 / 68.09 ±1.26 / 69.56 ms │ no change │
│ QQuery 22 │ 18.40 / 18.66 ±0.42 / 19.48 ms │ 18.51 / 19.12 ±0.55 / 19.81 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                │ 914.95ms │
│ Total Time (alamb_split_bloom)   │ 913.54ms │
│ Average Time (HEAD)              │  41.59ms │
│ Average Time (alamb_split_bloom) │  41.52ms │
│ 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	33.9s
CPU sys	3.1s
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	33.9s
CPU sys	2.9s
Disk read	0 B
Disk write	60.0 KiB

File an issue against this benchmark runner

adriangbot · 2026-03-31T13:21:48Z

🤖 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

Comparing HEAD and alamb_split_bloom
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃                        alamb_split_bloom ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           45.35 / 45.90 ±0.49 / 46.69 ms │           43.13 / 44.23 ±0.90 / 45.43 ms │     no change │
│ QQuery 2  │        150.98 / 153.00 ±2.24 / 157.35 ms │        145.61 / 146.59 ±0.58 / 147.19 ms │     no change │
│ QQuery 3  │        118.16 / 118.74 ±0.45 / 119.15 ms │        114.18 / 115.54 ±1.07 / 116.80 ms │     no change │
│ QQuery 4  │    1472.83 / 1489.94 ±12.71 / 1508.55 ms │    1308.67 / 1351.92 ±24.93 / 1382.03 ms │ +1.10x faster │
│ QQuery 5  │        178.70 / 180.70 ±1.89 / 183.92 ms │        176.28 / 177.47 ±0.89 / 178.55 ms │     no change │
│ QQuery 6  │    1050.56 / 1097.85 ±26.79 / 1120.79 ms │     998.85 / 1032.96 ±29.16 / 1073.04 ms │ +1.06x faster │
│ QQuery 7  │        364.98 / 371.52 ±8.02 / 387.18 ms │        356.98 / 360.41 ±2.64 / 364.09 ms │     no change │
│ QQuery 8  │        120.21 / 121.03 ±0.73 / 121.92 ms │        115.27 / 116.92 ±1.62 / 119.84 ms │     no change │
│ QQuery 9  │        103.04 / 106.94 ±2.68 / 109.56 ms │        101.53 / 104.58 ±2.53 / 107.74 ms │     no change │
│ QQuery 10 │        111.20 / 112.78 ±1.40 / 115.40 ms │        108.68 / 109.87 ±1.36 / 112.39 ms │     no change │
│ QQuery 11 │    1029.51 / 1046.01 ±10.17 / 1058.70 ms │        964.22 / 980.29 ±8.94 / 991.10 ms │ +1.07x faster │
│ QQuery 12 │           47.97 / 48.82 ±0.75 / 49.88 ms │           45.86 / 47.48 ±1.08 / 48.78 ms │     no change │
│ QQuery 13 │        410.71 / 416.30 ±3.34 / 420.26 ms │        406.42 / 411.09 ±2.81 / 413.55 ms │     no change │
│ QQuery 14 │     1050.97 / 1060.95 ±7.67 / 1071.99 ms │     1036.94 / 1045.89 ±6.58 / 1056.12 ms │     no change │
│ QQuery 15 │           17.54 / 18.78 ±0.84 / 19.92 ms │           16.55 / 17.36 ±0.90 / 19.02 ms │ +1.08x faster │
│ QQuery 16 │           42.33 / 43.08 ±0.63 / 44.23 ms │           40.45 / 43.17 ±2.03 / 46.02 ms │     no change │
│ QQuery 17 │        242.67 / 247.50 ±3.53 / 253.69 ms │        243.94 / 244.52 ±0.81 / 246.11 ms │     no change │
│ QQuery 18 │        131.11 / 132.23 ±0.88 / 133.67 ms │        130.74 / 131.62 ±0.85 / 133.17 ms │     no change │
│ QQuery 19 │        159.59 / 162.23 ±1.59 / 164.54 ms │        157.67 / 160.39 ±1.52 / 162.12 ms │     no change │
│ QQuery 20 │           14.15 / 14.83 ±0.56 / 15.76 ms │           14.45 / 15.12 ±0.36 / 15.50 ms │     no change │
│ QQuery 21 │           20.71 / 21.09 ±0.37 / 21.76 ms │           19.95 / 20.28 ±0.28 / 20.79 ms │     no change │
│ QQuery 22 │        499.60 / 501.91 ±1.86 / 504.81 ms │        500.78 / 505.88 ±5.81 / 515.23 ms │     no change │
│ QQuery 23 │        939.88 / 945.10 ±3.15 / 948.92 ms │        894.21 / 906.75 ±8.97 / 919.53 ms │     no change │
│ QQuery 24 │        426.24 / 428.12 ±2.00 / 430.63 ms │        421.57 / 423.55 ±1.49 / 425.79 ms │     no change │
│ QQuery 25 │        360.84 / 363.35 ±2.36 / 367.14 ms │        360.14 / 362.25 ±1.39 / 364.47 ms │     no change │
│ QQuery 26 │           83.66 / 84.86 ±0.91 / 86.12 ms │           83.20 / 85.11 ±1.53 / 87.16 ms │     no change │
│ QQuery 27 │        355.05 / 357.60 ±2.29 / 361.43 ms │        353.24 / 355.46 ±2.50 / 358.84 ms │     no change │
│ QQuery 28 │        151.16 / 152.22 ±0.74 / 153.19 ms │        150.46 / 152.09 ±1.37 / 154.49 ms │     no change │
│ QQuery 29 │        304.59 / 305.72 ±0.81 / 306.77 ms │        300.74 / 302.75 ±1.01 / 303.39 ms │     no change │
│ QQuery 30 │           44.54 / 46.51 ±2.62 / 51.62 ms │           44.37 / 45.47 ±1.13 / 47.46 ms │     no change │
│ QQuery 31 │        170.02 / 175.65 ±4.05 / 182.02 ms │        172.29 / 173.63 ±1.17 / 175.56 ms │     no change │
│ QQuery 32 │           58.93 / 59.71 ±0.96 / 61.44 ms │           56.90 / 59.05 ±1.52 / 61.65 ms │     no change │
│ QQuery 33 │        145.29 / 145.48 ±0.21 / 145.83 ms │        142.87 / 144.53 ±1.28 / 146.82 ms │     no change │
│ QQuery 34 │        109.06 / 109.83 ±0.54 / 110.51 ms │        107.24 / 108.53 ±0.80 / 109.57 ms │     no change │
│ QQuery 35 │        109.13 / 112.51 ±1.92 / 114.90 ms │        108.98 / 111.33 ±1.95 / 114.58 ms │     no change │
│ QQuery 36 │        220.01 / 227.74 ±4.34 / 232.48 ms │        219.90 / 223.87 ±3.64 / 228.49 ms │     no change │
│ QQuery 37 │        180.94 / 183.65 ±2.07 / 186.92 ms │        177.16 / 179.89 ±2.07 / 183.25 ms │     no change │
│ QQuery 38 │          90.54 / 93.78 ±3.53 / 100.29 ms │           84.58 / 89.63 ±3.63 / 95.57 ms │     no change │
│ QQuery 39 │        132.15 / 134.41 ±1.24 / 135.71 ms │        127.24 / 130.30 ±2.85 / 135.51 ms │     no change │
│ QQuery 40 │        115.43 / 120.54 ±4.30 / 125.82 ms │        111.41 / 118.87 ±5.11 / 127.41 ms │     no change │
│ QQuery 41 │           15.08 / 16.01 ±0.59 / 16.79 ms │           14.51 / 15.49 ±1.10 / 17.63 ms │     no change │
│ QQuery 42 │        111.19 / 112.49 ±1.27 / 114.92 ms │        107.69 / 109.06 ±1.22 / 110.86 ms │     no change │
│ QQuery 43 │           85.72 / 87.79 ±1.59 / 89.75 ms │           84.49 / 85.27 ±0.68 / 86.08 ms │     no change │
│ QQuery 44 │           12.14 / 12.68 ±0.58 / 13.48 ms │           11.56 / 11.98 ±0.33 / 12.41 ms │ +1.06x faster │
│ QQuery 45 │           55.79 / 57.04 ±0.89 / 57.99 ms │           53.21 / 55.24 ±2.07 / 57.81 ms │     no change │
│ QQuery 46 │        242.89 / 244.56 ±1.28 / 246.66 ms │        233.10 / 236.16 ±1.71 / 238.15 ms │     no change │
│ QQuery 47 │        771.64 / 781.84 ±7.77 / 790.66 ms │       694.44 / 710.79 ±10.88 / 726.06 ms │ +1.10x faster │
│ QQuery 48 │        287.35 / 296.53 ±5.09 / 302.56 ms │        290.13 / 293.01 ±2.71 / 297.34 ms │     no change │
│ QQuery 49 │        260.98 / 264.17 ±1.87 / 266.86 ms │        255.51 / 257.71 ±1.75 / 260.18 ms │     no change │
│ QQuery 50 │        243.93 / 249.24 ±3.11 / 253.59 ms │        226.64 / 235.17 ±5.51 / 243.94 ms │ +1.06x faster │
│ QQuery 51 │        188.09 / 191.18 ±2.73 / 196.08 ms │        182.69 / 186.13 ±1.99 / 188.59 ms │     no change │
│ QQuery 52 │        110.73 / 112.31 ±1.26 / 114.15 ms │        108.13 / 109.38 ±0.94 / 110.16 ms │     no change │
│ QQuery 53 │        105.52 / 106.95 ±1.15 / 108.61 ms │        104.65 / 106.09 ±1.12 / 107.60 ms │     no change │
│ QQuery 54 │        151.90 / 153.28 ±1.14 / 154.99 ms │        148.54 / 151.02 ±1.64 / 153.03 ms │     no change │
│ QQuery 55 │        110.51 / 112.35 ±2.33 / 116.82 ms │        108.61 / 110.39 ±1.39 / 112.49 ms │     no change │
│ QQuery 56 │        145.11 / 145.58 ±0.36 / 145.98 ms │        142.45 / 144.37 ±1.53 / 146.48 ms │     no change │
│ QQuery 57 │        179.19 / 180.25 ±0.66 / 180.96 ms │        176.58 / 178.10 ±1.60 / 180.08 ms │     no change │
│ QQuery 58 │        299.01 / 304.73 ±3.61 / 309.03 ms │       299.57 / 318.05 ±13.26 / 338.80 ms │     no change │
│ QQuery 59 │        206.86 / 209.34 ±1.93 / 211.81 ms │        200.91 / 202.68 ±1.27 / 204.40 ms │     no change │
│ QQuery 60 │        150.16 / 152.09 ±1.34 / 153.66 ms │        146.06 / 147.31 ±0.95 / 148.68 ms │     no change │
│ QQuery 61 │        177.72 / 179.10 ±0.97 / 180.64 ms │        172.49 / 174.23 ±1.99 / 177.69 ms │     no change │
│ QQuery 62 │      941.56 / 979.95 ±26.48 / 1010.49 ms │      934.97 / 969.21 ±29.62 / 1024.11 ms │     no change │
│ QQuery 63 │        108.10 / 110.71 ±1.52 / 112.20 ms │        104.72 / 107.98 ±3.08 / 113.79 ms │     no change │
│ QQuery 64 │        718.51 / 730.24 ±8.90 / 744.75 ms │        706.73 / 715.96 ±9.93 / 734.56 ms │     no change │
│ QQuery 65 │        255.19 / 262.18 ±4.46 / 268.68 ms │        250.01 / 253.15 ±3.24 / 257.18 ms │     no change │
│ QQuery 66 │       243.71 / 259.15 ±11.11 / 273.09 ms │       251.36 / 263.92 ±12.70 / 282.86 ms │     no change │
│ QQuery 67 │        323.78 / 336.27 ±8.80 / 347.48 ms │        319.64 / 331.57 ±6.97 / 338.81 ms │     no change │
│ QQuery 68 │        294.31 / 300.01 ±4.60 / 308.30 ms │        278.21 / 284.03 ±3.41 / 287.68 ms │ +1.06x faster │
│ QQuery 69 │        106.97 / 107.83 ±1.11 / 110.02 ms │        103.40 / 105.20 ±1.74 / 107.63 ms │     no change │
│ QQuery 70 │       334.30 / 347.65 ±10.29 / 364.92 ms │        345.61 / 358.69 ±8.19 / 369.98 ms │     no change │
│ QQuery 71 │        136.86 / 141.52 ±4.03 / 148.17 ms │        134.80 / 138.92 ±2.62 / 142.01 ms │     no change │
│ QQuery 72 │        736.00 / 751.53 ±8.63 / 760.01 ms │        727.02 / 735.54 ±6.40 / 744.32 ms │     no change │
│ QQuery 73 │        105.27 / 106.57 ±1.30 / 108.84 ms │        101.40 / 105.64 ±3.70 / 110.50 ms │     no change │
│ QQuery 74 │        571.58 / 580.13 ±6.46 / 590.93 ms │       556.75 / 589.26 ±32.57 / 644.41 ms │     no change │
│ QQuery 75 │        278.66 / 282.41 ±2.29 / 284.92 ms │        279.94 / 282.86 ±1.52 / 284.35 ms │     no change │
│ QQuery 76 │        137.72 / 139.26 ±1.25 / 140.82 ms │        133.23 / 135.50 ±1.85 / 138.52 ms │     no change │
│ QQuery 77 │        191.59 / 193.50 ±1.53 / 195.91 ms │        192.20 / 193.66 ±1.25 / 195.88 ms │     no change │
│ QQuery 78 │        355.40 / 358.65 ±3.44 / 364.59 ms │        354.55 / 358.81 ±3.87 / 363.75 ms │     no change │
│ QQuery 79 │        234.41 / 239.60 ±4.14 / 245.42 ms │        236.72 / 239.55 ±2.89 / 244.52 ms │     no change │
│ QQuery 80 │        335.04 / 338.50 ±2.11 / 341.33 ms │        331.49 / 335.18 ±3.77 / 342.41 ms │     no change │
│ QQuery 81 │           27.38 / 28.30 ±0.56 / 28.83 ms │           26.55 / 28.05 ±1.43 / 30.34 ms │     no change │
│ QQuery 82 │        201.65 / 204.58 ±1.85 / 207.21 ms │        203.90 / 205.66 ±1.29 / 207.40 ms │     no change │
│ QQuery 83 │           40.38 / 41.32 ±0.93 / 42.59 ms │           40.56 / 42.59 ±1.53 / 44.86 ms │     no change │
│ QQuery 84 │           49.34 / 50.39 ±1.15 / 52.44 ms │           48.36 / 49.76 ±1.15 / 51.87 ms │     no change │
│ QQuery 85 │        149.73 / 154.22 ±3.14 / 157.80 ms │        148.86 / 151.39 ±1.68 / 153.48 ms │     no change │
│ QQuery 86 │           39.54 / 40.35 ±0.77 / 41.52 ms │           39.45 / 40.48 ±0.70 / 41.47 ms │     no change │
│ QQuery 87 │           83.60 / 89.96 ±4.38 / 96.88 ms │           89.17 / 91.79 ±2.25 / 95.31 ms │     no change │
│ QQuery 88 │        102.51 / 103.39 ±1.32 / 105.98 ms │        100.84 / 101.52 ±0.47 / 102.26 ms │     no change │
│ QQuery 89 │        119.41 / 121.36 ±1.12 / 122.45 ms │        119.73 / 120.83 ±0.66 / 121.79 ms │     no change │
│ QQuery 90 │           24.41 / 24.80 ±0.45 / 25.61 ms │           24.26 / 24.91 ±0.65 / 25.92 ms │     no change │
│ QQuery 91 │           61.71 / 65.38 ±2.30 / 68.78 ms │           65.19 / 66.38 ±0.85 / 67.42 ms │     no change │
│ QQuery 92 │           59.06 / 60.08 ±0.84 / 61.51 ms │           57.69 / 58.99 ±0.82 / 60.25 ms │     no change │
│ QQuery 93 │        194.09 / 195.72 ±1.20 / 197.83 ms │        191.61 / 196.18 ±2.89 / 199.36 ms │     no change │
│ QQuery 94 │           63.43 / 64.39 ±1.06 / 65.98 ms │           61.79 / 63.30 ±1.18 / 64.69 ms │     no change │
│ QQuery 95 │        135.06 / 136.80 ±1.14 / 137.86 ms │        133.29 / 136.13 ±1.90 / 138.31 ms │     no change │
│ QQuery 96 │           76.17 / 77.14 ±0.80 / 78.39 ms │           72.32 / 75.21 ±1.90 / 78.30 ms │     no change │
│ QQuery 97 │        131.81 / 133.27 ±1.00 / 134.38 ms │        131.38 / 133.89 ±2.18 / 137.20 ms │     no change │
│ QQuery 98 │        156.20 / 158.79 ±1.38 / 160.08 ms │        156.37 / 158.02 ±2.03 / 162.03 ms │     no change │
│ QQuery 99 │ 10745.77 / 10791.94 ±41.43 / 10856.28 ms │ 10835.33 / 10863.00 ±20.89 / 10894.31 ms │     no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                │ 34636.22ms │
│ Total Time (alamb_split_bloom)   │ 34106.99ms │
│ Average Time (HEAD)              │   349.86ms │
│ Average Time (alamb_split_bloom) │   344.52ms │
│ Queries Faster                   │          8 │
│ Queries Slower                   │          0 │
│ Queries with No Change           │         91 │
│ Queries with Failure             │          0 │
└──────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	173.7s
Peak memory	5.8 GiB
Avg memory	4.8 GiB
CPU user	279.1s
CPU sys	19.4s
Disk read	0 B
Disk write	638.2 MiB

tpcds — branch

Metric	Value
Wall time	170.8s
Peak memory	5.3 GiB
Avg memory	4.4 GiB
CPU user	273.0s
CPU sys	20.0s
Disk read	0 B
Disk write	864.0 KiB

File an issue against this benchmark runner

adriangbot · 2026-03-31T13:22:19Z

🤖 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 alamb_split_bloom
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                      alamb_split_bloom ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.31 / 4.58 ±6.36 / 17.31 ms │           1.30 / 4.55 ±6.40 / 17.36 ms │     no change │
│ QQuery 1  │        14.35 / 14.81 ±0.24 / 15.00 ms │         14.43 / 14.69 ±0.15 / 14.88 ms │     no change │
│ QQuery 2  │        44.74 / 45.09 ±0.40 / 45.84 ms │         44.37 / 44.86 ±0.35 / 45.31 ms │     no change │
│ QQuery 3  │        44.22 / 45.19 ±0.54 / 45.89 ms │         43.21 / 45.22 ±1.31 / 46.98 ms │     no change │
│ QQuery 4  │     293.89 / 299.10 ±3.47 / 303.69 ms │     284.53 / 302.15 ±13.74 / 320.88 ms │     no change │
│ QQuery 5  │     347.45 / 352.66 ±3.00 / 356.26 ms │      342.45 / 351.52 ±4.70 / 355.97 ms │     no change │
│ QQuery 6  │          6.51 / 8.64 ±2.65 / 13.19 ms │            5.36 / 5.96 ±0.62 / 7.12 ms │ +1.45x faster │
│ QQuery 7  │        16.69 / 16.92 ±0.12 / 17.03 ms │         16.80 / 17.02 ±0.19 / 17.34 ms │     no change │
│ QQuery 8  │     410.76 / 421.69 ±7.65 / 431.64 ms │     414.72 / 432.89 ±13.75 / 453.99 ms │     no change │
│ QQuery 9  │    651.56 / 660.97 ±10.60 / 681.33 ms │     635.15 / 656.40 ±15.68 / 681.32 ms │     no change │
│ QQuery 10 │        91.95 / 95.82 ±3.15 / 99.85 ms │        90.02 / 94.51 ±4.47 / 102.98 ms │     no change │
│ QQuery 11 │     105.21 / 106.62 ±1.18 / 108.66 ms │      103.76 / 104.51 ±0.61 / 105.48 ms │     no change │
│ QQuery 12 │     349.12 / 359.11 ±9.91 / 376.50 ms │      346.22 / 360.33 ±7.96 / 368.13 ms │     no change │
│ QQuery 13 │    458.05 / 472.05 ±11.13 / 484.23 ms │      463.01 / 473.83 ±8.68 / 486.19 ms │     no change │
│ QQuery 14 │     347.37 / 352.27 ±2.91 / 355.21 ms │      348.88 / 354.79 ±3.09 / 357.64 ms │     no change │
│ QQuery 15 │    358.13 / 393.16 ±39.93 / 469.89 ms │     366.00 / 391.75 ±30.26 / 439.76 ms │     no change │
│ QQuery 16 │    715.62 / 744.36 ±16.53 / 765.85 ms │     727.59 / 756.16 ±23.37 / 796.42 ms │     no change │
│ QQuery 17 │     706.48 / 714.80 ±6.35 / 726.15 ms │     717.01 / 738.22 ±22.27 / 781.27 ms │     no change │
│ QQuery 18 │ 1418.45 / 1465.60 ±29.82 / 1508.34 ms │  1394.68 / 1476.02 ±41.73 / 1510.85 ms │     no change │
│ QQuery 19 │       35.54 / 43.86 ±12.43 / 68.59 ms │       34.15 / 62.82 ±30.07 / 106.27 ms │  1.43x slower │
│ QQuery 20 │    715.56 / 735.93 ±17.10 / 760.58 ms │     704.65 / 726.84 ±27.92 / 781.92 ms │     no change │
│ QQuery 21 │    762.98 / 780.57 ±12.76 / 801.03 ms │     750.04 / 769.09 ±13.92 / 786.24 ms │     no change │
│ QQuery 22 │  1127.67 / 1137.27 ±6.64 / 1143.54 ms │   1115.70 / 1122.45 ±9.15 / 1140.49 ms │     no change │
│ QQuery 23 │ 3078.41 / 3099.56 ±12.02 / 3112.98 ms │  3018.81 / 3040.78 ±13.38 / 3058.71 ms │     no change │
│ QQuery 24 │      98.25 / 100.53 ±2.70 / 105.22 ms │       97.19 / 101.42 ±4.20 / 109.48 ms │     no change │
│ QQuery 25 │     139.62 / 141.58 ±1.96 / 145.31 ms │      137.80 / 141.06 ±2.87 / 145.30 ms │     no change │
│ QQuery 26 │     101.88 / 103.34 ±0.95 / 104.41 ms │       96.83 / 102.47 ±3.59 / 107.18 ms │     no change │
│ QQuery 27 │     846.09 / 855.04 ±7.92 / 869.92 ms │      838.58 / 843.67 ±6.13 / 855.55 ms │     no change │
│ QQuery 28 │ 7723.30 / 7758.94 ±37.49 / 7807.64 ms │  7690.94 / 7748.73 ±42.40 / 7808.45 ms │     no change │
│ QQuery 29 │     53.92 / 110.66 ±63.02 / 202.47 ms │         50.52 / 54.99 ±5.19 / 64.81 ms │ +2.01x faster │
│ QQuery 30 │     361.94 / 371.24 ±7.13 / 382.37 ms │      359.86 / 367.78 ±5.60 / 374.60 ms │     no change │
│ QQuery 31 │    359.87 / 386.41 ±18.78 / 405.73 ms │      366.42 / 375.88 ±9.42 / 390.52 ms │     no change │
│ QQuery 32 │ 1123.85 / 1265.44 ±72.53 / 1331.36 ms │ 1100.58 / 1364.91 ±192.39 / 1672.93 ms │  1.08x slower │
│ QQuery 33 │ 1453.43 / 1501.59 ±42.26 / 1566.77 ms │  1491.96 / 1584.24 ±88.37 / 1752.11 ms │  1.06x slower │
│ QQuery 34 │ 1441.10 / 1459.35 ±13.31 / 1481.45 ms │  1458.01 / 1499.20 ±32.51 / 1548.29 ms │     no change │
│ QQuery 35 │     380.75 / 388.40 ±6.69 / 399.79 ms │      398.61 / 405.44 ±8.02 / 420.64 ms │     no change │
│ QQuery 36 │     112.33 / 120.17 ±5.91 / 129.14 ms │      122.98 / 124.71 ±1.81 / 127.22 ms │     no change │
│ QQuery 37 │        48.35 / 50.65 ±1.69 / 53.24 ms │         49.04 / 51.04 ±1.73 / 53.86 ms │     no change │
│ QQuery 38 │        75.14 / 76.50 ±1.08 / 78.47 ms │         76.28 / 77.55 ±1.02 / 78.90 ms │     no change │
│ QQuery 39 │     213.43 / 217.79 ±5.71 / 228.58 ms │      214.21 / 224.92 ±8.78 / 236.77 ms │     no change │
│ QQuery 40 │        22.76 / 26.41 ±2.25 / 29.42 ms │         23.60 / 26.57 ±2.01 / 29.30 ms │     no change │
│ QQuery 41 │        19.60 / 21.49 ±1.22 / 22.91 ms │         20.14 / 21.49 ±1.15 / 23.53 ms │     no change │
│ QQuery 42 │        19.23 / 19.59 ±0.23 / 19.84 ms │         19.14 / 20.19 ±0.66 / 20.92 ms │     no change │
└───────────┴───────────────────────────────────────┴────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                │ 27345.72ms │
│ Total Time (alamb_split_bloom)   │ 27483.64ms │
│ Average Time (HEAD)              │   635.95ms │
│ Average Time (alamb_split_bloom) │   639.15ms │
│ Queries Faster                   │          2 │
│ Queries Slower                   │          3 │
│ Queries with No Change           │         38 │
│ Queries with Failure             │          0 │
└──────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	137.9s
Peak memory	40.2 GiB
Avg memory	29.3 GiB
CPU user	1282.0s
CPU sys	102.1s
Disk read	0 B
Disk write	3.8 GiB

clickbench_partitioned — branch

Metric	Value
Wall time	138.6s
Peak memory	40.2 GiB
Avg memory	29.1 GiB
CPU user	1285.5s
CPU sys	106.9s
Disk read	0 B
Disk write	768.0 KiB

File an issue against this benchmark runner

Dandandan · 2026-03-31T17:02:03Z

run benchmarks

adriangbot · 2026-03-31T17:04:47Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4164094967-633-jznlb 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 alamb/split_bloom (bf0d992) to 6412c3a (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T17:04:57Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4164094967-631-2wp5r 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 alamb/split_bloom (bf0d992) to 6412c3a (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T17:04:57Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4164094967-632-4smwn 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 alamb/split_bloom (bf0d992) to 6412c3a (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-03-31T17:17:26Z

🤖 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

Comparing HEAD and alamb_split_bloom
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃              alamb_split_bloom ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 45.74 / 46.42 ±0.74 / 47.81 ms │ 45.76 / 46.31 ±0.82 / 47.94 ms │ no change │
│ QQuery 2  │ 21.82 / 22.35 ±0.54 / 23.38 ms │ 21.52 / 21.73 ±0.16 / 21.98 ms │ no change │
│ QQuery 3  │ 32.27 / 32.79 ±0.29 / 33.13 ms │ 32.23 / 32.40 ±0.19 / 32.77 ms │ no change │
│ QQuery 4  │ 20.72 / 21.48 ±0.84 / 22.99 ms │ 20.60 / 21.59 ±0.86 / 22.99 ms │ no change │
│ QQuery 5  │ 48.31 / 50.05 ±1.46 / 52.67 ms │ 48.12 / 50.07 ±1.37 / 51.76 ms │ no change │
│ QQuery 6  │ 17.52 / 17.65 ±0.09 / 17.75 ms │ 17.40 / 18.06 ±1.00 / 20.04 ms │ no change │
│ QQuery 7  │ 54.87 / 57.08 ±1.25 / 58.35 ms │ 55.33 / 55.84 ±0.67 / 57.08 ms │ no change │
│ QQuery 8  │ 48.88 / 49.80 ±1.27 / 52.30 ms │ 49.48 / 49.97 ±0.61 / 51.15 ms │ no change │
│ QQuery 9  │ 55.80 / 56.53 ±0.46 / 57.13 ms │ 54.55 / 55.87 ±1.08 / 57.66 ms │ no change │
│ QQuery 10 │ 73.35 / 74.75 ±1.69 / 77.77 ms │ 71.74 / 73.06 ±1.21 / 74.55 ms │ no change │
│ QQuery 11 │ 14.79 / 15.00 ±0.21 / 15.38 ms │ 14.55 / 14.78 ±0.14 / 14.95 ms │ no change │
│ QQuery 12 │ 28.47 / 29.35 ±0.88 / 30.96 ms │ 28.46 / 28.89 ±0.31 / 29.42 ms │ no change │
│ QQuery 13 │ 38.53 / 39.80 ±0.78 / 40.71 ms │ 39.29 / 39.64 ±0.50 / 40.62 ms │ no change │
│ QQuery 14 │ 28.95 / 29.21 ±0.25 / 29.60 ms │ 29.25 / 29.54 ±0.25 / 29.86 ms │ no change │
│ QQuery 15 │ 34.19 / 34.81 ±0.75 / 36.15 ms │ 34.56 / 35.52 ±0.73 / 36.79 ms │ no change │
│ QQuery 16 │ 16.80 / 17.09 ±0.19 / 17.27 ms │ 16.33 / 16.77 ±0.30 / 17.20 ms │ no change │
│ QQuery 17 │ 74.16 / 74.98 ±0.65 / 76.06 ms │ 76.36 / 76.66 ±0.38 / 77.41 ms │ no change │
│ QQuery 18 │ 80.97 / 81.31 ±0.18 / 81.47 ms │ 77.16 / 80.17 ±2.73 / 85.06 ms │ no change │
│ QQuery 19 │ 38.11 / 38.90 ±1.06 / 40.99 ms │ 38.04 / 38.49 ±0.32 / 38.98 ms │ no change │
│ QQuery 20 │ 40.93 / 41.79 ±0.58 / 42.41 ms │ 41.00 / 42.08 ±0.67 / 42.90 ms │ no change │
│ QQuery 21 │ 65.24 / 67.23 ±1.68 / 69.44 ms │ 65.42 / 68.89 ±2.58 / 72.08 ms │ no change │
│ QQuery 22 │ 18.26 / 18.86 ±0.56 / 19.84 ms │ 17.91 / 18.51 ±0.46 / 19.26 ms │ no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                │ 917.25ms │
│ Total Time (alamb_split_bloom)   │ 914.86ms │
│ Average Time (HEAD)              │  41.69ms │
│ Average Time (alamb_split_bloom) │  41.58ms │
│ 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	34.0s
CPU sys	3.0s
Disk read	0 B
Disk write	140.0 KiB

tpch — branch

Metric	Value
Wall time	4.8s
Peak memory	4.0 GiB
Avg memory	3.6 GiB
CPU user	34.0s
CPU sys	3.1s
Disk read	0 B
Disk write	60.0 KiB

File an issue against this benchmark runner

adriangbot · 2026-03-31T17:23:00Z

🤖 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 alamb_split_bloom
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                                     HEAD ┃                        alamb_split_bloom ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │           43.58 / 44.61 ±0.78 / 45.75 ms │           44.15 / 44.77 ±0.55 / 45.49 ms │    no change │
│ QQuery 2  │        146.74 / 147.69 ±0.63 / 148.61 ms │        146.12 / 146.80 ±0.73 / 148.16 ms │    no change │
│ QQuery 3  │        115.15 / 116.20 ±0.89 / 117.50 ms │        114.26 / 114.85 ±0.72 / 116.20 ms │    no change │
│ QQuery 4  │    1384.77 / 1419.65 ±25.08 / 1454.21 ms │    1375.40 / 1401.90 ±14.50 / 1419.46 ms │    no change │
│ QQuery 5  │        175.07 / 176.00 ±0.68 / 176.77 ms │        174.04 / 175.42 ±1.08 / 177.38 ms │    no change │
│ QQuery 6  │     997.74 / 1032.91 ±22.95 / 1054.79 ms │    1009.14 / 1027.08 ±21.68 / 1067.19 ms │    no change │
│ QQuery 7  │        355.49 / 359.20 ±2.29 / 361.84 ms │        354.09 / 358.03 ±3.20 / 363.49 ms │    no change │
│ QQuery 8  │        117.30 / 118.54 ±1.09 / 120.16 ms │        116.57 / 117.86 ±0.73 / 118.60 ms │    no change │
│ QQuery 9  │        101.49 / 105.14 ±2.73 / 107.55 ms │        101.16 / 103.57 ±2.59 / 107.64 ms │    no change │
│ QQuery 10 │        111.11 / 112.00 ±0.72 / 112.88 ms │        107.88 / 109.36 ±1.24 / 110.90 ms │    no change │
│ QQuery 11 │      992.84 / 1004.86 ±6.58 / 1010.53 ms │        954.81 / 964.46 ±7.73 / 978.44 ms │    no change │
│ QQuery 12 │           46.28 / 47.66 ±0.94 / 48.95 ms │           44.98 / 47.35 ±1.66 / 49.37 ms │    no change │
│ QQuery 13 │        406.79 / 412.62 ±4.62 / 420.86 ms │        402.10 / 403.35 ±0.79 / 404.59 ms │    no change │
│ QQuery 14 │     1039.94 / 1045.84 ±4.69 / 1051.73 ms │     1024.31 / 1032.82 ±5.66 / 1039.30 ms │    no change │
│ QQuery 15 │           16.57 / 17.97 ±1.01 / 19.57 ms │           16.50 / 17.23 ±0.67 / 18.28 ms │    no change │
│ QQuery 16 │           42.89 / 43.84 ±0.92 / 45.23 ms │           41.61 / 42.61 ±0.54 / 43.22 ms │    no change │
│ QQuery 17 │        244.31 / 246.03 ±1.74 / 249.11 ms │        242.09 / 245.25 ±2.86 / 250.57 ms │    no change │
│ QQuery 18 │        133.77 / 135.99 ±1.89 / 138.51 ms │        130.13 / 131.74 ±1.41 / 134.31 ms │    no change │
│ QQuery 19 │        160.47 / 161.78 ±0.81 / 162.95 ms │        157.27 / 159.24 ±1.22 / 160.75 ms │    no change │
│ QQuery 20 │           14.67 / 14.97 ±0.22 / 15.25 ms │           14.22 / 14.71 ±0.32 / 15.13 ms │    no change │
│ QQuery 21 │           20.22 / 20.68 ±0.35 / 21.22 ms │           19.91 / 21.13 ±1.02 / 22.84 ms │    no change │
│ QQuery 22 │        500.68 / 503.64 ±2.27 / 507.51 ms │        501.09 / 504.37 ±2.72 / 508.95 ms │    no change │
│ QQuery 23 │        922.79 / 932.39 ±8.15 / 946.05 ms │       905.06 / 921.52 ±14.03 / 947.41 ms │    no change │
│ QQuery 24 │        420.30 / 423.27 ±2.25 / 426.26 ms │        423.44 / 426.22 ±1.75 / 428.27 ms │    no change │
│ QQuery 25 │        356.40 / 359.99 ±2.11 / 362.91 ms │        355.37 / 357.22 ±1.86 / 359.97 ms │    no change │
│ QQuery 26 │           84.40 / 86.44 ±1.69 / 89.10 ms │           81.88 / 83.81 ±1.64 / 86.62 ms │    no change │
│ QQuery 27 │        350.91 / 353.87 ±2.41 / 357.43 ms │        351.01 / 356.04 ±3.02 / 359.69 ms │    no change │
│ QQuery 28 │        149.84 / 151.80 ±1.74 / 154.88 ms │        150.23 / 151.42 ±1.36 / 153.98 ms │    no change │
│ QQuery 29 │        300.93 / 305.14 ±3.03 / 309.98 ms │        302.74 / 304.74 ±1.93 / 308.42 ms │    no change │
│ QQuery 30 │           43.56 / 45.88 ±1.51 / 47.45 ms │           45.37 / 46.35 ±0.64 / 47.28 ms │    no change │
│ QQuery 31 │        171.55 / 176.16 ±2.92 / 180.36 ms │        173.09 / 174.70 ±1.44 / 177.27 ms │    no change │
│ QQuery 32 │           58.32 / 60.22 ±1.84 / 63.04 ms │           57.43 / 57.87 ±0.65 / 59.15 ms │    no change │
│ QQuery 33 │        141.78 / 144.25 ±1.35 / 145.61 ms │        141.19 / 143.04 ±1.17 / 144.33 ms │    no change │
│ QQuery 34 │        107.75 / 109.07 ±0.77 / 109.93 ms │        108.06 / 108.80 ±0.50 / 109.35 ms │    no change │
│ QQuery 35 │        108.32 / 111.17 ±1.76 / 113.20 ms │        111.11 / 112.00 ±0.94 / 113.72 ms │    no change │
│ QQuery 36 │        217.33 / 222.44 ±2.63 / 224.90 ms │        214.11 / 222.07 ±5.75 / 228.26 ms │    no change │
│ QQuery 37 │        180.37 / 182.10 ±2.22 / 186.22 ms │        178.75 / 181.64 ±1.59 / 183.53 ms │    no change │
│ QQuery 38 │           86.38 / 89.90 ±2.45 / 93.15 ms │           85.24 / 89.19 ±2.04 / 90.82 ms │    no change │
│ QQuery 39 │        130.73 / 133.24 ±2.83 / 136.88 ms │        131.47 / 133.64 ±1.21 / 135.11 ms │    no change │
│ QQuery 40 │        115.45 / 121.64 ±8.42 / 137.42 ms │        109.87 / 118.54 ±6.64 / 129.83 ms │    no change │
│ QQuery 41 │           15.54 / 16.00 ±0.32 / 16.42 ms │           14.58 / 16.37 ±2.15 / 20.22 ms │    no change │
│ QQuery 42 │        108.80 / 110.35 ±1.41 / 112.59 ms │        109.47 / 110.15 ±0.53 / 111.06 ms │    no change │
│ QQuery 43 │           84.22 / 86.61 ±2.65 / 91.73 ms │           84.96 / 85.25 ±0.24 / 85.60 ms │    no change │
│ QQuery 44 │           11.69 / 12.37 ±0.86 / 14.04 ms │           11.93 / 12.42 ±0.42 / 13.09 ms │    no change │
│ QQuery 45 │           54.28 / 55.55 ±1.76 / 59.04 ms │           54.27 / 55.52 ±1.04 / 57.09 ms │    no change │
│ QQuery 46 │        236.87 / 238.95 ±1.85 / 242.30 ms │        242.60 / 245.49 ±2.66 / 250.41 ms │    no change │
│ QQuery 47 │       733.59 / 749.35 ±12.32 / 763.95 ms │        792.82 / 800.82 ±6.64 / 812.08 ms │ 1.07x slower │
│ QQuery 48 │        288.55 / 292.21 ±3.23 / 297.04 ms │        293.71 / 298.79 ±3.26 / 303.01 ms │    no change │
│ QQuery 49 │        256.15 / 258.22 ±3.04 / 264.17 ms │        257.79 / 261.38 ±1.97 / 263.28 ms │    no change │
│ QQuery 50 │        237.87 / 242.61 ±3.10 / 247.36 ms │        234.43 / 241.81 ±4.72 / 246.95 ms │    no change │
│ QQuery 51 │        182.46 / 184.49 ±1.32 / 186.19 ms │        184.45 / 188.38 ±2.93 / 192.21 ms │    no change │
│ QQuery 52 │        108.61 / 109.83 ±0.77 / 110.73 ms │        108.17 / 110.13 ±1.66 / 112.82 ms │    no change │
│ QQuery 53 │        104.07 / 104.91 ±0.69 / 105.73 ms │        103.61 / 104.05 ±0.45 / 104.66 ms │    no change │
│ QQuery 54 │        149.70 / 150.99 ±0.74 / 151.87 ms │        149.44 / 150.75 ±0.99 / 152.11 ms │    no change │
│ QQuery 55 │        107.67 / 108.95 ±1.14 / 111.10 ms │        107.72 / 109.18 ±1.10 / 110.36 ms │    no change │
│ QQuery 56 │        142.03 / 143.38 ±1.97 / 147.28 ms │        142.97 / 144.07 ±0.74 / 145.09 ms │    no change │
│ QQuery 57 │        177.92 / 178.41 ±0.55 / 179.45 ms │        174.66 / 178.23 ±2.80 / 183.16 ms │    no change │
│ QQuery 58 │        296.99 / 307.48 ±5.52 / 312.00 ms │        300.29 / 303.71 ±3.88 / 309.46 ms │    no change │
│ QQuery 59 │        199.79 / 200.63 ±0.64 / 201.44 ms │        200.15 / 202.27 ±1.95 / 205.69 ms │    no change │
│ QQuery 60 │        145.11 / 146.32 ±1.10 / 147.94 ms │        146.95 / 148.00 ±0.92 / 149.51 ms │    no change │
│ QQuery 61 │        169.38 / 172.53 ±1.81 / 174.28 ms │        170.68 / 174.16 ±2.19 / 177.38 ms │    no change │
│ QQuery 62 │       896.56 / 920.21 ±16.94 / 938.81 ms │       911.73 / 933.83 ±16.20 / 958.26 ms │    no change │
│ QQuery 63 │        103.45 / 106.80 ±2.14 / 109.91 ms │        103.69 / 106.29 ±1.69 / 108.25 ms │    no change │
│ QQuery 64 │        709.23 / 713.82 ±3.92 / 719.94 ms │        713.49 / 722.08 ±7.75 / 736.05 ms │    no change │
│ QQuery 65 │        257.21 / 258.07 ±0.47 / 258.64 ms │        256.93 / 261.87 ±4.10 / 266.54 ms │    no change │
│ QQuery 66 │       245.82 / 257.82 ±12.22 / 279.12 ms │       234.96 / 264.15 ±15.07 / 276.89 ms │    no change │
│ QQuery 67 │        318.09 / 324.24 ±3.79 / 328.36 ms │        311.97 / 324.22 ±7.54 / 333.07 ms │    no change │
│ QQuery 68 │        286.46 / 288.80 ±1.85 / 291.35 ms │        286.58 / 291.55 ±4.45 / 298.49 ms │    no change │
│ QQuery 69 │        107.11 / 108.47 ±0.98 / 109.55 ms │        103.70 / 104.63 ±0.61 / 105.34 ms │    no change │
│ QQuery 70 │        332.72 / 346.42 ±8.87 / 355.77 ms │        344.84 / 350.70 ±3.51 / 354.55 ms │    no change │
│ QQuery 71 │        136.73 / 139.07 ±2.51 / 143.71 ms │        136.03 / 137.37 ±1.02 / 139.07 ms │    no change │
│ QQuery 72 │       709.86 / 733.00 ±11.87 / 743.62 ms │        743.36 / 747.62 ±3.48 / 753.23 ms │    no change │
│ QQuery 73 │        103.30 / 106.72 ±2.19 / 109.62 ms │        104.00 / 105.68 ±2.16 / 109.47 ms │    no change │
│ QQuery 74 │       603.45 / 615.34 ±18.80 / 652.85 ms │        623.85 / 631.15 ±5.36 / 636.35 ms │    no change │
│ QQuery 75 │        282.56 / 284.97 ±2.54 / 289.50 ms │        280.63 / 283.47 ±3.00 / 288.98 ms │    no change │
│ QQuery 76 │        134.72 / 136.88 ±2.26 / 141.02 ms │        135.07 / 137.23 ±1.56 / 139.42 ms │    no change │
│ QQuery 77 │        192.20 / 193.39 ±1.35 / 195.35 ms │        189.85 / 190.65 ±0.89 / 192.38 ms │    no change │
│ QQuery 78 │        355.64 / 360.49 ±3.67 / 364.85 ms │        353.48 / 359.84 ±3.59 / 363.18 ms │    no change │
│ QQuery 79 │        244.57 / 249.15 ±3.00 / 254.02 ms │        236.74 / 242.15 ±3.63 / 246.45 ms │    no change │
│ QQuery 80 │        329.35 / 334.57 ±3.13 / 337.72 ms │        333.02 / 334.76 ±1.61 / 336.77 ms │    no change │
│ QQuery 81 │           27.68 / 28.96 ±1.01 / 30.50 ms │           26.61 / 27.59 ±0.69 / 28.78 ms │    no change │
│ QQuery 82 │        200.41 / 203.92 ±2.98 / 207.76 ms │        199.43 / 203.96 ±2.64 / 207.40 ms │    no change │
│ QQuery 83 │           40.02 / 41.92 ±1.45 / 43.77 ms │           41.67 / 42.66 ±1.03 / 44.59 ms │    no change │
│ QQuery 84 │           50.38 / 50.88 ±0.75 / 52.34 ms │           49.66 / 50.63 ±0.76 / 51.95 ms │    no change │
│ QQuery 85 │        150.22 / 152.71 ±1.72 / 154.81 ms │        149.09 / 153.30 ±2.42 / 155.38 ms │    no change │
│ QQuery 86 │           40.52 / 40.86 ±0.21 / 41.14 ms │           39.32 / 40.12 ±0.47 / 40.60 ms │    no change │
│ QQuery 87 │           87.00 / 90.20 ±3.53 / 96.73 ms │           86.74 / 89.71 ±2.10 / 92.21 ms │    no change │
│ QQuery 88 │        101.56 / 102.60 ±1.03 / 104.40 ms │        101.16 / 102.43 ±1.48 / 105.23 ms │    no change │
│ QQuery 89 │        117.70 / 120.08 ±1.62 / 122.53 ms │        119.06 / 120.28 ±0.91 / 121.53 ms │    no change │
│ QQuery 90 │           24.42 / 24.84 ±0.29 / 25.22 ms │           24.50 / 24.86 ±0.27 / 25.16 ms │    no change │
│ QQuery 91 │           63.87 / 67.61 ±1.89 / 68.90 ms │           63.80 / 66.22 ±1.62 / 68.70 ms │    no change │
│ QQuery 92 │           58.46 / 59.35 ±0.61 / 60.35 ms │           57.81 / 58.95 ±0.74 / 59.79 ms │    no change │
│ QQuery 93 │        190.66 / 192.45 ±2.04 / 196.33 ms │        194.60 / 196.65 ±1.38 / 198.50 ms │    no change │
│ QQuery 94 │           61.90 / 63.38 ±0.77 / 64.07 ms │           63.32 / 64.44 ±1.36 / 67.11 ms │    no change │
│ QQuery 95 │        134.71 / 135.86 ±1.04 / 137.32 ms │        134.70 / 136.79 ±1.32 / 138.83 ms │    no change │
│ QQuery 96 │           70.49 / 74.48 ±2.76 / 78.55 ms │           73.97 / 75.16 ±0.83 / 76.23 ms │    no change │
│ QQuery 97 │        130.54 / 133.97 ±3.29 / 140.24 ms │        131.52 / 132.70 ±0.88 / 133.79 ms │    no change │
│ QQuery 98 │        156.48 / 158.45 ±2.37 / 162.90 ms │        153.49 / 158.10 ±2.47 / 160.56 ms │    no change │
│ QQuery 99 │ 10803.74 / 10845.06 ±35.91 / 10907.18 ms │ 10834.49 / 10860.20 ±19.12 / 10886.64 ms │    no change │
└───────────┴──────────────────────────────────────────┴──────────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                │ 34232.71ms │
│ Total Time (alamb_split_bloom)   │ 34247.58ms │
│ Average Time (HEAD)              │   345.78ms │
│ Average Time (alamb_split_bloom) │   345.94ms │
│ 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	171.5s
Peak memory	5.6 GiB
Avg memory	4.7 GiB
CPU user	275.3s
CPU sys	18.7s
Disk read	0 B
Disk write	638.1 MiB

tpcds — branch

Metric	Value
Wall time	171.5s
Peak memory	5.7 GiB
Avg memory	4.7 GiB
CPU user	275.4s
CPU sys	18.4s
Disk read	0 B
Disk write	152.0 KiB

File an issue against this benchmark runner

adriangbot · 2026-03-31T17:25:01Z

🤖 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 alamb_split_bloom
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                     alamb_split_bloom ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.34 / 4.62 ±6.44 / 17.50 ms │          1.32 / 4.63 ±6.46 / 17.55 ms │     no change │
│ QQuery 1  │        14.53 / 15.07 ±0.31 / 15.37 ms │        14.44 / 15.03 ±0.35 / 15.49 ms │     no change │
│ QQuery 2  │        44.98 / 45.34 ±0.28 / 45.65 ms │        45.33 / 45.78 ±0.35 / 46.38 ms │     no change │
│ QQuery 3  │        44.28 / 45.32 ±1.05 / 46.86 ms │        43.88 / 45.19 ±1.03 / 46.61 ms │     no change │
│ QQuery 4  │     294.20 / 298.12 ±4.28 / 306.10 ms │    295.58 / 305.94 ±10.18 / 322.19 ms │     no change │
│ QQuery 5  │     349.31 / 353.11 ±5.65 / 364.24 ms │     351.03 / 355.82 ±2.62 / 359.07 ms │     no change │
│ QQuery 6  │           6.38 / 7.20 ±0.93 / 8.97 ms │           6.04 / 7.32 ±1.06 / 9.20 ms │     no change │
│ QQuery 7  │        17.06 / 17.17 ±0.06 / 17.25 ms │        17.39 / 19.61 ±2.06 / 23.24 ms │  1.14x slower │
│ QQuery 8  │    427.83 / 439.79 ±12.62 / 458.29 ms │    439.84 / 455.26 ±14.20 / 482.09 ms │     no change │
│ QQuery 9  │     653.29 / 656.42 ±3.09 / 660.68 ms │     672.93 / 682.43 ±5.28 / 688.92 ms │     no change │
│ QQuery 10 │       91.39 / 95.61 ±3.15 / 100.89 ms │        91.27 / 94.01 ±2.13 / 97.08 ms │     no change │
│ QQuery 11 │     106.27 / 107.84 ±0.97 / 109.17 ms │     105.17 / 106.15 ±1.11 / 107.62 ms │     no change │
│ QQuery 12 │     336.16 / 347.67 ±6.94 / 357.75 ms │     348.95 / 354.26 ±4.35 / 360.47 ms │     no change │
│ QQuery 13 │     463.95 / 475.00 ±8.12 / 486.10 ms │     450.04 / 464.82 ±9.57 / 473.75 ms │     no change │
│ QQuery 14 │     351.03 / 355.90 ±3.74 / 360.37 ms │     354.47 / 362.47 ±6.47 / 370.81 ms │     no change │
│ QQuery 15 │    359.73 / 382.57 ±24.12 / 419.89 ms │    365.79 / 389.60 ±31.08 / 448.68 ms │     no change │
│ QQuery 16 │    724.40 / 751.95 ±17.47 / 773.49 ms │     726.59 / 736.85 ±8.98 / 751.43 ms │     no change │
│ QQuery 17 │     719.62 / 725.54 ±4.88 / 733.90 ms │    730.78 / 751.05 ±25.71 / 801.30 ms │     no change │
│ QQuery 18 │ 1444.88 / 1491.96 ±29.44 / 1534.23 ms │ 1469.34 / 1495.95 ±18.66 / 1522.55 ms │     no change │
│ QQuery 19 │        36.26 / 38.87 ±1.78 / 41.06 ms │        35.76 / 36.92 ±1.39 / 39.61 ms │ +1.05x faster │
│ QQuery 20 │    714.51 / 731.88 ±21.70 / 773.01 ms │    717.69 / 739.38 ±17.17 / 764.44 ms │     no change │
│ QQuery 21 │     759.29 / 762.06 ±3.09 / 767.75 ms │     759.35 / 766.94 ±4.87 / 772.43 ms │     no change │
│ QQuery 22 │  1127.39 / 1132.71 ±7.54 / 1147.57 ms │  1122.29 / 1128.24 ±4.46 / 1134.13 ms │     no change │
│ QQuery 23 │  3098.25 / 3108.70 ±8.90 / 3118.52 ms │  3071.38 / 3075.42 ±5.14 / 3085.44 ms │     no change │
│ QQuery 24 │      98.14 / 102.09 ±4.51 / 110.30 ms │     100.13 / 102.60 ±1.45 / 104.46 ms │     no change │
│ QQuery 25 │     138.00 / 142.72 ±2.49 / 144.81 ms │     139.84 / 141.88 ±1.20 / 143.37 ms │     no change │
│ QQuery 26 │      95.17 / 100.38 ±2.85 / 102.66 ms │       96.91 / 99.80 ±1.64 / 101.67 ms │     no change │
│ QQuery 27 │     849.06 / 853.61 ±4.61 / 861.40 ms │     845.72 / 852.14 ±6.60 / 863.14 ms │     no change │
│ QQuery 28 │ 7707.42 / 7777.06 ±37.18 / 7810.65 ms │ 7690.30 / 7761.35 ±36.25 / 7790.10 ms │     no change │
│ QQuery 29 │        50.06 / 55.69 ±4.06 / 60.79 ms │        51.30 / 56.27 ±7.52 / 71.02 ms │     no change │
│ QQuery 30 │    361.58 / 372.01 ±10.90 / 386.43 ms │     363.34 / 378.02 ±8.45 / 386.19 ms │     no change │
│ QQuery 31 │     373.88 / 388.61 ±9.83 / 401.38 ms │    360.89 / 389.81 ±15.14 / 402.18 ms │     no change │
│ QQuery 32 │ 1036.35 / 1053.54 ±10.74 / 1067.43 ms │ 1052.24 / 1087.12 ±48.91 / 1181.21 ms │     no change │
│ QQuery 33 │  1460.21 / 1466.01 ±4.75 / 1472.10 ms │ 1473.70 / 1495.79 ±13.16 / 1510.28 ms │     no change │
│ QQuery 34 │ 1498.70 / 1562.80 ±42.84 / 1625.08 ms │ 1463.86 / 1480.66 ±18.40 / 1506.53 ms │ +1.06x faster │
│ QQuery 35 │     394.50 / 400.57 ±4.88 / 407.29 ms │     394.80 / 406.56 ±8.69 / 421.06 ms │     no change │
│ QQuery 36 │     118.27 / 121.92 ±3.14 / 126.21 ms │     113.19 / 120.89 ±6.14 / 129.54 ms │     no change │
│ QQuery 37 │        47.81 / 48.90 ±0.90 / 50.02 ms │        49.17 / 51.76 ±1.95 / 54.11 ms │  1.06x slower │
│ QQuery 38 │        76.41 / 77.30 ±1.38 / 80.01 ms │        75.11 / 77.64 ±1.70 / 80.38 ms │     no change │
│ QQuery 39 │     216.37 / 225.26 ±5.10 / 231.89 ms │    204.76 / 222.17 ±10.61 / 231.48 ms │     no change │
│ QQuery 40 │        21.92 / 25.08 ±2.00 / 27.89 ms │        25.66 / 28.20 ±2.67 / 32.79 ms │  1.12x slower │
│ QQuery 41 │        21.39 / 22.29 ±0.72 / 23.42 ms │        20.39 / 21.44 ±0.72 / 22.67 ms │     no change │
│ QQuery 42 │        20.27 / 20.73 ±0.43 / 21.38 ms │        20.36 / 20.68 ±0.18 / 20.86 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                │ 27207.00ms │
│ Total Time (alamb_split_bloom)   │ 27233.86ms │
│ Average Time (HEAD)              │   632.72ms │
│ Average Time (alamb_split_bloom) │   633.35ms │
│ Queries Faster                   │          2 │
│ Queries Slower                   │          3 │
│ Queries with No Change           │         38 │
│ Queries with Failure             │          0 │
└──────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	137.2s
Peak memory	40.4 GiB
Avg memory	32.9 GiB
CPU user	1291.2s
CPU sys	90.0s
Disk read	0 B
Disk write	2.1 GiB

clickbench_partitioned — branch

Metric	Value
Wall time	137.4s
Peak memory	42.5 GiB
Avg memory	31.0 GiB
CPU user	1290.9s
CPU sys	91.2s
Disk read	0 B
Disk write	764.0 KiB

File an issue against this benchmark runner

adriangb

I like the direction these refactors are going in @alamb , this is the right way to organize opener.rs (I know at some point we spoke about splitting it up into more modules, this is either a good first step towards that or a better solution).

adriangb · 2026-04-01T12:34:33Z

datafusion/datasource-parquet/src/opener.rs

+                .into_iter()
+                .filter_map(|column_name| {
+                    let parquet_schema = builder.parquet_schema();
+                    let (column_idx, _) = parquet_column(


It'd be nice if we could get this to also support struct columns / struct filters cc @friendlymatthew

alamb · 2026-04-02T17:51:58Z

Thanks again @Dandandan and @adriangb

I am now working on the next steps

…pache#21285) ## Which issue does this PR close? - part of apache#20529 - Broken out of apache#20820 ## Rationale for this change We are try to be explicit about CPU and IO in the Parquet reader, however, the code for applying bloom filters still did both CPU and IO in the same future ## What changes are included in this PR? 1. Split the states for IO and CPU work when applying bloom filters ## Are these changes tested? Functionally by existing tests I also ran performance tests and they didn't show any substantial performance change ## Are there any user-facing changes? No, this is internal code reorganization

alamb mentioned this pull request Mar 31, 2026

Refactor parquet datasource into an explicit state machine #21190

Merged

2 tasks

github-actions bot added the datasource Changes to the datasource crate label Mar 31, 2026

alamb mentioned this pull request Mar 31, 2026

Sketch out a Morselize API #20820

Draft

5 tasks

refactor: Split Parquet BloomFilter CPU and IO into separate states

bf0d992

alamb force-pushed the alamb/split_bloom branch from ac46396 to bf0d992 Compare March 31, 2026 13:07

alamb marked this pull request as ready for review March 31, 2026 14:26

adriangb approved these changes Apr 1, 2026

View reviewed changes

Dandandan approved these changes Apr 2, 2026

View reviewed changes

alamb added this pull request to the merge queue Apr 2, 2026

Merged via the queue into apache:main with commit c8aefab Apr 2, 2026
30 checks passed

alamb deleted the alamb/split_bloom branch April 2, 2026 17:52

Conversation

alamb commented Mar 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

Uh oh!

alamb commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

Dandandan commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangbot commented Mar 31, 2026

Uh oh!

adriangb left a comment

Choose a reason for hiding this comment

Uh oh!

adriangb Apr 1, 2026

Choose a reason for hiding this comment

Uh oh!

alamb commented Apr 2, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

alamb commented Mar 31, 2026 •

edited

Loading