TEST: check performance with sparse masks by alamb · Pull Request #23728 · apache/datafusion

alamb · 2026-07-20T18:48:52Z

related to fix(parquet): support mask filtering across skipped pages arrow-rs#10288 from @hhhizzz

I want to run the datafusion performance tests with apache/arrow-rs#10288 to see how it helps (I don't expect to see any significant difference but I want to see)

alamb · 2026-07-20T18:49:20Z

run benchmarks

alamb · 2026-07-20T18:49:28Z

run benchmark clickbench_pushdown

adriangbot · 2026-07-20T18:51:22Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5025955782-1184-h86rf 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 alamb/test_sparse_mask2222323412 (fe1cedc) to 4184b07 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-20T18:51:52Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5025956778-1187-ckppc 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 alamb/test_sparse_mask2222323412 (fe1cedc) to 4184b07 (merge-base) diff using: clickbench_pushdown
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-20T18:52:11Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5025955782-1185-5sqxw 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 alamb/test_sparse_mask2222323412 (fe1cedc) to 4184b07 (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-20T18:52:17Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5025955782-1186-nxgpt 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 alamb/test_sparse_mask2222323412 (fe1cedc) to 4184b07 (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-20T19:09:18Z

🤖 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_test_sparse_mask2222323412
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ alamb_test_sparse_mask2222323412 ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 38.90 / 40.17 ±1.29 / 41.90 ms │   38.94 / 39.66 ±1.17 / 41.98 ms │ no change │
│ QQuery 2  │ 19.33 / 19.98 ±0.84 / 21.52 ms │   19.29 / 19.73 ±0.33 / 20.29 ms │ no change │
│ QQuery 3  │ 31.48 / 32.91 ±1.03 / 33.91 ms │   31.69 / 32.22 ±0.85 / 33.91 ms │ no change │
│ QQuery 4  │ 17.71 / 17.94 ±0.17 / 18.17 ms │   17.58 / 17.83 ±0.41 / 18.65 ms │ no change │
│ QQuery 5  │ 38.66 / 40.63 ±1.10 / 41.67 ms │   38.00 / 40.19 ±1.40 / 41.61 ms │ no change │
│ QQuery 6  │ 16.50 / 16.76 ±0.31 / 17.33 ms │   16.30 / 16.58 ±0.27 / 17.09 ms │ no change │
│ QQuery 7  │ 45.50 / 47.34 ±2.21 / 51.56 ms │   44.50 / 45.55 ±0.84 / 46.57 ms │ no change │
│ QQuery 8  │ 43.41 / 44.20 ±0.61 / 44.90 ms │   43.52 / 44.08 ±0.61 / 44.82 ms │ no change │
│ QQuery 9  │ 50.58 / 51.22 ±0.53 / 52.09 ms │   49.63 / 50.45 ±0.52 / 51.10 ms │ no change │
│ QQuery 10 │ 42.70 / 43.47 ±1.05 / 45.50 ms │   42.71 / 43.78 ±1.19 / 45.49 ms │ no change │
│ QQuery 11 │ 13.74 / 13.91 ±0.14 / 14.09 ms │   13.92 / 14.07 ±0.13 / 14.26 ms │ no change │
│ QQuery 12 │ 23.56 / 24.00 ±0.35 / 24.62 ms │   23.57 / 24.19 ±0.48 / 24.80 ms │ no change │
│ QQuery 13 │ 33.15 / 34.96 ±1.78 / 37.34 ms │   32.70 / 35.71 ±1.98 / 38.58 ms │ no change │
│ QQuery 14 │ 24.64 / 24.93 ±0.31 / 25.54 ms │   23.86 / 24.48 ±0.51 / 25.27 ms │ no change │
│ QQuery 15 │ 31.81 / 32.17 ±0.33 / 32.71 ms │   31.82 / 32.49 ±0.64 / 33.48 ms │ no change │
│ QQuery 16 │ 14.18 / 14.38 ±0.20 / 14.75 ms │   14.13 / 14.35 ±0.16 / 14.52 ms │ no change │
│ QQuery 17 │ 72.33 / 74.29 ±1.90 / 76.71 ms │   72.11 / 72.67 ±0.32 / 72.99 ms │ no change │
│ QQuery 18 │ 57.92 / 60.72 ±1.69 / 63.14 ms │   59.58 / 60.64 ±1.36 / 63.25 ms │ no change │
│ QQuery 19 │ 33.29 / 34.15 ±0.86 / 35.62 ms │   32.93 / 33.36 ±0.33 / 33.75 ms │ no change │
│ QQuery 20 │ 31.79 / 32.36 ±0.36 / 32.78 ms │   32.10 / 32.29 ±0.13 / 32.47 ms │ no change │
│ QQuery 21 │ 55.60 / 57.02 ±0.91 / 58.29 ms │   56.49 / 58.34 ±1.07 / 59.68 ms │ no change │
│ QQuery 22 │ 13.92 / 14.18 ±0.14 / 14.36 ms │   13.92 / 14.56 ±0.56 / 15.56 ms │ no change │
└───────────┴────────────────────────────────┴──────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                               ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                               │ 771.69ms │
│ Total Time (alamb_test_sparse_mask2222323412)   │ 767.24ms │
│ Average Time (HEAD)                             │  35.08ms │
│ Average Time (alamb_test_sparse_mask2222323412) │  34.87ms │
│ 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	5.0s
Peak memory	1.1 GiB
Avg memory	492.6 MiB
CPU user	22.2s
CPU sys	1.8s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	1.1 GiB
Avg memory	507.5 MiB
CPU user	22.0s
CPU sys	1.7s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-20T19:10:28Z

🤖 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

Comparing HEAD and alamb_test_sparse_mask2222323412
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃      alamb_test_sparse_mask2222323412 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.41 / 5.96 ±0.92 / 7.80 ms │           5.43 / 5.99 ±0.90 / 7.77 ms │     no change │
│ QQuery 2  │        81.13 / 81.55 ±0.30 / 81.91 ms │        80.83 / 81.06 ±0.19 / 81.30 ms │     no change │
│ QQuery 3  │        30.13 / 30.39 ±0.21 / 30.75 ms │        29.57 / 29.81 ±0.21 / 30.20 ms │     no change │
│ QQuery 4  │     492.77 / 497.80 ±4.61 / 505.89 ms │     488.11 / 492.87 ±3.56 / 499.04 ms │     no change │
│ QQuery 5  │        52.25 / 52.88 ±0.65 / 53.95 ms │        52.42 / 52.68 ±0.18 / 52.98 ms │     no change │
│ QQuery 6  │        36.97 / 37.19 ±0.25 / 37.65 ms │        36.94 / 37.24 ±0.18 / 37.45 ms │     no change │
│ QQuery 7  │       94.31 / 97.11 ±3.28 / 103.53 ms │        95.46 / 96.36 ±0.79 / 97.70 ms │     no change │
│ QQuery 8  │        36.95 / 38.10 ±0.81 / 39.29 ms │        37.75 / 38.06 ±0.33 / 38.64 ms │     no change │
│ QQuery 9  │        53.15 / 55.50 ±2.53 / 60.30 ms │        51.90 / 54.02 ±1.43 / 56.32 ms │     no change │
│ QQuery 10 │        63.92 / 64.65 ±0.63 / 65.60 ms │        64.17 / 64.38 ±0.19 / 64.68 ms │     no change │
│ QQuery 11 │     304.30 / 306.12 ±2.51 / 311.08 ms │     307.89 / 315.73 ±6.59 / 326.50 ms │     no change │
│ QQuery 12 │        29.01 / 29.52 ±0.39 / 29.98 ms │        29.09 / 29.74 ±0.34 / 30.05 ms │     no change │
│ QQuery 13 │     119.25 / 120.91 ±1.68 / 123.78 ms │     120.61 / 121.50 ±1.14 / 123.70 ms │     no change │
│ QQuery 14 │     416.16 / 418.49 ±1.73 / 421.01 ms │     422.37 / 426.31 ±3.29 / 430.84 ms │     no change │
│ QQuery 15 │        57.61 / 58.40 ±1.05 / 60.41 ms │        59.14 / 59.85 ±0.73 / 61.22 ms │     no change │
│ QQuery 16 │           6.61 / 6.75 ±0.20 / 7.13 ms │           6.70 / 6.84 ±0.24 / 7.31 ms │     no change │
│ QQuery 17 │        80.41 / 81.72 ±2.08 / 85.86 ms │        81.09 / 82.52 ±1.50 / 85.42 ms │     no change │
│ QQuery 18 │     122.68 / 124.59 ±1.35 / 126.83 ms │     124.86 / 127.00 ±1.18 / 128.33 ms │     no change │
│ QQuery 19 │        42.61 / 42.86 ±0.25 / 43.30 ms │        42.73 / 43.75 ±1.43 / 46.57 ms │     no change │
│ QQuery 20 │        36.18 / 37.17 ±1.26 / 39.65 ms │        36.58 / 37.00 ±0.41 / 37.51 ms │     no change │
│ QQuery 21 │        18.15 / 18.52 ±0.35 / 19.13 ms │        18.02 / 18.81 ±1.09 / 20.98 ms │     no change │
│ QQuery 22 │        62.46 / 63.63 ±0.63 / 64.23 ms │        64.13 / 64.68 ±0.44 / 65.14 ms │     no change │
│ QQuery 23 │     347.80 / 353.85 ±5.23 / 361.33 ms │     350.94 / 357.47 ±5.33 / 366.24 ms │     no change │
│ QQuery 24 │     226.83 / 230.31 ±5.03 / 240.23 ms │     225.48 / 230.95 ±4.50 / 238.71 ms │     no change │
│ QQuery 25 │     109.54 / 112.25 ±1.44 / 113.53 ms │     111.91 / 112.75 ±0.94 / 114.53 ms │     no change │
│ QQuery 26 │        57.76 / 61.22 ±3.65 / 68.27 ms │        58.18 / 59.20 ±1.29 / 61.75 ms │     no change │
│ QQuery 27 │           6.33 / 6.44 ±0.19 / 6.83 ms │           6.36 / 6.50 ±0.19 / 6.87 ms │     no change │
│ QQuery 28 │        61.69 / 62.50 ±0.52 / 63.17 ms │        60.87 / 61.73 ±0.47 / 62.22 ms │     no change │
│ QQuery 29 │      97.51 / 101.35 ±5.62 / 112.41 ms │      97.74 / 101.10 ±3.10 / 106.31 ms │     no change │
│ QQuery 30 │        32.77 / 34.17 ±1.46 / 36.81 ms │        32.68 / 33.06 ±0.37 / 33.76 ms │     no change │
│ QQuery 31 │     112.51 / 113.22 ±0.43 / 113.86 ms │     111.99 / 112.17 ±0.14 / 112.40 ms │     no change │
│ QQuery 32 │        20.32 / 20.73 ±0.38 / 21.44 ms │        20.68 / 22.10 ±2.14 / 26.32 ms │  1.07x slower │
│ QQuery 33 │        37.73 / 40.10 ±2.94 / 45.68 ms │        38.07 / 38.48 ±0.23 / 38.71 ms │     no change │
│ QQuery 34 │         9.82 / 10.46 ±0.91 / 12.23 ms │         9.97 / 10.26 ±0.30 / 10.65 ms │     no change │
│ QQuery 35 │        73.54 / 73.99 ±0.34 / 74.36 ms │        73.20 / 74.01 ±0.67 / 75.05 ms │     no change │
│ QQuery 36 │           5.82 / 5.92 ±0.14 / 6.18 ms │           5.77 / 5.92 ±0.18 / 6.27 ms │     no change │
│ QQuery 37 │           6.87 / 7.00 ±0.11 / 7.16 ms │           6.92 / 7.00 ±0.08 / 7.16 ms │     no change │
│ QQuery 38 │        62.38 / 62.53 ±0.10 / 62.66 ms │        62.81 / 63.59 ±0.90 / 65.31 ms │     no change │
│ QQuery 39 │        90.52 / 93.16 ±3.21 / 99.38 ms │        90.30 / 92.23 ±1.61 / 95.14 ms │     no change │
│ QQuery 40 │        23.79 / 24.37 ±0.40 / 25.00 ms │        23.40 / 23.66 ±0.20 / 23.97 ms │     no change │
│ QQuery 41 │        11.55 / 11.72 ±0.16 / 12.03 ms │        11.48 / 11.66 ±0.22 / 12.09 ms │     no change │
│ QQuery 42 │        24.31 / 24.71 ±0.40 / 25.47 ms │        24.04 / 24.35 ±0.23 / 24.69 ms │     no change │
│ QQuery 43 │           4.91 / 5.03 ±0.16 / 5.36 ms │           5.02 / 5.13 ±0.16 / 5.45 ms │     no change │
│ QQuery 44 │           9.14 / 9.26 ±0.12 / 9.48 ms │           9.44 / 9.50 ±0.08 / 9.66 ms │     no change │
│ QQuery 45 │        38.54 / 38.91 ±0.22 / 39.14 ms │        38.64 / 40.99 ±2.08 / 44.51 ms │  1.05x slower │
│ QQuery 46 │        11.76 / 13.05 ±1.72 / 16.40 ms │        12.14 / 12.96 ±0.87 / 14.57 ms │     no change │
│ QQuery 47 │     230.93 / 235.15 ±2.60 / 238.53 ms │     231.59 / 234.88 ±2.86 / 239.52 ms │     no change │
│ QQuery 48 │        96.47 / 97.63 ±0.76 / 98.74 ms │        95.53 / 96.27 ±0.44 / 96.79 ms │     no change │
│ QQuery 49 │        77.49 / 80.77 ±5.90 / 92.57 ms │        77.14 / 80.70 ±4.17 / 86.42 ms │     no change │
│ QQuery 50 │        59.82 / 61.80 ±1.96 / 65.06 ms │        59.67 / 60.28 ±0.46 / 60.76 ms │     no change │
│ QQuery 51 │        91.94 / 93.88 ±1.21 / 95.31 ms │        91.37 / 93.12 ±1.22 / 94.86 ms │     no change │
│ QQuery 52 │        24.64 / 24.85 ±0.20 / 25.20 ms │        24.47 / 26.65 ±3.06 / 32.58 ms │  1.07x slower │
│ QQuery 53 │        31.20 / 33.08 ±3.19 / 39.44 ms │        29.67 / 30.06 ±0.51 / 31.05 ms │ +1.10x faster │
│ QQuery 54 │        56.14 / 56.58 ±0.24 / 56.84 ms │        55.33 / 56.12 ±0.47 / 56.79 ms │     no change │
│ QQuery 55 │        23.81 / 24.11 ±0.29 / 24.66 ms │        23.73 / 23.94 ±0.27 / 24.46 ms │     no change │
│ QQuery 56 │        39.54 / 40.39 ±0.52 / 40.90 ms │        39.44 / 39.69 ±0.25 / 40.08 ms │     no change │
│ QQuery 57 │     178.71 / 179.97 ±1.91 / 183.74 ms │     177.48 / 181.83 ±3.90 / 188.20 ms │     no change │
│ QQuery 58 │     117.62 / 119.87 ±2.69 / 125.06 ms │     116.33 / 117.46 ±1.40 / 120.11 ms │     no change │
│ QQuery 59 │     118.56 / 119.07 ±0.33 / 119.55 ms │     118.61 / 118.96 ±0.19 / 119.15 ms │     no change │
│ QQuery 60 │        39.70 / 40.39 ±0.51 / 41.07 ms │        39.81 / 41.17 ±1.81 / 44.66 ms │     no change │
│ QQuery 61 │        12.11 / 12.22 ±0.15 / 12.51 ms │        12.18 / 12.35 ±0.20 / 12.72 ms │     no change │
│ QQuery 62 │        46.74 / 47.39 ±1.01 / 49.40 ms │        46.83 / 48.19 ±1.30 / 50.64 ms │     no change │
│ QQuery 63 │        29.86 / 30.58 ±0.43 / 30.97 ms │        29.82 / 30.33 ±0.45 / 30.89 ms │     no change │
│ QQuery 64 │     410.72 / 414.59 ±3.08 / 419.92 ms │     410.69 / 416.03 ±4.17 / 421.19 ms │     no change │
│ QQuery 65 │     126.93 / 131.60 ±3.03 / 135.07 ms │     128.80 / 131.08 ±2.07 / 134.90 ms │     no change │
│ QQuery 66 │        79.93 / 80.65 ±0.54 / 81.24 ms │        80.96 / 81.40 ±0.35 / 81.98 ms │     no change │
│ QQuery 67 │     243.26 / 246.74 ±2.22 / 249.56 ms │     240.33 / 244.86 ±2.83 / 248.30 ms │     no change │
│ QQuery 68 │        12.00 / 12.14 ±0.19 / 12.50 ms │        12.17 / 12.30 ±0.11 / 12.49 ms │     no change │
│ QQuery 69 │        58.31 / 59.19 ±0.72 / 60.45 ms │        57.65 / 61.03 ±4.30 / 69.29 ms │     no change │
│ QQuery 70 │     107.09 / 110.88 ±2.78 / 115.31 ms │     106.89 / 110.10 ±2.31 / 114.10 ms │     no change │
│ QQuery 71 │        36.30 / 36.54 ±0.19 / 36.79 ms │        35.83 / 36.19 ±0.37 / 36.89 ms │     no change │
│ QQuery 72 │ 2146.66 / 2208.23 ±52.08 / 2272.13 ms │ 2110.95 / 2172.05 ±39.14 / 2209.54 ms │     no change │
│ QQuery 73 │          9.56 / 9.75 ±0.23 / 10.16 ms │         9.90 / 10.32 ±0.32 / 10.81 ms │  1.06x slower │
│ QQuery 74 │     172.30 / 177.46 ±5.80 / 187.69 ms │     171.92 / 176.94 ±3.22 / 181.16 ms │     no change │
│ QQuery 75 │     150.15 / 153.69 ±4.84 / 162.66 ms │     149.89 / 151.89 ±2.37 / 156.47 ms │     no change │
│ QQuery 76 │        35.85 / 36.54 ±0.91 / 38.30 ms │        35.60 / 35.81 ±0.14 / 35.99 ms │     no change │
│ QQuery 77 │        62.09 / 64.04 ±2.91 / 69.80 ms │        62.77 / 64.61 ±2.78 / 70.11 ms │     no change │
│ QQuery 78 │     198.36 / 203.88 ±4.36 / 210.45 ms │     197.79 / 200.97 ±2.18 / 203.47 ms │     no change │
│ QQuery 79 │        68.09 / 69.11 ±0.88 / 70.63 ms │        68.30 / 69.17 ±0.81 / 70.63 ms │     no change │
│ QQuery 80 │     100.37 / 102.82 ±3.15 / 108.94 ms │     100.13 / 101.24 ±1.21 / 103.58 ms │     no change │
│ QQuery 81 │        26.47 / 27.91 ±1.71 / 31.17 ms │        26.23 / 27.71 ±2.49 / 32.67 ms │     no change │
│ QQuery 82 │        17.35 / 18.86 ±2.20 / 23.23 ms │        16.80 / 17.04 ±0.29 / 17.59 ms │ +1.11x faster │
│ QQuery 83 │        40.29 / 40.73 ±0.32 / 41.23 ms │        41.13 / 43.73 ±3.12 / 48.66 ms │  1.07x slower │
│ QQuery 84 │        30.47 / 30.93 ±0.32 / 31.40 ms │        30.62 / 31.01 ±0.37 / 31.70 ms │     no change │
│ QQuery 85 │     107.78 / 110.41 ±2.87 / 115.92 ms │     107.07 / 108.58 ±1.75 / 111.79 ms │     no change │
│ QQuery 86 │        25.91 / 26.83 ±1.09 / 28.93 ms │        26.11 / 26.70 ±0.68 / 27.83 ms │     no change │
│ QQuery 87 │        62.25 / 63.11 ±0.66 / 64.06 ms │        63.49 / 64.81 ±0.89 / 66.13 ms │     no change │
│ QQuery 88 │        64.38 / 64.58 ±0.21 / 64.92 ms │        63.88 / 64.38 ±0.85 / 66.08 ms │     no change │
│ QQuery 89 │        36.40 / 37.39 ±0.99 / 39.17 ms │        36.24 / 36.71 ±0.33 / 37.12 ms │     no change │
│ QQuery 90 │        17.20 / 17.48 ±0.22 / 17.86 ms │        17.22 / 18.18 ±1.82 / 21.82 ms │     no change │
│ QQuery 91 │        46.29 / 46.92 ±0.50 / 47.79 ms │        45.99 / 46.83 ±0.77 / 48.25 ms │     no change │
│ QQuery 92 │        29.64 / 30.28 ±0.68 / 31.57 ms │        30.11 / 30.70 ±0.40 / 31.27 ms │     no change │
│ QQuery 93 │        51.25 / 51.63 ±0.24 / 51.92 ms │        50.45 / 51.56 ±0.80 / 52.51 ms │     no change │
│ QQuery 94 │        38.41 / 39.91 ±2.06 / 43.99 ms │        38.54 / 39.30 ±0.51 / 40.02 ms │     no change │
│ QQuery 95 │        81.93 / 83.51 ±1.41 / 86.14 ms │        81.32 / 82.79 ±1.30 / 84.81 ms │     no change │
│ QQuery 96 │        24.39 / 24.62 ±0.32 / 25.24 ms │        24.31 / 24.62 ±0.23 / 25.00 ms │     no change │
│ QQuery 97 │        47.04 / 47.52 ±0.30 / 47.84 ms │        46.91 / 47.26 ±0.28 / 47.73 ms │     no change │
│ QQuery 98 │        43.42 / 44.87 ±2.26 / 49.36 ms │        43.37 / 43.99 ±0.65 / 45.24 ms │     no change │
│ QQuery 99 │        70.85 / 71.48 ±0.68 / 72.76 ms │        71.31 / 71.56 ±0.21 / 71.83 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                               ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                               │ 10080.63ms │
│ Total Time (alamb_test_sparse_mask2222323412)   │ 10048.44ms │
│ Average Time (HEAD)                             │   101.82ms │
│ Average Time (alamb_test_sparse_mask2222323412) │   101.50ms │
│ Queries Faster                                  │          2 │
│ Queries Slower                                  │          5 │
│ Queries with No Change                          │         92 │
│ Queries with Failure                            │          0 │
└─────────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	2.2 GiB
Avg memory	1.5 GiB
CPU user	224.3s
CPU sys	6.2s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	55.0s
Peak memory	2.1 GiB
Avg memory	1.5 GiB
CPU user	225.0s
CPU sys	6.1s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-20T19:13:01Z

🤖 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_test_sparse_mask2222323412
--------------------
Benchmark clickbench_pushdown.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃      alamb_test_sparse_mask2222323412 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.27 / 4.03 ±5.43 / 14.88 ms │          1.27 / 4.06 ±5.48 / 15.01 ms │     no change │
│ QQuery 1  │        13.37 / 13.47 ±0.08 / 13.59 ms │        13.23 / 13.34 ±0.08 / 13.42 ms │     no change │
│ QQuery 2  │        36.79 / 37.23 ±0.34 / 37.76 ms │        36.34 / 36.50 ±0.16 / 36.79 ms │     no change │
│ QQuery 3  │        31.47 / 32.21 ±0.87 / 33.67 ms │        31.00 / 31.18 ±0.20 / 31.56 ms │     no change │
│ QQuery 4  │     223.56 / 227.08 ±3.10 / 232.33 ms │     224.20 / 228.23 ±3.62 / 234.60 ms │     no change │
│ QQuery 5  │     271.54 / 273.49 ±1.98 / 276.57 ms │     269.81 / 272.94 ±2.49 / 277.02 ms │     no change │
│ QQuery 6  │           1.30 / 1.46 ±0.23 / 1.92 ms │           1.30 / 1.46 ±0.24 / 1.93 ms │     no change │
│ QQuery 7  │        16.56 / 16.81 ±0.15 / 16.93 ms │        16.66 / 16.86 ±0.11 / 16.95 ms │     no change │
│ QQuery 8  │     322.14 / 324.10 ±1.52 / 325.94 ms │     325.19 / 327.82 ±1.46 / 329.20 ms │     no change │
│ QQuery 9  │     448.45 / 456.98 ±6.22 / 464.84 ms │     459.28 / 466.62 ±6.67 / 476.72 ms │     no change │
│ QQuery 10 │        93.25 / 94.48 ±0.79 / 95.19 ms │        94.09 / 94.93 ±0.62 / 95.78 ms │     no change │
│ QQuery 11 │     105.65 / 106.24 ±0.45 / 106.90 ms │     105.07 / 106.66 ±1.10 / 107.79 ms │     no change │
│ QQuery 12 │     298.28 / 303.35 ±4.00 / 308.14 ms │     300.10 / 309.49 ±7.23 / 321.54 ms │     no change │
│ QQuery 13 │    407.78 / 433.74 ±16.90 / 452.73 ms │     417.44 / 423.15 ±5.29 / 429.66 ms │     no change │
│ QQuery 14 │     314.69 / 321.14 ±4.62 / 327.34 ms │    311.06 / 327.33 ±12.00 / 346.81 ms │     no change │
│ QQuery 15 │    271.60 / 283.38 ±12.06 / 305.54 ms │     271.77 / 283.06 ±8.70 / 292.53 ms │     no change │
│ QQuery 16 │     611.28 / 615.40 ±4.64 / 624.34 ms │     619.86 / 629.77 ±8.50 / 641.24 ms │     no change │
│ QQuery 17 │     620.49 / 630.90 ±9.94 / 646.80 ms │     625.39 / 632.92 ±8.83 / 650.16 ms │     no change │
│ QQuery 18 │ 1251.57 / 1267.22 ±10.66 / 1280.33 ms │ 1262.77 / 1280.52 ±20.36 / 1319.21 ms │     no change │
│ QQuery 19 │        28.93 / 32.63 ±5.90 / 44.36 ms │        30.12 / 30.88 ±0.71 / 32.09 ms │ +1.06x faster │
│ QQuery 20 │     513.29 / 520.50 ±8.65 / 536.99 ms │     517.40 / 525.64 ±5.98 / 532.08 ms │     no change │
│ QQuery 21 │     561.45 / 571.93 ±6.52 / 581.95 ms │     567.07 / 579.63 ±8.46 / 592.23 ms │     no change │
│ QQuery 22 │     927.60 / 932.49 ±3.75 / 936.93 ms │    925.19 / 939.17 ±14.32 / 965.18 ms │     no change │
│ QQuery 23 │     123.12 / 128.33 ±3.52 / 132.15 ms │     116.59 / 123.65 ±7.47 / 137.20 ms │     no change │
│ QQuery 24 │        38.64 / 41.18 ±1.82 / 42.92 ms │        37.35 / 38.15 ±0.87 / 39.83 ms │ +1.08x faster │
│ QQuery 25 │     144.53 / 149.28 ±5.15 / 157.20 ms │    144.45 / 150.84 ±10.35 / 171.46 ms │     no change │
│ QQuery 26 │        50.11 / 51.11 ±0.98 / 52.76 ms │        49.66 / 52.94 ±4.47 / 61.45 ms │     no change │
│ QQuery 27 │     566.65 / 575.28 ±5.34 / 582.75 ms │     562.91 / 575.64 ±9.34 / 590.59 ms │     no change │
│ QQuery 28 │ 2901.78 / 2920.47 ±11.30 / 2931.12 ms │ 2940.94 / 2979.74 ±34.72 / 3034.36 ms │     no change │
│ QQuery 29 │        41.55 / 41.81 ±0.30 / 42.38 ms │       41.42 / 54.35 ±16.97 / 85.80 ms │  1.30x slower │
│ QQuery 30 │    308.00 / 321.30 ±17.37 / 354.88 ms │     312.63 / 316.10 ±2.82 / 320.53 ms │     no change │
│ QQuery 31 │     288.76 / 295.96 ±4.39 / 300.64 ms │    287.82 / 309.78 ±14.65 / 329.01 ms │     no change │
│ QQuery 32 │    926.57 / 955.22 ±29.17 / 994.36 ms │    943.32 / 961.72 ±12.19 / 980.87 ms │     no change │
│ QQuery 33 │ 1460.25 / 1485.76 ±16.55 / 1506.02 ms │  1469.12 / 1477.07 ±9.32 / 1490.09 ms │     no change │
│ QQuery 34 │ 1459.46 / 1524.70 ±53.59 / 1606.45 ms │ 1466.56 / 1515.23 ±33.87 / 1552.66 ms │     no change │
│ QQuery 35 │    283.53 / 300.55 ±16.30 / 327.01 ms │    280.44 / 303.54 ±33.33 / 369.37 ms │     no change │
│ QQuery 36 │        69.33 / 73.41 ±3.51 / 77.51 ms │       65.98 / 75.17 ±10.35 / 95.14 ms │     no change │
│ QQuery 37 │        37.43 / 42.49 ±5.95 / 50.41 ms │        36.78 / 41.47 ±4.40 / 48.89 ms │     no change │
│ QQuery 38 │        35.41 / 40.47 ±4.27 / 47.76 ms │        35.37 / 38.90 ±3.02 / 44.40 ms │     no change │
│ QQuery 39 │     142.30 / 153.63 ±6.88 / 160.83 ms │    129.11 / 145.35 ±10.65 / 161.29 ms │ +1.06x faster │
│ QQuery 40 │        18.60 / 19.17 ±0.42 / 19.86 ms │        18.71 / 20.77 ±3.47 / 27.69 ms │  1.08x slower │
│ QQuery 41 │        17.20 / 20.20 ±5.09 / 30.35 ms │        16.95 / 18.49 ±2.72 / 23.93 ms │ +1.09x faster │
│ QQuery 42 │        14.72 / 16.48 ±3.41 / 23.29 ms │        14.47 / 19.87 ±6.43 / 27.97 ms │  1.21x slower │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                               ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                               │ 16657.07ms │
│ Total Time (alamb_test_sparse_mask2222323412)   │ 16780.96ms │
│ Average Time (HEAD)                             │   387.37ms │
│ Average Time (alamb_test_sparse_mask2222323412) │   390.25ms │
│ Queries Faster                                  │          4 │
│ Queries Slower                                  │          3 │
│ Queries with No Change                          │         36 │
│ Queries with Failure                            │          0 │
└─────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_pushdown — base (merge-base)

Metric	Value
Wall time	85.0s
Peak memory	11.0 GiB
Avg memory	4.9 GiB
CPU user	850.0s
CPU sys	58.0s
Peak spill	0 B

clickbench_pushdown — branch

Metric	Value
Wall time	85.0s
Peak memory	10.9 GiB
Avg memory	4.5 GiB
CPU user	851.2s
CPU sys	62.8s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-20T19:15:07Z

🤖 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_test_sparse_mask2222323412
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃       alamb_test_sparse_mask2222323412 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.25 / 3.98 ±5.38 / 14.74 ms │           1.29 / 4.23 ±5.70 / 15.63 ms │  1.06x slower │
│ QQuery 1  │        12.74 / 13.06 ±0.31 / 13.66 ms │         13.02 / 13.38 ±0.20 / 13.59 ms │     no change │
│ QQuery 2  │        37.16 / 37.49 ±0.19 / 37.70 ms │         36.48 / 37.66 ±0.71 / 38.67 ms │     no change │
│ QQuery 3  │        31.39 / 32.54 ±1.34 / 34.66 ms │         32.27 / 32.59 ±0.19 / 32.77 ms │     no change │
│ QQuery 4  │     221.03 / 224.82 ±2.77 / 228.38 ms │      227.31 / 236.69 ±9.03 / 249.17 ms │  1.05x slower │
│ QQuery 5  │     268.37 / 270.40 ±1.24 / 271.69 ms │      272.12 / 274.38 ±2.51 / 277.51 ms │     no change │
│ QQuery 6  │           1.27 / 1.42 ±0.23 / 1.88 ms │            1.28 / 1.44 ±0.25 / 1.94 ms │     no change │
│ QQuery 7  │        13.78 / 13.92 ±0.11 / 14.04 ms │         13.83 / 13.96 ±0.08 / 14.04 ms │     no change │
│ QQuery 8  │     320.41 / 325.22 ±2.74 / 328.50 ms │     325.64 / 361.49 ±27.84 / 393.19 ms │  1.11x slower │
│ QQuery 9  │     452.57 / 463.22 ±8.68 / 477.83 ms │     464.53 / 485.57 ±17.72 / 506.20 ms │     no change │
│ QQuery 10 │        68.99 / 71.30 ±2.44 / 75.67 ms │         69.71 / 71.34 ±1.25 / 73.40 ms │     no change │
│ QQuery 11 │        80.61 / 81.55 ±0.85 / 82.90 ms │         81.25 / 81.97 ±0.70 / 83.21 ms │     no change │
│ QQuery 12 │     266.53 / 270.99 ±5.07 / 279.93 ms │      266.96 / 271.05 ±3.64 / 276.16 ms │     no change │
│ QQuery 13 │     360.32 / 371.54 ±6.43 / 377.64 ms │     365.75 / 378.38 ±13.06 / 401.48 ms │     no change │
│ QQuery 14 │     276.60 / 281.04 ±3.51 / 286.41 ms │      279.81 / 288.21 ±6.85 / 300.00 ms │     no change │
│ QQuery 15 │     267.90 / 271.56 ±4.00 / 278.37 ms │      272.11 / 277.52 ±3.34 / 281.78 ms │     no change │
│ QQuery 16 │    609.21 / 642.39 ±35.90 / 699.11 ms │     611.64 / 651.38 ±36.68 / 705.60 ms │     no change │
│ QQuery 17 │     682.27 / 696.53 ±7.42 / 702.65 ms │     615.30 / 656.83 ±33.04 / 714.11 ms │ +1.06x faster │
│ QQuery 18 │ 1297.52 / 1375.43 ±59.42 / 1446.15 ms │  1257.73 / 1287.07 ±21.38 / 1319.11 ms │ +1.07x faster │
│ QQuery 19 │        28.17 / 28.76 ±0.64 / 29.95 ms │        28.33 / 34.01 ±11.03 / 56.06 ms │  1.18x slower │
│ QQuery 20 │     521.45 / 537.14 ±9.84 / 549.82 ms │      515.42 / 519.64 ±3.59 / 524.51 ms │     no change │
│ QQuery 21 │     521.64 / 524.51 ±3.11 / 530.24 ms │      514.08 / 521.06 ±4.23 / 525.04 ms │     no change │
│ QQuery 22 │    986.01 / 994.24 ±5.32 / 1002.40 ms │   992.36 / 1030.35 ±30.18 / 1066.96 ms │     no change │
│ QQuery 23 │ 3079.59 / 3119.62 ±24.78 / 3145.56 ms │ 3138.48 / 3264.56 ±102.80 / 3371.80 ms │     no change │
│ QQuery 24 │        41.43 / 46.01 ±5.34 / 52.65 ms │         41.39 / 44.17 ±4.93 / 54.00 ms │     no change │
│ QQuery 25 │     110.28 / 113.27 ±3.62 / 120.26 ms │      110.96 / 111.85 ±0.67 / 112.85 ms │     no change │
│ QQuery 26 │        41.55 / 45.09 ±5.79 / 56.52 ms │         41.91 / 44.82 ±3.09 / 49.17 ms │     no change │
│ QQuery 27 │    518.38 / 533.96 ±12.11 / 549.07 ms │      514.01 / 523.81 ±5.83 / 531.32 ms │     no change │
│ QQuery 28 │ 2876.61 / 2903.73 ±28.53 / 2947.06 ms │  2883.60 / 2940.32 ±32.48 / 2977.76 ms │     no change │
│ QQuery 29 │        40.99 / 41.39 ±0.49 / 42.33 ms │         40.93 / 41.29 ±0.27 / 41.69 ms │     no change │
│ QQuery 30 │    299.64 / 307.90 ±10.74 / 328.70 ms │     300.33 / 315.97 ±21.07 / 357.49 ms │     no change │
│ QQuery 31 │     281.51 / 294.30 ±8.91 / 305.66 ms │     283.73 / 305.10 ±24.37 / 349.40 ms │     no change │
│ QQuery 32 │  957.27 / 1066.53 ±64.73 / 1146.97 ms │    919.48 / 971.23 ±58.58 / 1066.10 ms │ +1.10x faster │
│ QQuery 33 │ 1505.06 / 1631.52 ±88.61 / 1739.94 ms │  1444.02 / 1469.20 ±27.58 / 1518.29 ms │ +1.11x faster │
│ QQuery 34 │ 1482.64 / 1546.41 ±60.14 / 1658.84 ms │  1490.34 / 1553.31 ±43.15 / 1605.77 ms │     no change │
│ QQuery 35 │    281.03 / 299.55 ±20.62 / 333.07 ms │     350.59 / 372.80 ±22.31 / 414.27 ms │  1.24x slower │
│ QQuery 36 │        69.53 / 74.82 ±3.84 / 81.52 ms │       76.17 / 86.40 ±11.43 / 108.70 ms │  1.15x slower │
│ QQuery 37 │        35.78 / 39.50 ±5.80 / 51.03 ms │         38.16 / 40.97 ±3.15 / 46.73 ms │     no change │
│ QQuery 38 │        40.32 / 43.61 ±3.96 / 50.91 ms │         42.52 / 46.89 ±7.97 / 62.82 ms │  1.08x slower │
│ QQuery 39 │     150.09 / 159.53 ±5.87 / 166.46 ms │     155.58 / 169.98 ±10.71 / 186.45 ms │  1.07x slower │
│ QQuery 40 │        14.43 / 15.06 ±0.51 / 15.65 ms │         15.75 / 16.18 ±0.34 / 16.58 ms │  1.07x slower │
│ QQuery 41 │        13.90 / 15.78 ±3.40 / 22.58 ms │         15.33 / 17.32 ±3.84 / 25.00 ms │  1.10x slower │
│ QQuery 42 │        13.33 / 13.49 ±0.14 / 13.70 ms │         14.84 / 19.07 ±7.61 / 34.25 ms │  1.41x slower │
└───────────┴───────────────────────────────────────┴────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                               ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                               │ 19844.12ms │
│ Total Time (alamb_test_sparse_mask2222323412)   │ 19885.43ms │
│ Average Time (HEAD)                             │   461.49ms │
│ Average Time (alamb_test_sparse_mask2222323412) │   462.45ms │
│ Queries Faster                                  │          4 │
│ Queries Slower                                  │         11 │
│ Queries with No Change                          │         28 │
│ Queries with Failure                            │          0 │
└─────────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	100.0s
Peak memory	11.8 GiB
Avg memory	4.7 GiB
CPU user	1021.4s
CPU sys	71.5s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	100.0s
Peak memory	10.7 GiB
Avg memory	4.6 GiB
CPU user	1016.6s
CPU sys	73.0s
Peak spill	0 B

File an issue against this benchmark runner

alamb · 2026-07-20T19:57:04Z

I don't see any substantial difference

chore: test with sparse page mask

fe1cedc

alamb marked this pull request as draft July 20, 2026 18:49

alamb mentioned this pull request Jul 20, 2026

fix(parquet): support mask filtering across skipped pages apache/arrow-rs#10288

Merged

alamb closed this Jul 20, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

TEST: check performance with sparse masks#23728

TEST: check performance with sparse masks#23728
alamb wants to merge 1 commit into
apache:mainfrom
alamb:alamb/test_sparse_mask2222323412

alamb commented Jul 20, 2026 •

edited

Loading

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

alamb commented Jul 20, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

adriangbot commented Jul 20, 2026

Uh oh!

alamb commented Jul 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

alamb commented Jul 20, 2026 •

edited

Loading