chore: refactor `MaterializingSortMergeJoinStream` into generators and simplify code to be textbook like as possible by rluvaton · Pull Request #23976 · apache/datafusion

rluvaton · 2026-07-29T15:05:05Z

Which issue does this PR close?

Part of:

[Status Update] Simplifing Streams to be more textbook-like and have less state while keeping the same perf #23974

Rationale for this change

Simplify the code and make code similar to textbook so it is easier to read

What changes are included in this PR?

Changed to async generators + simplified to be similar to textbook

Most of the code was written by Claude Fable 5 since it was just too large to comprehend in head

Are these changes tested?

existing tests + a little more tests to verify we are not counting the childs in the join time

Are there any user-facing changes?

The join_time now includes the time to read from the async spill stream between pending which is arguable more correct since this time is part of the operator, although long waits between pending calls will be counted in the op join_time while the alternative is not counting the read from file and decoding...

rluvaton · 2026-07-29T15:05:22Z

run benchmark smj

rluvaton · 2026-07-29T15:05:27Z

run benchmarks

adriangbot · 2026-07-29T15:07:30Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5119660671-1290-2d6h6 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 refactor-materializing-smj (ccdd363) to 68d5874 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:08:10Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5119660671-1291-bzfnj 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 refactor-materializing-smj (ccdd363) to 68d5874 (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:08:21Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5119660671-1292-rgbxd 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 refactor-materializing-smj (ccdd363) to 68d5874 (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:08:32Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5119659315-1289-ds2lx 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 refactor-materializing-smj (ccdd363) to 68d5874 (merge-base) diff using: smj
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:23:06Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃     refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.53 / 39.94 ±1.24 / 41.73 ms │ 38.67 / 39.75 ±1.80 / 43.34 ms │     no change │
│ QQuery 2  │ 19.38 / 19.57 ±0.19 / 19.91 ms │ 19.15 / 19.52 ±0.36 / 20.21 ms │     no change │
│ QQuery 3  │ 30.79 / 32.80 ±1.85 / 35.75 ms │ 31.42 / 33.54 ±1.60 / 36.38 ms │     no change │
│ QQuery 4  │ 17.35 / 17.53 ±0.12 / 17.68 ms │ 17.43 / 17.69 ±0.19 / 17.93 ms │     no change │
│ QQuery 5  │ 39.81 / 40.52 ±0.44 / 41.09 ms │ 37.69 / 39.88 ±2.27 / 43.81 ms │     no change │
│ QQuery 6  │ 16.42 / 16.97 ±0.90 / 18.77 ms │ 16.30 / 16.70 ±0.48 / 17.64 ms │     no change │
│ QQuery 7  │ 43.03 / 44.00 ±1.24 / 46.44 ms │ 43.19 / 45.03 ±1.41 / 47.18 ms │     no change │
│ QQuery 8  │ 42.62 / 42.78 ±0.21 / 43.18 ms │ 42.44 / 42.74 ±0.27 / 43.23 ms │     no change │
│ QQuery 9  │ 49.16 / 49.87 ±0.57 / 50.80 ms │ 48.70 / 51.32 ±1.82 / 53.55 ms │     no change │
│ QQuery 10 │ 41.78 / 42.20 ±0.33 / 42.76 ms │ 42.15 / 42.45 ±0.31 / 43.02 ms │     no change │
│ QQuery 11 │ 13.31 / 13.59 ±0.23 / 13.98 ms │ 13.40 / 14.00 ±0.68 / 14.93 ms │     no change │
│ QQuery 12 │ 23.69 / 24.54 ±1.03 / 26.48 ms │ 23.61 / 24.11 ±0.31 / 24.58 ms │     no change │
│ QQuery 13 │ 32.06 / 33.96 ±1.30 / 35.85 ms │ 33.04 / 34.17 ±0.80 / 35.08 ms │     no change │
│ QQuery 14 │ 23.45 / 23.72 ±0.20 / 23.95 ms │ 23.62 / 23.73 ±0.09 / 23.85 ms │     no change │
│ QQuery 15 │ 31.24 / 32.20 ±1.00 / 33.47 ms │ 31.05 / 31.36 ±0.21 / 31.68 ms │     no change │
│ QQuery 16 │ 13.94 / 14.01 ±0.06 / 14.08 ms │ 13.68 / 13.88 ±0.22 / 14.21 ms │     no change │
│ QQuery 17 │ 72.72 / 79.06 ±3.75 / 84.14 ms │ 70.47 / 70.79 ±0.29 / 71.30 ms │ +1.12x faster │
│ QQuery 18 │ 64.20 / 65.71 ±1.60 / 68.38 ms │ 57.86 / 60.07 ±1.33 / 61.57 ms │ +1.09x faster │
│ QQuery 19 │ 34.74 / 35.39 ±0.56 / 36.27 ms │ 32.99 / 33.43 ±0.40 / 34.07 ms │ +1.06x faster │
│ QQuery 20 │ 34.50 / 34.72 ±0.23 / 35.15 ms │ 31.81 / 32.05 ±0.14 / 32.20 ms │ +1.08x faster │
│ QQuery 21 │ 61.37 / 62.68 ±1.85 / 66.36 ms │ 55.02 / 57.26 ±1.71 / 59.59 ms │ +1.09x faster │
│ QQuery 22 │ 15.32 / 16.02 ±0.70 / 17.05 ms │ 13.90 / 14.13 ±0.16 / 14.38 ms │ +1.13x faster │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                         ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 781.79ms │
│ Total Time (refactor-materializing-smj)   │ 757.60ms │
│ Average Time (HEAD)                       │  35.54ms │
│ Average Time (refactor-materializing-smj) │  34.44ms │
│ Queries Faster                            │        6 │
│ Queries Slower                            │        0 │
│ Queries with No Change                    │       16 │
│ Queries with Failure                      │        0 │
└───────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.2 GiB
Avg memory	519.7 MiB
CPU user	22.8s
CPU sys	1.8s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	1.1 GiB
Avg memory	488.1 MiB
CPU user	21.7s
CPU sys	1.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:24:05Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark smj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                 HEAD ┃           refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │          8.52 / 8.72 ±0.14 / 8.92 ms │          8.00 / 8.21 ±0.19 / 8.45 ms │ +1.06x faster │
│ QQuery 2  │    163.16 / 164.09 ±0.80 / 165.22 ms │    158.54 / 159.63 ±1.12 / 161.78 ms │     no change │
│ QQuery 3  │    105.44 / 106.72 ±0.76 / 107.66 ms │    102.00 / 103.87 ±1.26 / 105.97 ms │     no change │
│ QQuery 4  │       28.36 / 28.52 ±0.15 / 28.75 ms │       28.20 / 28.33 ±0.12 / 28.51 ms │     no change │
│ QQuery 5  │       20.79 / 20.97 ±0.18 / 21.29 ms │       20.60 / 20.68 ±0.09 / 20.86 ms │     no change │
│ QQuery 6  │    160.06 / 163.18 ±3.53 / 170.04 ms │    155.67 / 157.04 ±1.40 / 159.42 ms │     no change │
│ QQuery 7  │    198.64 / 201.97 ±1.74 / 203.75 ms │    193.45 / 197.79 ±5.44 / 208.31 ms │     no change │
│ QQuery 8  │       19.05 / 19.25 ±0.17 / 19.52 ms │       18.75 / 19.78 ±1.05 / 21.46 ms │     no change │
│ QQuery 9  │   207.12 / 214.96 ±10.12 / 234.42 ms │    202.09 / 203.86 ±1.15 / 205.55 ms │ +1.05x faster │
│ QQuery 10 │       58.05 / 58.33 ±0.19 / 58.64 ms │       57.95 / 58.40 ±0.23 / 58.56 ms │     no change │
│ QQuery 11 │       27.10 / 30.49 ±6.31 / 43.10 ms │       27.36 / 27.51 ±0.15 / 27.78 ms │ +1.11x faster │
│ QQuery 12 │       70.76 / 72.19 ±1.62 / 75.27 ms │       67.13 / 71.77 ±2.56 / 74.92 ms │     no change │
│ QQuery 13 │     96.69 / 103.83 ±4.88 / 108.23 ms │     95.39 / 103.18 ±6.59 / 111.93 ms │     no change │
│ QQuery 14 │       58.48 / 61.55 ±4.42 / 70.11 ms │       57.75 / 59.64 ±2.40 / 64.35 ms │     no change │
│ QQuery 15 │       58.00 / 58.14 ±0.14 / 58.39 ms │       57.94 / 58.20 ±0.24 / 58.62 ms │     no change │
│ QQuery 16 │       11.55 / 11.83 ±0.19 / 12.14 ms │       11.55 / 11.72 ±0.12 / 11.84 ms │     no change │
│ QQuery 17 │    145.96 / 151.28 ±8.47 / 168.15 ms │    145.70 / 146.69 ±1.67 / 150.01 ms │     no change │
│ QQuery 18 │    111.04 / 111.46 ±0.35 / 111.98 ms │    111.97 / 112.33 ±0.25 / 112.73 ms │     no change │
│ QQuery 19 │    290.20 / 296.30 ±6.29 / 308.04 ms │    290.65 / 299.33 ±7.38 / 308.95 ms │     no change │
│ QQuery 20 │ 1262.70 / 1267.90 ±5.40 / 1274.78 ms │ 1236.53 / 1244.40 ±5.39 / 1251.81 ms │     no change │
│ QQuery 21 │       94.46 / 95.84 ±0.94 / 97.08 ms │       88.86 / 89.73 ±0.65 / 90.64 ms │ +1.07x faster │
│ QQuery 22 │    102.67 / 103.61 ±0.78 / 105.00 ms │      96.32 / 99.06 ±3.98 / 106.82 ms │     no change │
│ QQuery 23 │    110.90 / 112.48 ±1.39 / 114.29 ms │    104.36 / 105.48 ±0.72 / 106.57 ms │ +1.07x faster │
│ QQuery 24 │       27.46 / 27.54 ±0.08 / 27.66 ms │       27.41 / 27.84 ±0.42 / 28.63 ms │     no change │
│ QQuery 25 │       65.74 / 71.26 ±3.85 / 77.16 ms │       69.69 / 73.68 ±3.53 / 80.15 ms │     no change │
│ QQuery 26 │     98.03 / 107.04 ±7.11 / 118.25 ms │     98.33 / 109.05 ±5.52 / 113.61 ms │     no change │
└───────────┴──────────────────────────────────────┴──────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 3669.47ms │
│ Total Time (refactor-materializing-smj)   │ 3597.21ms │
│ Average Time (HEAD)                       │  141.13ms │
│ Average Time (refactor-materializing-smj) │  138.35ms │
│ Queries Faster                            │         5 │
│ Queries Slower                            │         0 │
│ Queries with No Change                    │        21 │
│ Queries with Failure                      │         0 │
└───────────────────────────────────────────┴───────────┘

Resource Usage

smj — base (merge-base)

Metric	Value
Wall time	20.0s
Peak memory	1.1 GiB
Avg memory	445.0 MiB
CPU user	156.9s
CPU sys	3.5s
Peak spill	0 B

smj — branch

Metric	Value
Wall time	20.0s
Peak memory	1.1 GiB
Avg memory	441.0 MiB
CPU user	157.2s
CPU sys	3.5s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:24:56Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃            refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.56 / 6.07 ±0.89 / 7.85 ms │           5.67 / 6.18 ±0.92 / 8.01 ms │     no change │
│ QQuery 2  │        82.05 / 82.18 ±0.13 / 82.40 ms │        82.55 / 83.15 ±0.45 / 83.83 ms │     no change │
│ QQuery 3  │        30.29 / 30.92 ±1.14 / 33.19 ms │        30.34 / 30.62 ±0.21 / 30.92 ms │     no change │
│ QQuery 4  │     506.64 / 512.45 ±3.50 / 517.53 ms │     511.17 / 515.43 ±3.90 / 522.68 ms │     no change │
│ QQuery 5  │        52.84 / 53.53 ±0.79 / 54.97 ms │        54.39 / 54.93 ±0.34 / 55.32 ms │     no change │
│ QQuery 6  │        36.77 / 37.14 ±0.30 / 37.63 ms │        38.22 / 38.65 ±0.49 / 39.46 ms │     no change │
│ QQuery 7  │       95.44 / 97.83 ±3.57 / 104.90 ms │       96.70 / 98.57 ±2.42 / 103.25 ms │     no change │
│ QQuery 8  │        37.37 / 37.80 ±0.30 / 38.19 ms │        37.96 / 38.52 ±0.39 / 38.97 ms │     no change │
│ QQuery 9  │        53.98 / 56.29 ±1.21 / 57.23 ms │        55.49 / 56.53 ±0.64 / 57.30 ms │     no change │
│ QQuery 10 │        63.74 / 63.89 ±0.13 / 64.07 ms │        64.84 / 64.99 ±0.15 / 65.25 ms │     no change │
│ QQuery 11 │    315.17 / 326.75 ±11.14 / 340.57 ms │     320.07 / 324.01 ±3.92 / 330.58 ms │     no change │
│ QQuery 12 │        29.41 / 29.98 ±0.40 / 30.65 ms │        29.33 / 29.88 ±0.35 / 30.37 ms │     no change │
│ QQuery 13 │     120.83 / 122.10 ±1.52 / 125.01 ms │     120.62 / 122.07 ±1.60 / 125.11 ms │     no change │
│ QQuery 14 │     415.42 / 421.13 ±6.62 / 433.50 ms │     422.50 / 426.05 ±5.25 / 436.48 ms │     no change │
│ QQuery 15 │        59.34 / 60.01 ±0.42 / 60.48 ms │        58.76 / 60.36 ±1.44 / 62.93 ms │     no change │
│ QQuery 16 │           6.66 / 6.86 ±0.21 / 7.24 ms │           6.69 / 6.92 ±0.20 / 7.23 ms │     no change │
│ QQuery 17 │        81.08 / 83.04 ±2.66 / 88.18 ms │        82.68 / 84.16 ±1.78 / 87.42 ms │     no change │
│ QQuery 18 │     123.47 / 124.44 ±0.85 / 125.98 ms │     124.00 / 125.51 ±1.23 / 127.75 ms │     no change │
│ QQuery 19 │        42.37 / 42.77 ±0.38 / 43.31 ms │        42.99 / 44.01 ±1.60 / 47.19 ms │     no change │
│ QQuery 20 │        38.13 / 38.74 ±0.71 / 40.07 ms │        37.15 / 38.15 ±0.60 / 38.86 ms │     no change │
│ QQuery 21 │        18.09 / 18.33 ±0.23 / 18.70 ms │        18.37 / 18.50 ±0.14 / 18.77 ms │     no change │
│ QQuery 22 │        64.84 / 65.70 ±0.49 / 66.23 ms │        63.72 / 64.85 ±0.83 / 66.28 ms │     no change │
│ QQuery 23 │     354.84 / 361.42 ±5.56 / 369.49 ms │     357.80 / 361.45 ±2.41 / 365.11 ms │     no change │
│ QQuery 24 │     227.22 / 229.30 ±2.24 / 232.81 ms │     229.87 / 231.30 ±0.79 / 232.17 ms │     no change │
│ QQuery 25 │     111.28 / 113.67 ±1.97 / 117.11 ms │     112.76 / 114.88 ±1.58 / 116.92 ms │     no change │
│ QQuery 26 │        58.50 / 59.02 ±0.29 / 59.38 ms │        58.67 / 58.97 ±0.23 / 59.25 ms │     no change │
│ QQuery 27 │           6.52 / 6.78 ±0.18 / 7.06 ms │           6.28 / 6.39 ±0.12 / 6.62 ms │ +1.06x faster │
│ QQuery 28 │        57.29 / 61.69 ±2.24 / 63.58 ms │        62.15 / 62.64 ±0.35 / 63.04 ms │     no change │
│ QQuery 29 │      98.68 / 101.85 ±2.84 / 105.53 ms │      99.38 / 102.55 ±2.58 / 107.00 ms │     no change │
│ QQuery 30 │        32.92 / 33.21 ±0.39 / 33.98 ms │        33.07 / 33.23 ±0.12 / 33.42 ms │     no change │
│ QQuery 31 │     113.69 / 114.27 ±0.45 / 114.80 ms │     114.56 / 116.95 ±2.70 / 122.21 ms │     no change │
│ QQuery 32 │        21.08 / 21.68 ±0.51 / 22.35 ms │        21.57 / 21.89 ±0.20 / 22.18 ms │     no change │
│ QQuery 33 │        38.56 / 39.14 ±0.63 / 40.29 ms │        39.40 / 39.78 ±0.23 / 40.06 ms │     no change │
│ QQuery 34 │        10.11 / 10.58 ±0.39 / 11.22 ms │        10.29 / 10.60 ±0.36 / 11.18 ms │     no change │
│ QQuery 35 │        74.16 / 75.08 ±0.82 / 76.30 ms │        73.93 / 75.12 ±0.82 / 76.45 ms │     no change │
│ QQuery 36 │           5.86 / 6.06 ±0.22 / 6.47 ms │           6.10 / 6.27 ±0.19 / 6.62 ms │     no change │
│ QQuery 37 │           6.88 / 7.01 ±0.08 / 7.14 ms │           7.00 / 7.15 ±0.09 / 7.26 ms │     no change │
│ QQuery 38 │        63.40 / 65.04 ±2.84 / 70.71 ms │        65.23 / 66.40 ±1.06 / 68.07 ms │     no change │
│ QQuery 39 │        90.78 / 92.28 ±0.83 / 93.15 ms │        91.34 / 92.65 ±0.90 / 94.13 ms │     no change │
│ QQuery 40 │        23.80 / 24.92 ±1.05 / 26.84 ms │        23.89 / 24.28 ±0.36 / 24.79 ms │     no change │
│ QQuery 41 │        11.84 / 12.01 ±0.15 / 12.27 ms │        11.85 / 12.07 ±0.21 / 12.43 ms │     no change │
│ QQuery 42 │        24.43 / 24.95 ±0.50 / 25.81 ms │        24.87 / 25.17 ±0.32 / 25.77 ms │     no change │
│ QQuery 43 │           4.81 / 4.99 ±0.19 / 5.36 ms │           5.11 / 6.19 ±1.90 / 9.98 ms │  1.24x slower │
│ QQuery 44 │           9.24 / 9.34 ±0.06 / 9.39 ms │         9.48 / 10.29 ±1.22 / 12.71 ms │  1.10x slower │
│ QQuery 45 │        38.54 / 41.51 ±3.03 / 47.18 ms │        40.55 / 40.99 ±0.34 / 41.29 ms │     no change │
│ QQuery 46 │        12.16 / 12.58 ±0.29 / 12.95 ms │        12.19 / 12.52 ±0.25 / 12.84 ms │     no change │
│ QQuery 47 │     235.27 / 241.14 ±3.87 / 246.06 ms │     237.16 / 243.41 ±5.86 / 250.96 ms │     no change │
│ QQuery 48 │        96.82 / 97.84 ±0.96 / 99.65 ms │       98.76 / 99.90 ±1.24 / 102.15 ms │     no change │
│ QQuery 49 │        78.01 / 79.00 ±0.84 / 80.17 ms │        78.07 / 79.59 ±1.60 / 82.04 ms │     no change │
│ QQuery 50 │        59.72 / 60.26 ±0.43 / 61.03 ms │        60.36 / 60.96 ±0.37 / 61.46 ms │     no change │
│ QQuery 51 │        93.79 / 95.03 ±1.24 / 96.92 ms │       93.78 / 98.43 ±4.16 / 105.61 ms │     no change │
│ QQuery 52 │        24.44 / 25.11 ±0.45 / 25.83 ms │        25.36 / 25.54 ±0.11 / 25.65 ms │     no change │
│ QQuery 53 │        29.97 / 30.33 ±0.20 / 30.59 ms │        30.52 / 30.84 ±0.22 / 31.17 ms │     no change │
│ QQuery 54 │        56.13 / 56.46 ±0.25 / 56.85 ms │        56.71 / 57.26 ±0.37 / 57.76 ms │     no change │
│ QQuery 55 │        23.97 / 24.18 ±0.22 / 24.60 ms │        24.66 / 24.85 ±0.22 / 25.24 ms │     no change │
│ QQuery 56 │        40.19 / 40.67 ±0.48 / 41.41 ms │        40.13 / 40.68 ±0.67 / 41.93 ms │     no change │
│ QQuery 57 │     178.37 / 182.02 ±3.70 / 189.12 ms │     180.49 / 182.62 ±2.16 / 186.53 ms │     no change │
│ QQuery 58 │     116.98 / 119.47 ±2.61 / 124.41 ms │     118.05 / 119.53 ±1.17 / 121.25 ms │     no change │
│ QQuery 59 │     119.16 / 119.72 ±0.47 / 120.52 ms │     119.45 / 120.40 ±0.82 / 121.44 ms │     no change │
│ QQuery 60 │        40.49 / 42.83 ±3.37 / 49.53 ms │        40.79 / 41.94 ±0.95 / 43.57 ms │     no change │
│ QQuery 61 │        12.45 / 12.60 ±0.14 / 12.85 ms │        12.65 / 12.75 ±0.13 / 12.96 ms │     no change │
│ QQuery 62 │        47.02 / 47.63 ±0.61 / 48.66 ms │        47.35 / 47.65 ±0.32 / 48.16 ms │     no change │
│ QQuery 63 │        30.05 / 30.26 ±0.16 / 30.46 ms │        30.83 / 30.97 ±0.15 / 31.17 ms │     no change │
│ QQuery 64 │     414.93 / 417.96 ±2.89 / 422.88 ms │     413.83 / 416.58 ±3.02 / 422.36 ms │     no change │
│ QQuery 65 │     129.21 / 134.21 ±5.96 / 145.62 ms │     131.41 / 132.73 ±1.54 / 134.70 ms │     no change │
│ QQuery 66 │        80.69 / 82.72 ±1.92 / 85.99 ms │        82.40 / 83.63 ±1.34 / 86.24 ms │     no change │
│ QQuery 67 │     253.51 / 257.34 ±3.89 / 264.12 ms │     252.92 / 257.05 ±5.19 / 266.30 ms │     no change │
│ QQuery 68 │        12.22 / 12.39 ±0.12 / 12.55 ms │        12.28 / 12.56 ±0.20 / 12.80 ms │     no change │
│ QQuery 69 │        58.07 / 58.41 ±0.27 / 58.82 ms │        58.58 / 59.26 ±0.50 / 59.99 ms │     no change │
│ QQuery 70 │     106.46 / 108.12 ±1.28 / 109.76 ms │     107.83 / 109.22 ±1.19 / 111.23 ms │     no change │
│ QQuery 71 │        37.10 / 40.13 ±5.60 / 51.32 ms │        36.39 / 38.78 ±4.29 / 47.35 ms │     no change │
│ QQuery 72 │ 2231.14 / 2331.94 ±80.31 / 2436.04 ms │ 2186.64 / 2244.81 ±45.53 / 2326.38 ms │     no change │
│ QQuery 73 │          9.65 / 9.89 ±0.18 / 10.16 ms │         9.83 / 10.13 ±0.19 / 10.41 ms │     no change │
│ QQuery 74 │     178.57 / 183.13 ±4.64 / 192.03 ms │     181.90 / 185.94 ±4.91 / 195.56 ms │     no change │
│ QQuery 75 │     149.26 / 151.51 ±2.06 / 155.38 ms │     152.99 / 154.78 ±1.29 / 156.42 ms │     no change │
│ QQuery 76 │        35.93 / 36.28 ±0.31 / 36.81 ms │        36.40 / 36.65 ±0.23 / 36.95 ms │     no change │
│ QQuery 77 │        62.65 / 66.30 ±4.93 / 75.34 ms │        63.37 / 64.00 ±0.63 / 65.11 ms │     no change │
│ QQuery 78 │     202.48 / 209.00 ±5.22 / 217.82 ms │     203.98 / 206.51 ±2.26 / 210.35 ms │     no change │
│ QQuery 79 │        68.61 / 69.52 ±0.63 / 70.60 ms │        69.96 / 72.06 ±2.53 / 76.89 ms │     no change │
│ QQuery 80 │     101.57 / 102.10 ±0.32 / 102.54 ms │     100.35 / 102.67 ±1.82 / 105.51 ms │     no change │
│ QQuery 81 │        26.69 / 30.61 ±5.00 / 40.16 ms │        27.09 / 27.35 ±0.14 / 27.50 ms │ +1.12x faster │
│ QQuery 82 │        17.08 / 17.40 ±0.18 / 17.61 ms │        17.41 / 17.65 ±0.17 / 17.82 ms │     no change │
│ QQuery 83 │        40.77 / 41.33 ±0.31 / 41.64 ms │        41.70 / 44.12 ±3.43 / 50.95 ms │  1.07x slower │
│ QQuery 84 │        30.54 / 30.87 ±0.29 / 31.20 ms │        31.27 / 31.57 ±0.28 / 32.08 ms │     no change │
│ QQuery 85 │     107.73 / 111.68 ±5.80 / 123.02 ms │     108.70 / 111.01 ±2.97 / 116.85 ms │     no change │
│ QQuery 86 │        25.96 / 26.78 ±0.55 / 27.54 ms │        26.54 / 28.50 ±2.14 / 31.72 ms │  1.06x slower │
│ QQuery 87 │        63.29 / 64.84 ±1.19 / 66.29 ms │        65.10 / 66.91 ±2.04 / 70.76 ms │     no change │
│ QQuery 88 │        64.38 / 64.72 ±0.21 / 64.99 ms │        65.11 / 66.54 ±2.20 / 70.89 ms │     no change │
│ QQuery 89 │        36.17 / 38.85 ±3.42 / 45.45 ms │        37.08 / 37.54 ±0.25 / 37.85 ms │     no change │
│ QQuery 90 │        17.32 / 17.72 ±0.27 / 18.07 ms │        17.96 / 18.54 ±0.86 / 20.24 ms │     no change │
│ QQuery 91 │        46.02 / 46.81 ±0.77 / 48.28 ms │        46.60 / 48.23 ±1.55 / 51.14 ms │     no change │
│ QQuery 92 │        29.90 / 30.56 ±0.56 / 31.24 ms │        30.85 / 31.55 ±0.59 / 32.55 ms │     no change │
│ QQuery 93 │        50.73 / 51.55 ±0.44 / 52.01 ms │        51.17 / 52.28 ±0.82 / 53.67 ms │     no change │
│ QQuery 94 │        38.87 / 40.38 ±2.62 / 45.60 ms │        39.61 / 39.97 ±0.38 / 40.67 ms │     no change │
│ QQuery 95 │        83.00 / 85.13 ±3.42 / 91.93 ms │        83.87 / 85.47 ±1.51 / 87.93 ms │     no change │
│ QQuery 96 │        24.80 / 24.93 ±0.08 / 25.03 ms │        25.24 / 25.35 ±0.07 / 25.44 ms │     no change │
│ QQuery 97 │        47.11 / 47.43 ±0.19 / 47.65 ms │        47.94 / 48.43 ±0.36 / 48.86 ms │     no change │
│ QQuery 98 │        43.43 / 45.01 ±1.39 / 47.63 ms │        43.98 / 44.56 ±0.39 / 45.18 ms │     no change │
│ QQuery 99 │        70.74 / 71.16 ±0.42 / 71.94 ms │        71.65 / 73.14 ±1.02 / 74.56 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 10309.60ms │
│ Total Time (refactor-materializing-smj)   │ 10275.15ms │
│ Average Time (HEAD)                       │   104.14ms │
│ Average Time (refactor-materializing-smj) │   103.79ms │
│ Queries Faster                            │          2 │
│ Queries Slower                            │          4 │
│ Queries with No Change                    │         93 │
│ Queries with Failure                      │          0 │
└───────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	2.0 GiB
Avg memory	1.3 GiB
CPU user	233.9s
CPU sys	6.2s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	55.0s
Peak memory	2.3 GiB
Avg memory	1.6 GiB
CPU user	233.9s
CPU sys	6.2s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-29T15:29:45Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃             refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.63 / 4.66 ±5.87 / 16.40 ms │           1.59 / 4.53 ±5.71 / 15.94 ms │     no change │
│ QQuery 1  │        13.91 / 14.41 ±0.42 / 15.02 ms │         13.85 / 13.98 ±0.07 / 14.05 ms │     no change │
│ QQuery 2  │        37.67 / 38.39 ±0.97 / 40.31 ms │         37.37 / 37.69 ±0.32 / 38.13 ms │     no change │
│ QQuery 3  │        33.17 / 34.20 ±0.95 / 35.79 ms │         32.72 / 33.22 ±0.30 / 33.53 ms │     no change │
│ QQuery 4  │    243.44 / 255.91 ±12.23 / 278.48 ms │      224.13 / 242.47 ±9.34 / 249.44 ms │ +1.06x faster │
│ QQuery 5  │     272.87 / 276.46 ±3.67 / 283.27 ms │      271.69 / 273.61 ±1.95 / 277.17 ms │     no change │
│ QQuery 6  │           1.30 / 1.45 ±0.22 / 1.88 ms │            1.32 / 1.47 ±0.23 / 1.92 ms │     no change │
│ QQuery 7  │        14.34 / 15.53 ±2.17 / 19.86 ms │         13.70 / 13.87 ±0.12 / 14.08 ms │ +1.12x faster │
│ QQuery 8  │     328.55 / 331.08 ±3.24 / 336.97 ms │      325.45 / 330.70 ±4.08 / 337.23 ms │     no change │
│ QQuery 9  │     456.72 / 464.27 ±8.16 / 479.43 ms │      451.42 / 464.30 ±9.06 / 478.37 ms │     no change │
│ QQuery 10 │        69.67 / 71.36 ±1.33 / 73.64 ms │         70.16 / 70.79 ±0.74 / 71.82 ms │     no change │
│ QQuery 11 │        81.74 / 82.85 ±0.64 / 83.51 ms │         81.47 / 83.41 ±1.59 / 86.23 ms │     no change │
│ QQuery 12 │     269.60 / 280.21 ±8.97 / 296.27 ms │      267.10 / 273.34 ±5.31 / 279.36 ms │     no change │
│ QQuery 13 │    407.77 / 433.96 ±24.27 / 477.12 ms │      370.22 / 373.68 ±4.70 / 382.73 ms │ +1.16x faster │
│ QQuery 14 │     287.17 / 295.88 ±9.05 / 313.31 ms │      280.01 / 285.63 ±5.44 / 293.29 ms │     no change │
│ QQuery 15 │     274.47 / 280.46 ±5.36 / 288.68 ms │      266.48 / 272.33 ±4.48 / 277.55 ms │     no change │
│ QQuery 16 │     610.22 / 623.80 ±7.33 / 632.51 ms │     611.15 / 633.43 ±14.07 / 654.39 ms │     no change │
│ QQuery 17 │    613.35 / 628.81 ±11.68 / 639.60 ms │     628.88 / 664.48 ±34.48 / 718.14 ms │  1.06x slower │
│ QQuery 18 │ 1248.43 / 1325.31 ±81.62 / 1449.44 ms │  1264.37 / 1308.50 ±33.77 / 1363.06 ms │     no change │
│ QQuery 19 │        29.37 / 29.87 ±0.59 / 30.98 ms │         28.42 / 30.63 ±3.09 / 36.59 ms │     no change │
│ QQuery 20 │    517.98 / 532.48 ±18.05 / 567.68 ms │     521.28 / 553.41 ±28.56 / 605.93 ms │     no change │
│ QQuery 21 │     515.15 / 522.74 ±5.82 / 532.19 ms │      520.27 / 531.61 ±8.41 / 544.66 ms │     no change │
│ QQuery 22 │    989.05 / 997.40 ±7.35 / 1010.64 ms │     989.80 / 998.89 ±9.03 / 1014.92 ms │     no change │
│ QQuery 23 │ 3124.26 / 3157.92 ±39.32 / 3233.59 ms │  3090.65 / 3136.35 ±63.85 / 3262.52 ms │     no change │
│ QQuery 24 │       41.71 / 47.75 ±10.33 / 68.33 ms │         41.43 / 42.56 ±0.80 / 43.86 ms │ +1.12x faster │
│ QQuery 25 │     111.07 / 115.70 ±5.10 / 122.43 ms │      111.05 / 113.39 ±1.75 / 116.35 ms │     no change │
│ QQuery 26 │       41.87 / 50.27 ±10.01 / 68.56 ms │         41.68 / 42.17 ±0.46 / 43.02 ms │ +1.19x faster │
│ QQuery 27 │     512.16 / 529.64 ±9.39 / 537.03 ms │      517.91 / 531.34 ±9.42 / 543.82 ms │     no change │
│ QQuery 28 │ 2891.26 / 2968.39 ±50.51 / 3023.80 ms │  2917.22 / 2932.91 ±17.86 / 2966.65 ms │     no change │
│ QQuery 29 │        42.31 / 45.24 ±4.83 / 54.89 ms │         41.09 / 41.38 ±0.23 / 41.76 ms │ +1.09x faster │
│ QQuery 30 │    298.31 / 317.51 ±23.68 / 363.93 ms │      303.90 / 316.00 ±9.84 / 328.66 ms │     no change │
│ QQuery 31 │     275.74 / 288.49 ±7.98 / 299.54 ms │     284.87 / 299.69 ±14.41 / 326.54 ms │     no change │
│ QQuery 32 │    921.87 / 939.13 ±11.14 / 954.71 ms │    921.71 / 967.84 ±41.36 / 1028.23 ms │     no change │
│ QQuery 33 │ 1462.13 / 1534.72 ±97.05 / 1721.29 ms │ 1496.32 / 1575.32 ±110.15 / 1789.57 ms │     no change │
│ QQuery 34 │ 1466.24 / 1519.28 ±34.52 / 1561.38 ms │  1526.35 / 1600.91 ±89.65 / 1760.60 ms │  1.05x slower │
│ QQuery 35 │    281.91 / 300.12 ±21.02 / 335.30 ms │     278.54 / 296.00 ±25.29 / 346.15 ms │     no change │
│ QQuery 36 │        67.21 / 73.40 ±4.65 / 79.05 ms │         65.31 / 72.03 ±3.50 / 75.31 ms │     no change │
│ QQuery 37 │        35.83 / 37.77 ±2.54 / 42.75 ms │         36.42 / 40.14 ±3.97 / 45.93 ms │  1.06x slower │
│ QQuery 38 │        40.53 / 45.19 ±5.32 / 55.09 ms │         41.22 / 45.13 ±2.67 / 49.51 ms │     no change │
│ QQuery 39 │     137.16 / 151.07 ±8.54 / 161.35 ms │      143.36 / 156.30 ±7.56 / 166.80 ms │     no change │
│ QQuery 40 │        14.65 / 16.68 ±3.00 / 22.53 ms │         14.78 / 17.32 ±4.50 / 26.30 ms │     no change │
│ QQuery 41 │       14.07 / 20.00 ±10.93 / 41.82 ms │         13.99 / 14.14 ±0.13 / 14.35 ms │ +1.41x faster │
│ QQuery 42 │        13.39 / 15.48 ±2.44 / 19.18 ms │         13.16 / 14.11 ±1.45 / 16.98 ms │ +1.10x faster │
└───────────┴───────────────────────────────────────┴────────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 19715.23ms │
│ Total Time (refactor-materializing-smj)   │ 19750.99ms │
│ Average Time (HEAD)                       │   458.49ms │
│ Average Time (refactor-materializing-smj) │   459.33ms │
│ Queries Faster                            │          8 │
│ Queries Slower                            │          3 │
│ Queries with No Change                    │         32 │
│ Queries with Failure                      │          0 │
└───────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	100.0s
Peak memory	10.5 GiB
Avg memory	4.5 GiB
CPU user	1009.2s
CPU sys	73.8s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	100.0s
Peak memory	11.9 GiB
Avg memory	4.6 GiB
CPU user	1007.7s
CPU sys	75.9s
Peak spill	0 B

File an issue against this benchmark runner

codecov-commenter · 2026-07-29T15:55:37Z

Codecov Report

❌ Patch coverage is 87.75510% with 42 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.84%. Comparing base (68d5874) to head (cb881c7).
⚠️ Report is 25 commits behind head on main.

Files with missing lines	Patch %	Lines
.../src/joins/sort_merge_join/materializing_stream.rs	87.71%	11 Missing and 31 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #23976      +/-   ##
==========================================
+ Coverage   80.75%   80.84%   +0.08%     
==========================================
  Files        1096     1096              
  Lines      373282   373902     +620     
  Branches   373282   373902     +620     
==========================================
+ Hits       301440   302274     +834     
+ Misses      53869    53575     -294     
- Partials    17973    18053      +80

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

rluvaton · 2026-07-30T15:09:49Z

run benchmark smj

rluvaton · 2026-07-30T15:10:58Z

run benchmarks

adriangbot · 2026-07-30T15:12:35Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5132631045-1301-xr9hs 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 refactor-materializing-smj (5bab8f5) to 68d5874 (merge-base) diff using: smj
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:13:07Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5132649709-1302-krvm7 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 refactor-materializing-smj (5bab8f5) to 68d5874 (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:13:17Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5132649709-1303-44pmk 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 refactor-materializing-smj (5bab8f5) to 68d5874 (merge-base) diff using: tpcds
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:13:44Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5132649709-1304-pvp6s 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 refactor-materializing-smj (5bab8f5) to 68d5874 (merge-base) diff using: tpch
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:27:21Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark smj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃           refactor-materializing-smj ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │           8.64 / 8.73 ±0.15 / 9.02 ms │          8.09 / 8.40 ±0.24 / 8.82 ms │    no change │
│ QQuery 2  │     162.14 / 164.25 ±1.69 / 166.49 ms │    161.72 / 163.73 ±1.94 / 166.85 ms │    no change │
│ QQuery 3  │     103.51 / 105.17 ±0.99 / 106.60 ms │    102.30 / 105.50 ±1.95 / 107.96 ms │    no change │
│ QQuery 4  │        28.49 / 28.84 ±0.25 / 29.15 ms │       28.02 / 28.18 ±0.11 / 28.32 ms │    no change │
│ QQuery 5  │        20.52 / 20.76 ±0.14 / 20.94 ms │       20.43 / 20.47 ±0.05 / 20.55 ms │    no change │
│ QQuery 6  │     158.00 / 162.26 ±4.05 / 169.96 ms │    155.49 / 159.62 ±2.51 / 162.73 ms │    no change │
│ QQuery 7  │     200.87 / 203.32 ±1.57 / 205.41 ms │    199.07 / 200.54 ±0.80 / 201.37 ms │    no change │
│ QQuery 8  │        19.67 / 19.93 ±0.14 / 20.10 ms │       19.10 / 19.26 ±0.13 / 19.40 ms │    no change │
│ QQuery 9  │     207.98 / 214.00 ±3.48 / 217.52 ms │    206.72 / 208.83 ±2.23 / 212.94 ms │    no change │
│ QQuery 10 │        57.65 / 58.16 ±0.32 / 58.48 ms │       59.15 / 59.45 ±0.23 / 59.78 ms │    no change │
│ QQuery 11 │        27.30 / 30.07 ±4.43 / 38.87 ms │       27.63 / 31.22 ±6.00 / 43.15 ms │    no change │
│ QQuery 12 │        66.72 / 69.85 ±2.25 / 73.20 ms │       70.66 / 73.03 ±1.33 / 74.59 ms │    no change │
│ QQuery 13 │      96.82 / 105.70 ±4.56 / 109.12 ms │    101.37 / 108.36 ±3.67 / 111.89 ms │    no change │
│ QQuery 14 │        58.22 / 60.00 ±2.90 / 65.77 ms │       59.37 / 63.08 ±5.35 / 73.40 ms │ 1.05x slower │
│ QQuery 15 │        57.69 / 58.27 ±0.34 / 58.69 ms │       59.08 / 59.40 ±0.26 / 59.81 ms │    no change │
│ QQuery 16 │        11.41 / 11.58 ±0.09 / 11.67 ms │       11.83 / 11.92 ±0.06 / 11.98 ms │    no change │
│ QQuery 17 │     145.77 / 147.05 ±0.81 / 148.16 ms │    147.91 / 152.74 ±7.95 / 168.60 ms │    no change │
│ QQuery 18 │     111.31 / 112.11 ±0.54 / 112.81 ms │    112.44 / 112.94 ±0.32 / 113.36 ms │    no change │
│ QQuery 19 │     289.62 / 295.00 ±4.97 / 303.07 ms │    292.38 / 302.20 ±7.69 / 311.75 ms │    no change │
│ QQuery 20 │ 1262.67 / 1274.66 ±14.71 / 1301.61 ms │ 1244.53 / 1250.58 ±4.67 / 1256.28 ms │    no change │
│ QQuery 21 │       94.18 / 98.10 ±6.23 / 110.50 ms │      89.32 / 96.57 ±8.72 / 113.56 ms │    no change │
│ QQuery 22 │     100.32 / 103.06 ±1.51 / 104.43 ms │    100.72 / 101.92 ±0.96 / 103.60 ms │    no change │
│ QQuery 23 │     108.67 / 110.06 ±1.41 / 112.62 ms │    104.59 / 107.16 ±2.77 / 112.45 ms │    no change │
│ QQuery 24 │        27.35 / 27.44 ±0.09 / 27.58 ms │       27.95 / 28.07 ±0.07 / 28.14 ms │    no change │
│ QQuery 25 │        70.79 / 73.17 ±3.14 / 79.35 ms │       70.70 / 73.43 ±1.68 / 75.66 ms │    no change │
│ QQuery 26 │     100.92 / 109.92 ±4.83 / 115.57 ms │    109.89 / 113.36 ±2.40 / 116.18 ms │    no change │
└───────────┴───────────────────────────────────────┴──────────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 3671.48ms │
│ Total Time (refactor-materializing-smj)   │ 3659.97ms │
│ Average Time (HEAD)                       │  141.21ms │
│ Average Time (refactor-materializing-smj) │  140.77ms │
│ Queries Faster                            │         0 │
│ Queries Slower                            │         1 │
│ Queries with No Change                    │        25 │
│ Queries with Failure                      │         0 │
└───────────────────────────────────────────┴───────────┘

Resource Usage

smj — base (merge-base)

Metric	Value
Wall time	20.0s
Peak memory	1.2 GiB
Avg memory	460.1 MiB
CPU user	157.3s
CPU sys	3.3s
Peak spill	0 B

smj — branch

Metric	Value
Wall time	20.0s
Peak memory	1.2 GiB
Avg memory	454.1 MiB
CPU user	156.1s
CPU sys	3.6s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:27:35Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃     refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 39.54 / 40.62 ±1.36 / 43.29 ms │ 39.42 / 40.11 ±1.07 / 42.23 ms │     no change │
│ QQuery 2  │ 19.95 / 20.28 ±0.30 / 20.84 ms │ 19.85 / 20.53 ±0.65 / 21.76 ms │     no change │
│ QQuery 3  │ 32.09 / 34.73 ±2.41 / 38.69 ms │ 33.52 / 34.35 ±0.47 / 34.94 ms │     no change │
│ QQuery 4  │ 18.05 / 18.53 ±0.61 / 19.71 ms │ 18.09 / 18.46 ±0.51 / 19.46 ms │     no change │
│ QQuery 5  │ 39.90 / 43.12 ±2.26 / 45.91 ms │ 42.20 / 42.53 ±0.32 / 43.10 ms │     no change │
│ QQuery 6  │ 17.57 / 17.84 ±0.28 / 18.33 ms │ 17.00 / 17.26 ±0.21 / 17.53 ms │     no change │
│ QQuery 7  │ 46.39 / 48.43 ±1.13 / 49.71 ms │ 47.15 / 49.01 ±1.02 / 50.06 ms │     no change │
│ QQuery 8  │ 44.39 / 45.45 ±0.88 / 46.91 ms │ 44.31 / 45.24 ±0.91 / 46.93 ms │     no change │
│ QQuery 9  │ 51.39 / 53.93 ±1.73 / 56.58 ms │ 51.51 / 52.47 ±0.64 / 53.07 ms │     no change │
│ QQuery 10 │ 44.97 / 45.58 ±0.39 / 46.20 ms │ 43.92 / 44.58 ±0.46 / 45.35 ms │     no change │
│ QQuery 11 │ 14.29 / 14.76 ±0.26 / 15.05 ms │ 14.11 / 14.33 ±0.17 / 14.50 ms │     no change │
│ QQuery 12 │ 25.26 / 26.69 ±1.98 / 30.60 ms │ 24.33 / 25.94 ±1.72 / 29.20 ms │     no change │
│ QQuery 13 │ 32.86 / 35.84 ±2.43 / 39.94 ms │ 32.54 / 35.03 ±3.05 / 40.97 ms │     no change │
│ QQuery 14 │ 24.46 / 25.33 ±1.27 / 27.83 ms │ 24.32 / 24.63 ±0.20 / 24.86 ms │     no change │
│ QQuery 15 │ 32.50 / 32.88 ±0.20 / 33.12 ms │ 32.12 / 32.85 ±0.81 / 34.40 ms │     no change │
│ QQuery 16 │ 14.50 / 14.62 ±0.11 / 14.79 ms │ 14.70 / 14.84 ±0.17 / 15.16 ms │     no change │
│ QQuery 17 │ 77.91 / 79.75 ±1.28 / 81.82 ms │ 75.08 / 75.83 ±0.56 / 76.38 ms │     no change │
│ QQuery 18 │ 63.38 / 64.93 ±1.38 / 66.75 ms │ 61.52 / 62.86 ±0.73 / 63.60 ms │     no change │
│ QQuery 19 │ 34.50 / 35.06 ±0.50 / 35.99 ms │ 33.74 / 34.23 ±0.42 / 35.01 ms │     no change │
│ QQuery 20 │ 33.33 / 34.08 ±0.55 / 34.97 ms │ 33.42 / 33.69 ±0.32 / 34.27 ms │     no change │
│ QQuery 21 │ 58.84 / 59.40 ±0.75 / 60.88 ms │ 57.96 / 58.81 ±0.58 / 59.69 ms │     no change │
│ QQuery 22 │ 14.95 / 16.41 ±2.51 / 21.42 ms │ 14.64 / 14.99 ±0.27 / 15.32 ms │ +1.09x faster │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                         ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 808.28ms │
│ Total Time (refactor-materializing-smj)   │ 792.57ms │
│ Average Time (HEAD)                       │  36.74ms │
│ Average Time (refactor-materializing-smj) │  36.03ms │
│ Queries Faster                            │        1 │
│ Queries Slower                            │        0 │
│ Queries with No Change                    │       21 │
│ Queries with Failure                      │        0 │
└───────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	1.2 GiB
Avg memory	687.5 MiB
CPU user	23.4s
CPU sys	1.9s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	1.2 GiB
Avg memory	514.6 MiB
CPU user	23.1s
CPU sys	1.9s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:29:03Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃            refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.60 / 6.12 ±0.90 / 7.91 ms │           5.50 / 6.09 ±0.92 / 7.89 ms │     no change │
│ QQuery 2  │        81.56 / 81.82 ±0.20 / 82.11 ms │        81.52 / 82.06 ±0.42 / 82.54 ms │     no change │
│ QQuery 3  │        29.80 / 30.08 ±0.19 / 30.28 ms │        29.82 / 30.14 ±0.21 / 30.44 ms │     no change │
│ QQuery 4  │     496.11 / 499.65 ±2.07 / 501.86 ms │     494.59 / 500.83 ±4.93 / 507.93 ms │     no change │
│ QQuery 5  │        53.14 / 53.59 ±0.41 / 54.34 ms │        52.29 / 52.77 ±0.26 / 53.09 ms │     no change │
│ QQuery 6  │        36.66 / 37.16 ±0.28 / 37.43 ms │        36.57 / 36.84 ±0.21 / 37.14 ms │     no change │
│ QQuery 7  │       95.26 / 97.85 ±2.58 / 102.80 ms │       95.92 / 97.68 ±2.10 / 101.57 ms │     no change │
│ QQuery 8  │        37.26 / 37.80 ±0.60 / 38.96 ms │        37.20 / 37.61 ±0.25 / 37.91 ms │     no change │
│ QQuery 9  │        54.75 / 56.58 ±2.38 / 61.22 ms │        52.40 / 55.17 ±1.66 / 57.35 ms │     no change │
│ QQuery 10 │        63.87 / 64.58 ±0.38 / 64.91 ms │        63.22 / 63.72 ±0.26 / 63.91 ms │     no change │
│ QQuery 11 │     309.84 / 312.33 ±2.30 / 315.56 ms │     306.55 / 312.37 ±5.53 / 319.30 ms │     no change │
│ QQuery 12 │        28.77 / 29.18 ±0.26 / 29.55 ms │        28.88 / 29.62 ±0.68 / 30.47 ms │     no change │
│ QQuery 13 │     119.96 / 121.70 ±1.64 / 124.29 ms │     119.72 / 121.91 ±3.36 / 128.58 ms │     no change │
│ QQuery 14 │     418.39 / 423.35 ±3.61 / 428.98 ms │     415.75 / 419.43 ±3.96 / 426.62 ms │     no change │
│ QQuery 15 │        57.62 / 59.15 ±1.22 / 61.30 ms │        58.73 / 59.92 ±0.68 / 60.61 ms │     no change │
│ QQuery 16 │           6.84 / 6.99 ±0.18 / 7.34 ms │           6.70 / 6.86 ±0.24 / 7.33 ms │     no change │
│ QQuery 17 │        79.60 / 81.54 ±2.13 / 85.65 ms │        80.68 / 81.88 ±1.09 / 83.66 ms │     no change │
│ QQuery 18 │     123.05 / 125.37 ±1.80 / 128.07 ms │     123.75 / 125.91 ±1.16 / 127.16 ms │     no change │
│ QQuery 19 │        42.17 / 42.58 ±0.36 / 43.14 ms │        42.48 / 42.75 ±0.17 / 42.96 ms │     no change │
│ QQuery 20 │        36.76 / 37.91 ±1.24 / 39.68 ms │        35.88 / 37.84 ±1.49 / 40.18 ms │     no change │
│ QQuery 21 │        17.87 / 18.22 ±0.25 / 18.54 ms │        17.74 / 18.21 ±0.46 / 19.01 ms │     no change │
│ QQuery 22 │        63.10 / 64.19 ±0.76 / 65.12 ms │        63.80 / 64.31 ±0.62 / 65.52 ms │     no change │
│ QQuery 23 │    345.04 / 359.96 ±16.57 / 391.48 ms │     347.63 / 352.58 ±5.77 / 363.65 ms │     no change │
│ QQuery 24 │     226.14 / 229.08 ±2.24 / 231.79 ms │     226.67 / 230.16 ±5.07 / 240.19 ms │     no change │
│ QQuery 25 │     110.36 / 111.55 ±1.24 / 113.85 ms │     111.57 / 112.41 ±1.32 / 115.00 ms │     no change │
│ QQuery 26 │        58.72 / 59.12 ±0.30 / 59.58 ms │        58.26 / 61.26 ±3.41 / 67.38 ms │     no change │
│ QQuery 27 │           6.45 / 6.59 ±0.17 / 6.91 ms │           6.38 / 6.51 ±0.13 / 6.73 ms │     no change │
│ QQuery 28 │        57.94 / 61.11 ±2.41 / 64.00 ms │        58.07 / 61.64 ±1.93 / 63.96 ms │     no change │
│ QQuery 29 │       97.81 / 99.16 ±1.79 / 102.61 ms │       98.20 / 99.71 ±1.66 / 102.80 ms │     no change │
│ QQuery 30 │        32.65 / 33.01 ±0.37 / 33.61 ms │        33.75 / 35.39 ±2.01 / 39.26 ms │  1.07x slower │
│ QQuery 31 │     112.25 / 112.53 ±0.23 / 112.89 ms │     112.56 / 113.38 ±0.90 / 115.11 ms │     no change │
│ QQuery 32 │        20.57 / 20.80 ±0.17 / 21.02 ms │        20.65 / 21.04 ±0.23 / 21.36 ms │     no change │
│ QQuery 33 │        38.46 / 38.66 ±0.27 / 39.19 ms │        38.58 / 40.01 ±1.35 / 42.26 ms │     no change │
│ QQuery 34 │         9.83 / 10.14 ±0.26 / 10.45 ms │         9.95 / 10.27 ±0.20 / 10.44 ms │     no change │
│ QQuery 35 │        73.53 / 73.84 ±0.33 / 74.41 ms │        73.33 / 74.13 ±0.41 / 74.43 ms │     no change │
│ QQuery 36 │           5.85 / 5.97 ±0.18 / 6.32 ms │           5.82 / 5.95 ±0.18 / 6.30 ms │     no change │
│ QQuery 37 │           6.92 / 7.00 ±0.07 / 7.10 ms │           6.96 / 7.02 ±0.05 / 7.09 ms │     no change │
│ QQuery 38 │        62.88 / 63.00 ±0.15 / 63.27 ms │        62.02 / 62.30 ±0.24 / 62.58 ms │     no change │
│ QQuery 39 │        91.69 / 92.69 ±1.32 / 95.27 ms │        90.76 / 92.22 ±1.70 / 95.47 ms │     no change │
│ QQuery 40 │        23.60 / 23.99 ±0.37 / 24.47 ms │        23.79 / 24.07 ±0.22 / 24.37 ms │     no change │
│ QQuery 41 │        11.68 / 11.89 ±0.24 / 12.34 ms │        11.72 / 12.09 ±0.30 / 12.58 ms │     no change │
│ QQuery 42 │        24.35 / 24.69 ±0.36 / 25.37 ms │        24.37 / 24.88 ±0.47 / 25.57 ms │     no change │
│ QQuery 43 │           5.10 / 5.23 ±0.19 / 5.60 ms │           5.16 / 5.27 ±0.20 / 5.67 ms │     no change │
│ QQuery 44 │           9.48 / 9.54 ±0.06 / 9.64 ms │           9.56 / 9.66 ±0.09 / 9.82 ms │     no change │
│ QQuery 45 │        38.31 / 39.32 ±1.43 / 42.13 ms │        38.57 / 40.34 ±2.04 / 44.22 ms │     no change │
│ QQuery 46 │        11.88 / 12.23 ±0.44 / 13.10 ms │        12.19 / 12.83 ±0.46 / 13.38 ms │     no change │
│ QQuery 47 │     231.15 / 235.85 ±4.81 / 244.52 ms │     234.12 / 237.48 ±2.20 / 239.92 ms │     no change │
│ QQuery 48 │        97.40 / 98.15 ±0.54 / 99.08 ms │        98.01 / 98.48 ±0.25 / 98.71 ms │     no change │
│ QQuery 49 │        77.16 / 78.02 ±1.65 / 81.32 ms │        78.15 / 80.51 ±3.30 / 86.91 ms │     no change │
│ QQuery 50 │        59.77 / 60.68 ±0.69 / 61.82 ms │        60.57 / 60.97 ±0.48 / 61.88 ms │     no change │
│ QQuery 51 │        93.03 / 93.90 ±0.72 / 94.98 ms │        93.92 / 94.47 ±0.38 / 94.93 ms │     no change │
│ QQuery 52 │        24.80 / 25.02 ±0.22 / 25.38 ms │        24.97 / 27.04 ±3.42 / 33.82 ms │  1.08x slower │
│ QQuery 53 │        30.57 / 32.18 ±2.80 / 37.76 ms │        30.71 / 31.39 ±0.51 / 32.03 ms │     no change │
│ QQuery 54 │        56.80 / 57.26 ±0.38 / 57.89 ms │        57.00 / 57.50 ±0.41 / 57.93 ms │     no change │
│ QQuery 55 │        23.92 / 24.55 ±0.79 / 26.10 ms │        24.47 / 24.59 ±0.14 / 24.79 ms │     no change │
│ QQuery 56 │        39.35 / 39.99 ±0.53 / 40.74 ms │        39.72 / 40.24 ±0.34 / 40.74 ms │     no change │
│ QQuery 57 │     176.89 / 179.16 ±2.20 / 183.34 ms │     179.21 / 182.31 ±3.52 / 189.18 ms │     no change │
│ QQuery 58 │     116.43 / 118.71 ±3.23 / 125.06 ms │     117.45 / 119.88 ±2.26 / 123.56 ms │     no change │
│ QQuery 59 │     118.77 / 119.26 ±0.38 / 119.86 ms │     120.48 / 120.87 ±0.29 / 121.31 ms │     no change │
│ QQuery 60 │        40.07 / 40.58 ±0.27 / 40.87 ms │        40.70 / 42.16 ±1.87 / 45.78 ms │     no change │
│ QQuery 61 │        12.66 / 13.32 ±1.20 / 15.73 ms │        12.26 / 12.46 ±0.18 / 12.77 ms │ +1.07x faster │
│ QQuery 62 │        47.17 / 47.60 ±0.33 / 48.07 ms │        47.37 / 47.78 ±0.29 / 48.15 ms │     no change │
│ QQuery 63 │        30.48 / 30.68 ±0.24 / 31.00 ms │        30.54 / 31.03 ±0.28 / 31.38 ms │     no change │
│ QQuery 64 │     407.58 / 410.76 ±2.36 / 413.44 ms │     416.24 / 422.95 ±6.60 / 432.69 ms │     no change │
│ QQuery 65 │     127.76 / 130.35 ±2.41 / 133.66 ms │     128.24 / 131.95 ±2.47 / 135.60 ms │     no change │
│ QQuery 66 │        80.38 / 81.43 ±0.83 / 82.80 ms │        82.86 / 83.02 ±0.13 / 83.22 ms │     no change │
│ QQuery 67 │     244.27 / 247.12 ±3.18 / 251.02 ms │     244.37 / 249.10 ±4.22 / 256.90 ms │     no change │
│ QQuery 68 │        11.88 / 12.15 ±0.24 / 12.55 ms │        12.09 / 13.83 ±3.10 / 20.00 ms │  1.14x slower │
│ QQuery 69 │        58.34 / 61.10 ±2.14 / 63.71 ms │        58.73 / 60.42 ±1.77 / 63.78 ms │     no change │
│ QQuery 70 │     106.86 / 108.55 ±2.41 / 113.28 ms │     107.61 / 109.18 ±1.65 / 111.92 ms │     no change │
│ QQuery 71 │        35.66 / 36.40 ±0.56 / 37.33 ms │        35.71 / 37.92 ±3.83 / 45.58 ms │     no change │
│ QQuery 72 │ 2148.41 / 2190.76 ±22.54 / 2214.29 ms │ 1984.41 / 2124.42 ±83.77 / 2230.21 ms │     no change │
│ QQuery 73 │          9.53 / 9.75 ±0.29 / 10.33 ms │          9.71 / 9.94 ±0.29 / 10.51 ms │     no change │
│ QQuery 74 │     171.57 / 175.52 ±2.60 / 177.96 ms │     172.89 / 177.96 ±5.56 / 188.42 ms │     no change │
│ QQuery 75 │     149.40 / 154.04 ±6.87 / 167.64 ms │     151.37 / 152.64 ±0.92 / 154.11 ms │     no change │
│ QQuery 76 │        35.82 / 36.10 ±0.37 / 36.81 ms │        36.06 / 36.45 ±0.42 / 37.15 ms │     no change │
│ QQuery 77 │        62.34 / 63.51 ±1.68 / 66.83 ms │        62.21 / 62.43 ±0.17 / 62.68 ms │     no change │
│ QQuery 78 │     198.51 / 200.52 ±2.01 / 204.33 ms │     201.98 / 203.83 ±2.45 / 208.56 ms │     no change │
│ QQuery 79 │        68.09 / 69.50 ±1.09 / 71.46 ms │        68.54 / 71.18 ±4.26 / 79.67 ms │     no change │
│ QQuery 80 │      98.92 / 100.58 ±1.32 / 102.49 ms │     101.78 / 103.92 ±1.42 / 106.05 ms │     no change │
│ QQuery 81 │        26.06 / 27.66 ±2.56 / 32.76 ms │        26.17 / 26.40 ±0.13 / 26.54 ms │     no change │
│ QQuery 82 │        17.28 / 19.04 ±3.12 / 25.27 ms │        16.48 / 17.38 ±1.01 / 18.83 ms │ +1.10x faster │
│ QQuery 83 │        40.57 / 41.47 ±0.63 / 42.35 ms │        40.86 / 43.50 ±3.52 / 50.48 ms │     no change │
│ QQuery 84 │        30.84 / 31.16 ±0.23 / 31.53 ms │        30.65 / 31.16 ±0.49 / 31.92 ms │     no change │
│ QQuery 85 │     107.24 / 109.25 ±1.53 / 111.90 ms │     107.13 / 108.67 ±1.91 / 112.16 ms │     no change │
│ QQuery 86 │        26.33 / 28.29 ±2.97 / 34.19 ms │        25.33 / 25.95 ±0.41 / 26.56 ms │ +1.09x faster │
│ QQuery 87 │        63.15 / 64.22 ±0.86 / 65.61 ms │        63.38 / 64.90 ±2.16 / 69.17 ms │     no change │
│ QQuery 88 │        64.80 / 65.09 ±0.28 / 65.48 ms │        63.90 / 64.86 ±1.08 / 66.95 ms │     no change │
│ QQuery 89 │        36.46 / 36.55 ±0.08 / 36.69 ms │        36.48 / 36.67 ±0.12 / 36.82 ms │     no change │
│ QQuery 90 │        17.44 / 18.52 ±1.93 / 22.38 ms │        17.20 / 17.41 ±0.16 / 17.63 ms │ +1.06x faster │
│ QQuery 91 │        46.70 / 47.42 ±0.51 / 47.99 ms │        45.72 / 47.31 ±2.22 / 51.72 ms │     no change │
│ QQuery 92 │        30.02 / 30.76 ±0.44 / 31.37 ms │        29.90 / 30.60 ±0.36 / 30.91 ms │     no change │
│ QQuery 93 │        50.61 / 50.95 ±0.26 / 51.27 ms │        50.53 / 51.23 ±0.50 / 52.00 ms │     no change │
│ QQuery 94 │        38.68 / 39.06 ±0.32 / 39.51 ms │        38.39 / 38.81 ±0.34 / 39.22 ms │     no change │
│ QQuery 95 │        81.31 / 82.62 ±0.96 / 83.64 ms │        80.41 / 82.66 ±2.12 / 86.42 ms │     no change │
│ QQuery 96 │        24.37 / 24.58 ±0.15 / 24.81 ms │        24.49 / 24.70 ±0.19 / 24.96 ms │     no change │
│ QQuery 97 │        46.91 / 47.25 ±0.22 / 47.58 ms │        47.24 / 47.73 ±0.67 / 49.06 ms │     no change │
│ QQuery 98 │        43.48 / 43.99 ±0.47 / 44.82 ms │        43.35 / 43.84 ±0.58 / 44.91 ms │     no change │
│ QQuery 99 │        71.14 / 72.66 ±1.57 / 75.53 ms │        70.90 / 72.91 ±2.57 / 77.39 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 10062.44ms │
│ Total Time (refactor-materializing-smj)   │ 10036.05ms │
│ Average Time (HEAD)                       │   101.64ms │
│ Average Time (refactor-materializing-smj) │   101.37ms │
│ Queries Faster                            │          4 │
│ Queries Slower                            │          3 │
│ Queries with No Change                    │         92 │
│ Queries with Failure                      │          0 │
└───────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	2.4 GiB
Avg memory	1.6 GiB
CPU user	226.7s
CPU sys	6.2s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	55.0s
Peak memory	2.3 GiB
Avg memory	1.5 GiB
CPU user	225.6s
CPU sys	6.0s
Peak spill	0 B

File an issue against this benchmark runner

adriangbot · 2026-07-30T15:31:39Z

🤖 Benchmark completed (GKE) | trigger

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

CPU Details (lscpu)

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

Details

Comparing HEAD and refactor-materializing-smj
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃            refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.28 / 4.05 ±5.45 / 14.95 ms │          1.28 / 4.05 ±5.47 / 14.98 ms │     no change │
│ QQuery 1  │        13.49 / 13.68 ±0.22 / 14.09 ms │        13.21 / 13.61 ±0.21 / 13.80 ms │     no change │
│ QQuery 2  │        36.71 / 36.96 ±0.27 / 37.40 ms │        36.31 / 36.55 ±0.17 / 36.84 ms │     no change │
│ QQuery 3  │        31.25 / 32.01 ±0.63 / 33.17 ms │        30.77 / 31.16 ±0.31 / 31.63 ms │     no change │
│ QQuery 4  │     224.68 / 227.21 ±3.35 / 233.73 ms │     223.02 / 225.75 ±2.98 / 231.18 ms │     no change │
│ QQuery 5  │     271.23 / 274.16 ±1.58 / 275.57 ms │     273.01 / 274.05 ±1.20 / 276.27 ms │     no change │
│ QQuery 6  │           1.30 / 1.45 ±0.23 / 1.90 ms │           1.32 / 1.51 ±0.30 / 2.11 ms │     no change │
│ QQuery 7  │        14.42 / 14.46 ±0.03 / 14.51 ms │        14.52 / 14.65 ±0.07 / 14.75 ms │     no change │
│ QQuery 8  │     321.40 / 327.76 ±3.97 / 333.55 ms │     322.82 / 326.39 ±2.10 / 328.17 ms │     no change │
│ QQuery 9  │     453.55 / 460.03 ±5.96 / 467.49 ms │     453.85 / 461.66 ±7.59 / 474.75 ms │     no change │
│ QQuery 10 │        69.82 / 70.69 ±0.83 / 71.72 ms │        69.75 / 70.67 ±0.77 / 71.73 ms │     no change │
│ QQuery 11 │        81.47 / 82.19 ±0.46 / 82.87 ms │        82.02 / 83.11 ±0.96 / 84.81 ms │     no change │
│ QQuery 12 │     263.56 / 268.19 ±4.32 / 275.12 ms │     268.01 / 271.93 ±4.10 / 278.53 ms │     no change │
│ QQuery 13 │     362.10 / 368.88 ±5.26 / 375.62 ms │    372.95 / 386.95 ±11.96 / 408.80 ms │     no change │
│ QQuery 14 │     282.92 / 287.59 ±5.53 / 298.02 ms │     283.73 / 285.07 ±0.98 / 285.99 ms │     no change │
│ QQuery 15 │     265.14 / 277.39 ±7.05 / 285.64 ms │     267.25 / 281.30 ±9.69 / 297.18 ms │     no change │
│ QQuery 16 │     611.43 / 619.09 ±5.86 / 628.07 ms │     608.25 / 619.72 ±6.53 / 626.43 ms │     no change │
│ QQuery 17 │     613.93 / 622.20 ±6.59 / 629.33 ms │     619.03 / 626.82 ±5.18 / 632.30 ms │     no change │
│ QQuery 18 │ 1249.80 / 1266.69 ±15.90 / 1291.56 ms │ 1240.61 / 1274.58 ±17.90 / 1292.19 ms │     no change │
│ QQuery 19 │        28.62 / 31.88 ±5.93 / 43.73 ms │       28.75 / 39.45 ±15.90 / 69.82 ms │  1.24x slower │
│ QQuery 20 │     517.08 / 520.67 ±6.20 / 533.03 ms │     520.44 / 529.72 ±5.13 / 535.01 ms │     no change │
│ QQuery 21 │     514.37 / 522.37 ±6.25 / 533.09 ms │     512.62 / 518.56 ±3.54 / 522.54 ms │     no change │
│ QQuery 22 │   981.94 / 993.03 ±11.08 / 1014.28 ms │    988.46 / 996.86 ±5.92 / 1005.34 ms │     no change │
│ QQuery 23 │ 3010.96 / 3066.67 ±29.29 / 3094.84 ms │ 3021.81 / 3072.66 ±31.96 / 3106.05 ms │     no change │
│ QQuery 24 │        41.33 / 41.99 ±0.67 / 43.14 ms │       42.48 / 50.70 ±14.37 / 79.36 ms │  1.21x slower │
│ QQuery 25 │     112.35 / 114.13 ±2.01 / 117.95 ms │     111.86 / 115.52 ±6.22 / 127.91 ms │     no change │
│ QQuery 26 │        41.95 / 44.28 ±3.95 / 52.16 ms │        41.94 / 46.14 ±7.81 / 61.76 ms │     no change │
│ QQuery 27 │     510.07 / 516.23 ±3.23 / 518.79 ms │     511.59 / 526.15 ±7.91 / 533.40 ms │     no change │
│ QQuery 28 │ 2866.58 / 2909.48 ±23.60 / 2934.20 ms │ 2875.78 / 2920.68 ±24.27 / 2945.53 ms │     no change │
│ QQuery 29 │       41.53 / 51.95 ±20.14 / 92.23 ms │       41.44 / 51.71 ±14.25 / 78.41 ms │     no change │
│ QQuery 30 │     309.44 / 312.45 ±2.27 / 315.80 ms │     299.90 / 310.35 ±6.39 / 319.43 ms │     no change │
│ QQuery 31 │     284.43 / 288.85 ±2.62 / 292.46 ms │     287.33 / 295.02 ±6.01 / 302.91 ms │     no change │
│ QQuery 32 │    925.78 / 966.34 ±23.17 / 997.17 ms │    924.24 / 954.56 ±19.49 / 975.43 ms │     no change │
│ QQuery 33 │ 1436.40 / 1465.37 ±19.05 / 1488.07 ms │ 1446.18 / 1483.20 ±23.03 / 1513.43 ms │     no change │
│ QQuery 34 │ 1464.85 / 1495.14 ±20.14 / 1524.96 ms │ 1451.97 / 1480.90 ±19.08 / 1503.03 ms │     no change │
│ QQuery 35 │    280.49 / 314.00 ±40.01 / 370.05 ms │    279.46 / 317.30 ±36.35 / 378.68 ms │     no change │
│ QQuery 36 │        66.88 / 74.66 ±6.56 / 83.52 ms │        65.59 / 72.63 ±6.10 / 80.13 ms │     no change │
│ QQuery 37 │        35.42 / 37.93 ±3.48 / 44.76 ms │        36.02 / 37.64 ±1.67 / 40.70 ms │     no change │
│ QQuery 38 │        40.08 / 44.69 ±3.96 / 49.53 ms │        43.06 / 45.63 ±2.05 / 48.46 ms │     no change │
│ QQuery 39 │     148.86 / 153.47 ±2.72 / 156.24 ms │     136.01 / 149.49 ±7.73 / 157.19 ms │     no change │
│ QQuery 40 │        14.30 / 15.95 ±2.18 / 20.22 ms │        14.65 / 16.29 ±1.80 / 19.70 ms │     no change │
│ QQuery 41 │        14.04 / 14.35 ±0.26 / 14.75 ms │        14.30 / 15.89 ±1.52 / 18.43 ms │  1.11x slower │
│ QQuery 42 │        13.42 / 15.55 ±3.84 / 23.23 ms │        13.40 / 13.80 ±0.33 / 14.25 ms │ +1.13x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 19266.13ms │
│ Total Time (refactor-materializing-smj)   │ 19350.36ms │
│ Average Time (HEAD)                       │   448.05ms │
│ Average Time (refactor-materializing-smj) │   450.01ms │
│ Queries Faster                            │          1 │
│ Queries Slower                            │          3 │
│ Queries with No Change                    │         39 │
│ Queries with Failure                      │          0 │
└───────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	100.0s
Peak memory	11.9 GiB
Avg memory	4.5 GiB
CPU user	993.6s
CPU sys	66.6s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	100.0s
Peak memory	12.2 GiB
Avg memory	4.4 GiB
CPU user	991.2s
CPU sys	68.7s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-30T15:53:26Z

run benchmark smj

adriangbot · 2026-07-30T15:56:13Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5133171057-1305-7zf7z 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 refactor-materializing-smj (5bab8f5) to 68d5874 (merge-base) diff using: smj
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T16:06:17Z

🤖 Benchmark completed (GKE) | trigger

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

CPU Details (lscpu)

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

Details

Comparing HEAD and refactor-materializing-smj
--------------------
Benchmark smj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃           refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           8.48 / 8.82 ±0.24 / 9.15 ms │          8.36 / 8.49 ±0.15 / 8.78 ms │     no change │
│ QQuery 2  │     162.13 / 163.05 ±0.82 / 164.30 ms │    160.51 / 162.01 ±1.20 / 163.95 ms │     no change │
│ QQuery 3  │     103.43 / 104.76 ±1.14 / 106.87 ms │   101.78 / 115.74 ±14.44 / 133.85 ms │  1.10x slower │
│ QQuery 4  │        28.31 / 28.51 ±0.11 / 28.64 ms │       28.07 / 28.23 ±0.09 / 28.32 ms │     no change │
│ QQuery 5  │        20.70 / 21.00 ±0.20 / 21.24 ms │       20.31 / 20.64 ±0.17 / 20.74 ms │     no change │
│ QQuery 6  │     158.78 / 160.40 ±2.04 / 164.34 ms │    157.78 / 160.52 ±3.11 / 166.48 ms │     no change │
│ QQuery 7  │     199.92 / 201.71 ±0.98 / 202.57 ms │    195.61 / 198.03 ±1.75 / 200.62 ms │     no change │
│ QQuery 8  │        18.80 / 19.16 ±0.22 / 19.48 ms │       18.59 / 18.90 ±0.21 / 19.21 ms │     no change │
│ QQuery 9  │     206.47 / 208.02 ±0.91 / 209.22 ms │   204.07 / 210.90 ±10.89 / 232.52 ms │     no change │
│ QQuery 10 │        57.96 / 58.12 ±0.09 / 58.22 ms │       57.94 / 58.12 ±0.10 / 58.25 ms │     no change │
│ QQuery 11 │        27.44 / 29.96 ±4.98 / 39.91 ms │       27.25 / 31.52 ±8.25 / 48.02 ms │  1.05x slower │
│ QQuery 12 │        65.20 / 70.15 ±2.75 / 73.23 ms │       63.17 / 66.37 ±2.60 / 68.90 ms │ +1.06x faster │
│ QQuery 13 │      96.78 / 106.52 ±5.29 / 112.18 ms │    104.60 / 106.92 ±1.55 / 108.47 ms │     no change │
│ QQuery 14 │        57.11 / 59.63 ±2.32 / 63.98 ms │       58.29 / 62.13 ±5.96 / 73.86 ms │     no change │
│ QQuery 15 │        57.96 / 58.40 ±0.27 / 58.69 ms │       58.04 / 58.28 ±0.28 / 58.78 ms │     no change │
│ QQuery 16 │        11.53 / 11.66 ±0.10 / 11.85 ms │       11.44 / 11.60 ±0.13 / 11.78 ms │     no change │
│ QQuery 17 │     146.60 / 148.89 ±3.33 / 155.46 ms │    145.32 / 147.83 ±3.50 / 154.76 ms │     no change │
│ QQuery 18 │     111.70 / 112.16 ±0.29 / 112.51 ms │    111.86 / 112.36 ±0.31 / 112.78 ms │     no change │
│ QQuery 19 │     290.45 / 298.34 ±6.59 / 305.40 ms │    289.52 / 297.47 ±7.10 / 309.21 ms │     no change │
│ QQuery 20 │ 1259.66 / 1269.68 ±16.25 / 1302.08 ms │ 1240.74 / 1247.42 ±4.40 / 1252.58 ms │     no change │
│ QQuery 21 │       95.16 / 99.36 ±6.11 / 111.38 ms │      90.75 / 95.16 ±7.21 / 109.54 ms │     no change │
│ QQuery 22 │     102.37 / 103.64 ±0.83 / 104.87 ms │      97.28 / 98.83 ±1.05 / 100.39 ms │     no change │
│ QQuery 23 │     109.77 / 110.93 ±1.07 / 112.77 ms │    102.75 / 103.19 ±0.54 / 104.26 ms │ +1.08x faster │
│ QQuery 24 │        27.37 / 27.45 ±0.08 / 27.58 ms │       27.44 / 27.60 ±0.13 / 27.84 ms │     no change │
│ QQuery 25 │        65.24 / 70.88 ±4.78 / 79.12 ms │       68.39 / 72.37 ±3.44 / 78.69 ms │     no change │
│ QQuery 26 │     103.95 / 107.66 ±2.70 / 111.97 ms │     96.81 / 107.37 ±6.04 / 112.68 ms │     no change │
└───────────┴───────────────────────────────────────┴──────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 3658.87ms │
│ Total Time (refactor-materializing-smj)   │ 3628.01ms │
│ Average Time (HEAD)                       │  140.73ms │
│ Average Time (refactor-materializing-smj) │  139.54ms │
│ Queries Faster                            │         2 │
│ Queries Slower                            │         2 │
│ Queries with No Change                    │        22 │
│ Queries with Failure                      │         0 │
└───────────────────────────────────────────┴───────────┘

Resource Usage

smj — base (merge-base)

Metric	Value
Wall time	20.0s
Peak memory	1.2 GiB
Avg memory	462.6 MiB
CPU user	157.3s
CPU sys	3.3s
Peak spill	0 B

smj — branch

Metric	Value
Wall time	20.0s
Peak memory	1.2 GiB
Avg memory	468.8 MiB
CPU user	155.7s
CPU sys	3.6s
Peak spill	0 B

File an issue against this benchmark runner

rluvaton · 2026-07-30T16:25:29Z

run benchmark smj

adriangbot · 2026-07-30T16:28:16Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5133512422-1306-lhxv7 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 refactor-materializing-smj (cb881c7) to 68d5874 (merge-base) diff using: smj
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T16:42:39Z

🤖 Benchmark completed (GKE) | trigger

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

CPU Details (lscpu)

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

Details

Comparing HEAD and refactor-materializing-smj
--------------------
Benchmark smj.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                 HEAD ┃           refactor-materializing-smj ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │          8.37 / 8.78 ±0.27 / 9.19 ms │          8.65 / 8.81 ±0.22 / 9.24 ms │     no change │
│ QQuery 2  │    162.12 / 164.55 ±1.93 / 167.19 ms │    158.47 / 160.95 ±2.06 / 164.32 ms │     no change │
│ QQuery 3  │    102.55 / 105.30 ±1.44 / 106.30 ms │    103.17 / 104.74 ±1.03 / 106.24 ms │     no change │
│ QQuery 4  │       28.08 / 28.35 ±0.18 / 28.66 ms │       28.20 / 28.52 ±0.19 / 28.76 ms │     no change │
│ QQuery 5  │       20.75 / 21.12 ±0.21 / 21.35 ms │       20.41 / 20.59 ±0.17 / 20.91 ms │     no change │
│ QQuery 6  │    159.14 / 164.41 ±5.76 / 175.18 ms │    156.46 / 157.60 ±1.01 / 159.42 ms │     no change │
│ QQuery 7  │    198.07 / 200.74 ±1.84 / 202.48 ms │    195.58 / 198.12 ±1.81 / 200.44 ms │     no change │
│ QQuery 8  │       18.96 / 19.17 ±0.16 / 19.34 ms │       18.43 / 18.67 ±0.16 / 18.86 ms │     no change │
│ QQuery 9  │   206.62 / 213.19 ±11.50 / 236.12 ms │    202.32 / 203.59 ±1.50 / 206.44 ms │     no change │
│ QQuery 10 │       57.75 / 57.96 ±0.25 / 58.42 ms │       58.10 / 58.42 ±0.33 / 59.05 ms │     no change │
│ QQuery 11 │       27.13 / 31.70 ±9.02 / 49.74 ms │       27.31 / 27.40 ±0.06 / 27.50 ms │ +1.16x faster │
│ QQuery 12 │       64.81 / 70.00 ±3.29 / 74.21 ms │       66.48 / 70.23 ±2.26 / 73.50 ms │     no change │
│ QQuery 13 │     95.63 / 104.60 ±6.78 / 113.43 ms │    103.15 / 104.56 ±1.01 / 106.20 ms │     no change │
│ QQuery 14 │       58.25 / 61.93 ±6.05 / 73.94 ms │       58.12 / 61.33 ±4.95 / 71.11 ms │     no change │
│ QQuery 15 │       57.86 / 58.54 ±0.55 / 59.26 ms │       58.48 / 58.65 ±0.11 / 58.76 ms │     no change │
│ QQuery 16 │       11.50 / 11.63 ±0.10 / 11.79 ms │       11.45 / 11.65 ±0.11 / 11.76 ms │     no change │
│ QQuery 17 │    146.42 / 151.27 ±8.59 / 168.43 ms │    146.00 / 149.50 ±6.23 / 161.94 ms │     no change │
│ QQuery 18 │    111.50 / 111.70 ±0.19 / 112.02 ms │    112.37 / 113.11 ±0.59 / 113.75 ms │     no change │
│ QQuery 19 │    288.35 / 296.50 ±8.25 / 311.51 ms │    292.84 / 309.56 ±9.65 / 318.62 ms │     no change │
│ QQuery 20 │ 1258.96 / 1266.00 ±7.98 / 1281.22 ms │ 1240.29 / 1247.36 ±6.44 / 1255.92 ms │     no change │
│ QQuery 21 │       94.92 / 96.10 ±1.37 / 98.56 ms │       90.92 / 91.80 ±0.58 / 92.54 ms │     no change │
│ QQuery 22 │     98.82 / 102.74 ±2.24 / 105.39 ms │     98.09 / 100.39 ±1.47 / 101.69 ms │     no change │
│ QQuery 23 │    107.14 / 110.51 ±2.08 / 113.63 ms │    103.01 / 105.08 ±1.21 / 106.70 ms │     no change │
│ QQuery 24 │       27.33 / 27.49 ±0.08 / 27.57 ms │       27.78 / 27.97 ±0.16 / 28.15 ms │     no change │
│ QQuery 25 │       63.53 / 67.59 ±3.36 / 71.09 ms │       69.38 / 72.32 ±4.20 / 80.64 ms │  1.07x slower │
│ QQuery 26 │     96.58 / 108.68 ±7.70 / 119.06 ms │    94.61 / 112.70 ±17.19 / 145.44 ms │     no change │
└───────────┴──────────────────────────────────────┴──────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 3660.55ms │
│ Total Time (refactor-materializing-smj)   │ 3623.60ms │
│ Average Time (HEAD)                       │  140.79ms │
│ Average Time (refactor-materializing-smj) │  139.37ms │
│ Queries Faster                            │         1 │
│ Queries Slower                            │         1 │
│ Queries with No Change                    │        24 │
│ Queries with Failure                      │         0 │
└───────────────────────────────────────────┴───────────┘

Resource Usage

smj — base (merge-base)

Metric	Value
Wall time	20.0s
Peak memory	1.1 GiB
Avg memory	449.4 MiB
CPU user	157.0s
CPU sys	3.6s
Peak spill	0 B

smj — branch

Metric	Value
Wall time	20.0s
Peak memory	1.1 GiB
Avg memory	450.5 MiB
CPU user	155.9s
CPU sys	3.6s
Peak spill	0 B

File an issue against this benchmark runner

mbutrovich · 2026-07-30T18:06:16Z

run benchmark tpch10

env:
   PREFER_HASH_JOIN: false

adriangbot · 2026-07-30T18:08:24Z

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c5134520421-1307-v5r7k 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 refactor-materializing-smj (cb881c7) to 68d5874 (merge-base) diff using: tpch10
Results will be posted here when complete

File an issue against this benchmark runner

adriangbot · 2026-07-30T18:19:22Z

🤖 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 refactor-materializing-smj
--------------------
Benchmark tpch_sf10.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃                               HEAD ┃         refactor-materializing-smj ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │  315.56 / 317.48 ±1.64 / 320.36 ms │  316.61 / 317.92 ±1.96 / 321.81 ms │ no change │
│ QQuery 2  │  103.40 / 106.34 ±2.32 / 110.41 ms │  105.55 / 107.13 ±2.11 / 111.23 ms │ no change │
│ QQuery 3  │  236.21 / 239.54 ±2.93 / 243.74 ms │  231.76 / 238.24 ±4.40 / 243.78 ms │ no change │
│ QQuery 4  │  114.08 / 115.93 ±1.29 / 117.93 ms │  113.54 / 115.91 ±1.72 / 118.57 ms │ no change │
│ QQuery 5  │ 353.13 / 370.64 ±10.89 / 382.47 ms │  359.34 / 362.17 ±2.41 / 365.91 ms │ no change │
│ QQuery 6  │  128.88 / 131.80 ±4.03 / 139.74 ms │  129.12 / 131.88 ±3.47 / 138.68 ms │ no change │
│ QQuery 7  │  466.09 / 474.65 ±5.35 / 480.92 ms │  473.74 / 484.60 ±8.25 / 494.95 ms │ no change │
│ QQuery 8  │  387.93 / 397.03 ±5.63 / 404.50 ms │  387.85 / 393.57 ±4.33 / 399.15 ms │ no change │
│ QQuery 9  │ 563.23 / 572.05 ±12.20 / 596.14 ms │  557.78 / 568.11 ±6.47 / 577.13 ms │ no change │
│ QQuery 10 │  296.74 / 306.42 ±5.48 / 312.21 ms │  296.33 / 305.54 ±7.17 / 318.22 ms │ no change │
│ QQuery 11 │    82.81 / 92.37 ±8.52 / 107.87 ms │     86.40 / 88.67 ±2.75 / 93.93 ms │ no change │
│ QQuery 12 │ 178.46 / 186.48 ±10.83 / 207.88 ms │  182.03 / 189.62 ±9.42 / 207.74 ms │ no change │
│ QQuery 13 │ 289.23 / 304.24 ±18.99 / 339.49 ms │ 293.91 / 306.49 ±11.98 / 329.18 ms │ no change │
│ QQuery 14 │  173.23 / 177.62 ±5.11 / 187.36 ms │  173.33 / 177.85 ±3.87 / 182.78 ms │ no change │
│ QQuery 15 │  307.13 / 311.03 ±2.41 / 314.45 ms │  308.33 / 310.47 ±1.70 / 312.60 ms │ no change │
│ QQuery 16 │     65.50 / 67.79 ±2.02 / 71.58 ms │     64.02 / 67.72 ±2.35 / 70.73 ms │ no change │
│ QQuery 17 │ 650.14 / 662.01 ±10.13 / 678.05 ms │  640.47 / 652.84 ±9.54 / 663.01 ms │ no change │
│ QQuery 18 │ 707.26 / 724.25 ±20.75 / 760.77 ms │ 699.20 / 717.38 ±19.23 / 752.03 ms │ no change │
│ QQuery 19 │ 247.08 / 262.60 ±13.93 / 283.77 ms │ 249.84 / 263.01 ±14.09 / 285.24 ms │ no change │
│ QQuery 20 │  287.60 / 295.91 ±4.68 / 300.20 ms │ 273.91 / 292.31 ±12.44 / 311.82 ms │ no change │
│ QQuery 21 │  668.39 / 673.81 ±4.02 / 680.35 ms │  665.72 / 677.55 ±7.75 / 685.02 ms │ no change │
│ QQuery 22 │     59.94 / 63.87 ±3.99 / 71.32 ms │     61.61 / 64.63 ±2.75 / 68.03 ms │ no change │
└───────────┴────────────────────────────────────┴────────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                         ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                         │ 6853.86ms │
│ Total Time (refactor-materializing-smj)   │ 6833.63ms │
│ Average Time (HEAD)                       │  311.54ms │
│ Average Time (refactor-materializing-smj) │  310.62ms │
│ Queries Faster                            │         0 │
│ Queries Slower                            │         0 │
│ Queries with No Change                    │        22 │
│ Queries with Failure                      │         0 │
└───────────────────────────────────────────┴───────────┘

Resource Usage

tpch10 — base (merge-base)

Metric	Value
Wall time	35.0s
Peak memory	5.0 GiB
Avg memory	1.5 GiB
CPU user	344.8s
CPU sys	19.6s
Peak spill	0 B

tpch10 — branch

Metric	Value
Wall time	35.0s
Peak memory	5.1 GiB
Avg memory	1.6 GiB
CPU user	345.8s
CPU sys	19.8s
Peak spill	0 B

File an issue against this benchmark runner

mbutrovich

Tracks with what I saw in #23761. Tests are green, added tests look good, and performance looks the same. Thanks @rluvaton!

mbutrovich · 2026-07-30T21:35:20Z


-                match ready!(self.spill_stream.as_mut().unwrap().poll_next_unpin(cx)) {
-                    Some(Ok(batch)) => {
+                match spill_stream.next().await.transpose()? {


restore_spilled_batches awaits the spill stream here without stop_join_time()/start_join_time() around it, per the stop_join_time doc comment: "The join's own spill write and read-back are NOT excluded — that time is join work." That's a real change from the old poll_next, which returned out via ready!(self.poll_spilled_batches(cx, ...)) on a genuine Poll::Pending, stopping _timer for that wait.

This is the same join_time semantics change #23761 required calling out before merge (that PR's description was updated to note the metric now includes spill read-back time). This PR's description currently says "Are there any user-facing changes? nope" — worth adding the same note here, since join_time is visible in EXPLAIN ANALYZE.

rluvaton added 4 commits July 29, 2026 14:45

chore: refactor MaterializingSortMergeJoinStream into generators

000b8a7

remove duplicate functions

88e2471

extract on exhausted

de03b85

simplify

ccdd363

github-actions Bot added the physical-plan Changes to the physical-plan crate label Jul 29, 2026

rluvaton mentioned this pull request Jul 30, 2026

[Status Update] Simplifing Streams to be more textbook-like and have less state while keeping the same perf #23974

Open

13 tasks

rluvaton added 2 commits July 30, 2026 17:51

simplify to be text book like

fc208f5

comments

5bab8f5

update comments and add timing

2f132c4

rluvaton marked this pull request as ready for review July 30, 2026 16:25

rluvaton requested a review from mbutrovich July 30, 2026 16:25

rluvaton changed the title ~~chore: refactor MaterializingSortMergeJoinStream into generators~~ chore: refactor MaterializingSortMergeJoinStream into generators and simplify code to be textbook like as possible Jul 30, 2026

update comments

cb881c7

mbutrovich approved these changes Jul 30, 2026

View reviewed changes

rluvaton added this pull request to the merge queue Jul 30, 2026

Merged via the queue into apache:main with commit 39d5064 Jul 30, 2026
42 checks passed

rluvaton deleted the refactor-materializing-smj branch July 30, 2026 22:10

Uh oh!

Conversation

rluvaton commented Jul 29, 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!

rluvaton commented Jul 29, 2026

Uh oh!

rluvaton commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

adriangbot commented Jul 29, 2026

Uh oh!

codecov-commenter commented Jul 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

rluvaton commented Jul 30, 2026

Uh oh!

rluvaton commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

rluvaton commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

rluvaton commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

mbutrovich commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

adriangbot commented Jul 30, 2026

Uh oh!

mbutrovich left a comment

Choose a reason for hiding this comment

Uh oh!

mbutrovich Jul 30, 2026

Choose a reason for hiding this comment

Uh oh!

rluvaton Jul 30, 2026

Choose a reason for hiding this comment

rluvaton commented Jul 29, 2026 •

edited

Loading

codecov-commenter commented Jul 29, 2026 •

edited

Loading