Lazy register parquet file metrics

[xudong963]

Which issue does this PR close?

• Closes Make parquet metrics use lazy-registration pattern #22189.

Rationale for this change

Parquet scans currently register a full set of per-file metrics when each
ParquetFileMetrics is created, even when most of those metrics remain zero.
This is costly for scans over many parquet files because metric registration
clones labels such as filename and appends many unused metrics to the plan
metrics set.

What changes are included in this PR?

• Create parquet file metric handles without immediately registering every
  metric.
• Add a lazy registration guard that registers only non-zero parquet file
  metrics when the file metrics are dropped.
• Keep the registration guard alive for push decoder streams so cloned metric
  handles can be updated throughout streaming before registration occurs.
• Preserve the existing lazy behavior for
  page_index_pages_skipped_by_fully_matched.
• Update parquet tests to allow zero-valued metrics to be absent while still
  requiring metrics that are expected to have non-zero values.
• Add focused unit coverage for lazy registration of Count, Gauge, Time,
  PruningMetrics, and RatioMetrics.

Are these changes tested?

Yes.

cargo fmt --all
cargo clippy --all-targets --all-features -- -D warnings
cargo test -p datafusion-datasource-parquet --lib
cargo test -p datafusion --test parquet_integration

I also ran a local release microbenchmark for repeated ParquetFileMetrics
creation against origin/main; for 50k file metrics, zero-metric registration
improved from about 57.9ms to 25.2ms, and the bytes_scanned case improved
from about 57.7ms to 28.4ms.

Are there any user-facing changes?

There are no SQL result or schema changes.

The observable metrics output changes: parquet file metrics that remain zero
may now be absent from the metrics set instead of being present with value 0.
Low-level integrations should treat missing parquet file metrics as zero when
appropriate.

ParquetFileMetrics is a low-level public struct that is documented as subject
to change. This PR adds an internal registration guard field, so external Rust
code should construct it via ParquetFileMetrics::new.

[xudong963]

run benchmarks

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4485097144-204-c2mvf 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 xudong963/issue-22189 (eb2ee9b) to c8b784a (merge-base) diff using: clickbench_partitioned
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4485097144-205-nkv27 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 xudong963/issue-22189 (eb2ee9b) to c8b784a (merge-base) diff using: tpcds
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4485097144-206-hqswn 6.12.68+ #1 SMP Wed Apr 1 02:23:28 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 xudong963/issue-22189 (eb2ee9b) to c8b784a (merge-base) diff using: tpch
Results will be posted here when complete

---

File an issue against this benchmark runner

[adriangbot]

🤖 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 xudong963_issue-22189
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃          xudong963_issue-22189 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.43 / 39.44 ±1.02 / 41.26 ms │ 37.97 / 39.27 ±1.43 / 41.87 ms │     no change │
│ QQuery 2  │ 20.38 / 20.81 ±0.42 / 21.55 ms │ 19.55 / 19.68 ±0.17 / 20.01 ms │ +1.06x faster │
│ QQuery 3  │ 32.98 / 34.17 ±1.32 / 36.27 ms │ 32.41 / 33.40 ±1.18 / 34.86 ms │     no change │
│ QQuery 4  │ 17.41 / 17.69 ±0.39 / 18.44 ms │ 16.58 / 17.56 ±0.82 / 18.49 ms │     no change │
│ QQuery 5  │ 41.90 / 44.14 ±1.80 / 47.24 ms │ 39.73 / 41.55 ±1.22 / 43.44 ms │ +1.06x faster │
│ QQuery 6  │ 16.55 / 17.13 ±0.92 / 18.96 ms │ 16.00 / 16.34 ±0.55 / 17.43 ms │     no change │
│ QQuery 7  │ 45.51 / 47.71 ±1.49 / 49.89 ms │ 47.32 / 48.25 ±0.75 / 49.25 ms │     no change │
│ QQuery 8  │ 44.97 / 45.27 ±0.18 / 45.48 ms │ 43.99 / 44.16 ±0.26 / 44.67 ms │     no change │
│ QQuery 9  │ 49.30 / 50.32 ±0.64 / 50.97 ms │ 48.33 / 48.93 ±0.51 / 49.56 ms │     no change │
│ QQuery 10 │ 63.37 / 64.60 ±1.19 / 66.23 ms │ 62.37 / 62.86 ±0.31 / 63.34 ms │     no change │
│ QQuery 11 │ 13.37 / 13.60 ±0.20 / 13.86 ms │ 12.73 / 12.81 ±0.07 / 12.91 ms │ +1.06x faster │
│ QQuery 12 │ 24.93 / 25.60 ±0.48 / 26.42 ms │ 23.91 / 24.83 ±0.97 / 26.13 ms │     no change │
│ QQuery 13 │ 33.58 / 35.09 ±1.60 / 37.87 ms │ 33.63 / 34.55 ±0.58 / 35.29 ms │     no change │
│ QQuery 14 │ 25.43 / 25.67 ±0.19 / 25.95 ms │ 24.98 / 25.20 ±0.25 / 25.69 ms │     no change │
│ QQuery 15 │ 31.33 / 31.86 ±0.63 / 33.10 ms │ 30.51 / 30.80 ±0.26 / 31.26 ms │     no change │
│ QQuery 16 │ 15.19 / 15.29 ±0.08 / 15.42 ms │ 14.56 / 14.67 ±0.10 / 14.83 ms │     no change │
│ QQuery 17 │ 74.48 / 75.85 ±1.58 / 78.84 ms │ 73.54 / 74.35 ±0.54 / 75.01 ms │     no change │
│ QQuery 18 │ 62.24 / 63.67 ±1.10 / 65.02 ms │ 60.03 / 62.12 ±1.30 / 63.70 ms │     no change │
│ QQuery 19 │ 35.14 / 36.65 ±1.23 / 37.93 ms │ 34.53 / 35.50 ±1.23 / 37.89 ms │     no change │
│ QQuery 20 │ 37.41 / 38.33 ±0.67 / 39.47 ms │ 37.08 / 37.48 ±0.59 / 38.64 ms │     no change │
│ QQuery 21 │ 55.98 / 58.08 ±1.51 / 60.30 ms │ 53.97 / 55.77 ±1.39 / 58.00 ms │     no change │
│ QQuery 22 │ 23.63 / 24.17 ±0.37 / 24.75 ms │ 22.65 / 23.81 ±1.82 / 27.44 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                    ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                    │ 825.14ms │
│ Total Time (xudong963_issue-22189)   │ 803.88ms │
│ Average Time (HEAD)                  │  37.51ms │
│ Average Time (xudong963_issue-22189) │  36.54ms │
│ Queries Faster                       │        3 │
│ Queries Slower                       │        0 │
│ Queries with No Change               │       19 │
│ Queries with Failure                 │        0 │
└──────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric	Value
Wall time	5.0s
Peak memory	5.5 GiB
Avg memory	5.0 GiB
CPU user	29.5s
CPU sys	2.3s
Peak spill	0 B

tpch — branch

Metric	Value
Wall time	5.0s
Peak memory	5.6 GiB
Avg memory	5.0 GiB
CPU user	29.3s
CPU sys	2.1s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 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 xudong963_issue-22189
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                 xudong963_issue-22189 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           6.41 / 6.89 ±0.89 / 8.68 ms │           6.24 / 6.82 ±0.92 / 8.65 ms │     no change │
│ QQuery 2  │        81.47 / 81.85 ±0.21 / 82.09 ms │        81.14 / 81.58 ±0.34 / 82.09 ms │     no change │
│ QQuery 3  │        29.39 / 29.62 ±0.20 / 29.90 ms │        29.12 / 29.45 ±0.16 / 29.55 ms │     no change │
│ QQuery 4  │     521.29 / 523.04 ±1.72 / 525.12 ms │     518.30 / 523.87 ±4.15 / 528.35 ms │     no change │
│ QQuery 5  │        53.46 / 53.85 ±0.29 / 54.33 ms │        51.50 / 52.08 ±0.30 / 52.35 ms │     no change │
│ QQuery 6  │        36.05 / 36.24 ±0.17 / 36.52 ms │        35.48 / 35.95 ±0.48 / 36.84 ms │     no change │
│ QQuery 7  │     109.45 / 111.98 ±2.17 / 115.54 ms │     107.64 / 108.54 ±1.10 / 110.65 ms │     no change │
│ QQuery 8  │        39.86 / 40.07 ±0.13 / 40.22 ms │        39.33 / 39.53 ±0.14 / 39.68 ms │     no change │
│ QQuery 9  │        53.34 / 55.58 ±1.58 / 57.97 ms │        52.92 / 55.03 ±1.66 / 57.37 ms │     no change │
│ QQuery 10 │        82.36 / 83.15 ±1.40 / 85.94 ms │        80.65 / 81.40 ±1.07 / 83.52 ms │     no change │
│ QQuery 11 │     318.37 / 325.00 ±5.01 / 332.17 ms │     320.75 / 326.26 ±3.16 / 329.38 ms │     no change │
│ QQuery 12 │        28.93 / 29.39 ±0.37 / 29.82 ms │        28.96 / 29.33 ±0.40 / 30.09 ms │     no change │
│ QQuery 13 │     129.23 / 130.20 ±1.00 / 131.93 ms │     128.26 / 129.36 ±0.71 / 130.40 ms │     no change │
│ QQuery 14 │     508.00 / 510.66 ±2.56 / 514.06 ms │     503.77 / 506.78 ±2.65 / 510.96 ms │     no change │
│ QQuery 15 │        61.58 / 62.05 ±0.34 / 62.59 ms │        60.96 / 62.35 ±1.12 / 64.31 ms │     no change │
│ QQuery 16 │           6.88 / 7.19 ±0.37 / 7.91 ms │           7.17 / 7.34 ±0.20 / 7.68 ms │     no change │
│ QQuery 17 │        81.00 / 82.09 ±0.61 / 82.83 ms │        81.04 / 81.56 ±0.63 / 82.69 ms │     no change │
│ QQuery 18 │     152.47 / 153.31 ±0.72 / 154.28 ms │     150.80 / 152.13 ±1.04 / 153.89 ms │     no change │
│ QQuery 19 │        41.85 / 41.98 ±0.18 / 42.33 ms │        41.73 / 41.85 ±0.12 / 42.07 ms │     no change │
│ QQuery 20 │        35.68 / 36.17 ±0.38 / 36.80 ms │        35.28 / 35.63 ±0.36 / 36.12 ms │     no change │
│ QQuery 21 │        17.78 / 18.20 ±0.30 / 18.58 ms │        17.99 / 18.37 ±0.21 / 18.62 ms │     no change │
│ QQuery 22 │        62.58 / 63.04 ±0.35 / 63.60 ms │        61.76 / 62.01 ±0.24 / 62.37 ms │     no change │
│ QQuery 23 │     480.86 / 484.33 ±2.83 / 489.47 ms │     480.95 / 482.54 ±1.14 / 484.46 ms │     no change │
│ QQuery 24 │     233.85 / 236.01 ±1.56 / 238.59 ms │     233.12 / 237.14 ±3.64 / 243.63 ms │     no change │
│ QQuery 25 │     114.64 / 117.05 ±2.20 / 120.64 ms │     113.31 / 116.12 ±3.21 / 122.40 ms │     no change │
│ QQuery 26 │        71.88 / 72.15 ±0.28 / 72.66 ms │        70.12 / 70.45 ±0.25 / 70.84 ms │     no change │
│ QQuery 27 │           7.12 / 7.32 ±0.16 / 7.59 ms │           7.17 / 7.32 ±0.20 / 7.71 ms │     no change │
│ QQuery 28 │        58.11 / 62.10 ±2.64 / 64.42 ms │        61.66 / 62.58 ±0.59 / 63.29 ms │     no change │
│ QQuery 29 │       99.10 / 99.81 ±0.59 / 100.69 ms │      98.74 / 100.85 ±1.99 / 104.47 ms │     no change │
│ QQuery 30 │        31.32 / 31.52 ±0.22 / 31.86 ms │        30.70 / 31.00 ±0.39 / 31.71 ms │     no change │
│ QQuery 31 │     113.53 / 115.40 ±2.30 / 119.80 ms │     112.09 / 113.73 ±1.40 / 116.15 ms │     no change │
│ QQuery 32 │        20.77 / 20.99 ±0.20 / 21.36 ms │        20.85 / 20.91 ±0.08 / 21.04 ms │     no change │
│ QQuery 33 │        39.22 / 39.83 ±0.40 / 40.31 ms │        38.92 / 39.19 ±0.26 / 39.59 ms │     no change │
│ QQuery 34 │         9.90 / 10.21 ±0.22 / 10.47 ms │         9.84 / 10.21 ±0.28 / 10.61 ms │     no change │
│ QQuery 35 │        81.72 / 84.23 ±1.80 / 86.62 ms │        80.16 / 81.45 ±1.03 / 82.85 ms │     no change │
│ QQuery 36 │           6.53 / 6.69 ±0.15 / 6.97 ms │           6.44 / 6.62 ±0.20 / 7.00 ms │     no change │
│ QQuery 37 │           7.30 / 7.67 ±0.31 / 8.18 ms │           7.22 / 7.44 ±0.21 / 7.72 ms │     no change │
│ QQuery 38 │        68.54 / 69.65 ±1.00 / 71.17 ms │        68.51 / 69.22 ±0.62 / 69.96 ms │     no change │
│ QQuery 39 │      99.45 / 102.63 ±3.04 / 107.34 ms │       98.73 / 99.54 ±0.48 / 100.11 ms │     no change │
│ QQuery 40 │        24.18 / 24.66 ±0.35 / 25.26 ms │        23.38 / 23.71 ±0.54 / 24.79 ms │     no change │
│ QQuery 41 │        14.56 / 14.80 ±0.20 / 15.05 ms │        14.42 / 14.60 ±0.17 / 14.84 ms │     no change │
│ QQuery 42 │        24.50 / 24.97 ±0.35 / 25.42 ms │        24.22 / 24.77 ±0.59 / 25.87 ms │     no change │
│ QQuery 43 │           5.42 / 5.55 ±0.13 / 5.79 ms │           5.46 / 5.59 ±0.13 / 5.83 ms │     no change │
│ QQuery 44 │        11.02 / 11.27 ±0.17 / 11.54 ms │        10.53 / 10.73 ±0.11 / 10.87 ms │     no change │
│ QQuery 45 │        40.16 / 40.90 ±0.72 / 42.14 ms │        41.07 / 41.57 ±0.32 / 42.09 ms │     no change │
│ QQuery 46 │        13.81 / 14.25 ±0.52 / 15.11 ms │        13.44 / 13.84 ±0.25 / 14.13 ms │     no change │
│ QQuery 47 │     233.92 / 237.29 ±3.19 / 242.69 ms │     230.09 / 233.35 ±2.76 / 238.06 ms │     no change │
│ QQuery 48 │     105.07 / 105.69 ±0.55 / 106.72 ms │     104.36 / 105.56 ±1.18 / 107.39 ms │     no change │
│ QQuery 49 │        81.46 / 82.04 ±0.58 / 83.13 ms │        80.81 / 81.43 ±0.52 / 82.08 ms │     no change │
│ QQuery 50 │        60.89 / 63.00 ±2.20 / 66.73 ms │        59.96 / 60.61 ±0.80 / 62.17 ms │     no change │
│ QQuery 51 │        92.55 / 94.51 ±1.83 / 97.46 ms │       91.57 / 95.13 ±3.68 / 101.80 ms │     no change │
│ QQuery 52 │        24.39 / 24.62 ±0.22 / 25.01 ms │        24.07 / 24.40 ±0.33 / 24.95 ms │     no change │
│ QQuery 53 │        30.67 / 31.30 ±1.04 / 33.35 ms │        30.01 / 30.23 ±0.16 / 30.46 ms │     no change │
│ QQuery 54 │        55.52 / 57.23 ±2.85 / 62.92 ms │        54.93 / 56.31 ±2.31 / 60.92 ms │     no change │
│ QQuery 55 │        24.25 / 24.58 ±0.38 / 25.08 ms │        23.89 / 24.37 ±0.38 / 24.94 ms │     no change │
│ QQuery 56 │        39.93 / 40.69 ±0.40 / 41.09 ms │        39.31 / 39.85 ±0.38 / 40.50 ms │     no change │
│ QQuery 57 │     180.07 / 182.06 ±1.12 / 183.54 ms │     177.42 / 178.98 ±0.84 / 179.73 ms │     no change │
│ QQuery 58 │     116.69 / 117.69 ±0.76 / 118.61 ms │     116.15 / 117.41 ±0.90 / 118.29 ms │     no change │
│ QQuery 59 │     118.25 / 119.62 ±0.88 / 120.97 ms │     117.85 / 119.06 ±0.96 / 120.80 ms │     no change │
│ QQuery 60 │        40.41 / 40.85 ±0.27 / 41.22 ms │        39.23 / 39.51 ±0.19 / 39.80 ms │     no change │
│ QQuery 61 │        13.98 / 14.74 ±1.28 / 17.29 ms │        13.88 / 14.02 ±0.22 / 14.45 ms │     no change │
│ QQuery 62 │        47.28 / 49.23 ±2.12 / 52.57 ms │        46.38 / 46.89 ±0.33 / 47.41 ms │     no change │
│ QQuery 63 │        31.17 / 31.60 ±0.33 / 32.01 ms │        30.57 / 30.96 ±0.36 / 31.52 ms │     no change │
│ QQuery 64 │     463.27 / 472.55 ±6.07 / 479.19 ms │     461.66 / 468.85 ±6.54 / 478.70 ms │     no change │
│ QQuery 65 │     148.01 / 149.83 ±1.63 / 152.47 ms │     147.00 / 147.65 ±0.56 / 148.60 ms │     no change │
│ QQuery 66 │        83.57 / 84.28 ±0.51 / 85.13 ms │        82.15 / 83.05 ±0.62 / 83.97 ms │     no change │
│ QQuery 67 │     252.26 / 257.36 ±3.12 / 261.02 ms │     243.24 / 248.77 ±4.71 / 254.69 ms │     no change │
│ QQuery 68 │        14.20 / 14.37 ±0.17 / 14.67 ms │        13.73 / 13.89 ±0.22 / 14.29 ms │     no change │
│ QQuery 69 │        78.33 / 78.64 ±0.32 / 79.20 ms │        76.48 / 77.92 ±2.48 / 82.86 ms │     no change │
│ QQuery 70 │     108.10 / 113.72 ±7.87 / 128.84 ms │     105.13 / 107.42 ±2.16 / 111.11 ms │ +1.06x faster │
│ QQuery 71 │        36.57 / 37.13 ±0.44 / 37.91 ms │        35.17 / 35.48 ±0.35 / 36.14 ms │     no change │
│ QQuery 72 │ 2176.23 / 2230.24 ±56.06 / 2326.35 ms │ 2093.73 / 2139.32 ±53.22 / 2237.71 ms │     no change │
│ QQuery 73 │         9.70 / 10.48 ±0.93 / 12.20 ms │          9.52 / 9.76 ±0.22 / 10.06 ms │ +1.07x faster │
│ QQuery 74 │     180.69 / 185.97 ±3.72 / 190.96 ms │     181.38 / 185.05 ±3.92 / 192.57 ms │     no change │
│ QQuery 75 │     149.40 / 151.70 ±2.82 / 156.81 ms │     147.48 / 151.13 ±3.76 / 158.19 ms │     no change │
│ QQuery 76 │        36.40 / 36.76 ±0.26 / 37.13 ms │        35.39 / 35.65 ±0.13 / 35.75 ms │     no change │
│ QQuery 77 │        62.46 / 63.14 ±0.64 / 64.24 ms │        61.32 / 61.86 ±0.51 / 62.82 ms │     no change │
│ QQuery 78 │     191.52 / 194.26 ±2.59 / 199.07 ms │     188.05 / 190.04 ±2.12 / 193.91 ms │     no change │
│ QQuery 79 │        67.95 / 69.16 ±1.19 / 71.42 ms │        67.60 / 70.60 ±3.75 / 77.56 ms │     no change │
│ QQuery 80 │     102.94 / 104.41 ±1.01 / 105.75 ms │     101.39 / 103.15 ±1.53 / 105.30 ms │     no change │
│ QQuery 81 │        25.05 / 25.37 ±0.18 / 25.58 ms │        24.94 / 25.29 ±0.24 / 25.67 ms │     no change │
│ QQuery 82 │        16.90 / 17.23 ±0.22 / 17.50 ms │        16.28 / 17.48 ±1.15 / 18.94 ms │     no change │
│ QQuery 83 │        37.89 / 38.09 ±0.18 / 38.41 ms │        37.98 / 38.69 ±1.01 / 40.66 ms │     no change │
│ QQuery 84 │        43.93 / 46.49 ±3.06 / 52.02 ms │        43.91 / 45.09 ±1.21 / 47.41 ms │     no change │
│ QQuery 85 │     138.07 / 139.56 ±1.29 / 141.65 ms │     136.77 / 137.91 ±0.78 / 139.20 ms │     no change │
│ QQuery 86 │        25.53 / 25.74 ±0.18 / 26.07 ms │        25.79 / 26.93 ±1.86 / 30.64 ms │     no change │
│ QQuery 87 │        70.42 / 73.35 ±1.91 / 76.27 ms │        69.30 / 70.49 ±1.01 / 72.11 ms │     no change │
│ QQuery 88 │        65.60 / 66.25 ±0.65 / 67.27 ms │        64.17 / 64.39 ±0.18 / 64.64 ms │     no change │
│ QQuery 89 │        36.46 / 36.71 ±0.13 / 36.82 ms │        36.64 / 37.13 ±0.25 / 37.30 ms │     no change │
│ QQuery 90 │        17.55 / 17.74 ±0.18 / 18.04 ms │        17.61 / 17.80 ±0.14 / 18.01 ms │     no change │
│ QQuery 91 │        53.18 / 54.02 ±0.88 / 55.53 ms │        52.76 / 53.10 ±0.46 / 53.98 ms │     no change │
│ QQuery 92 │        29.81 / 30.37 ±0.47 / 31.04 ms │        30.12 / 30.31 ±0.15 / 30.51 ms │     no change │
│ QQuery 93 │        51.15 / 51.50 ±0.26 / 51.95 ms │        50.64 / 51.41 ±0.42 / 51.87 ms │     no change │
│ QQuery 94 │        38.27 / 38.80 ±0.33 / 39.22 ms │        38.01 / 38.86 ±0.83 / 40.32 ms │     no change │
│ QQuery 95 │        85.63 / 86.26 ±0.70 / 87.30 ms │        84.92 / 85.69 ±0.57 / 86.67 ms │     no change │
│ QQuery 96 │        24.48 / 24.72 ±0.26 / 25.22 ms │        24.14 / 24.58 ±0.35 / 24.89 ms │     no change │
│ QQuery 97 │        46.22 / 46.58 ±0.32 / 46.99 ms │        45.69 / 46.17 ±0.29 / 46.59 ms │     no change │
│ QQuery 98 │        42.22 / 43.00 ±0.70 / 44.14 ms │        42.44 / 43.30 ±0.52 / 43.98 ms │     no change │
│ QQuery 99 │        70.46 / 71.59 ±1.08 / 73.23 ms │        70.44 / 71.42 ±1.38 / 74.17 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                    ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                    │ 10689.68ms │
│ Total Time (xudong963_issue-22189)   │ 10512.04ms │
│ Average Time (HEAD)                  │   107.98ms │
│ Average Time (xudong963_issue-22189) │   106.18ms │
│ Queries Faster                       │          2 │
│ Queries Slower                       │          0 │
│ Queries with No Change               │         97 │
│ Queries with Failure                 │          0 │
└──────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric	Value
Wall time	55.0s
Peak memory	6.9 GiB
Avg memory	6.1 GiB
CPU user	238.3s
CPU sys	6.0s
Peak spill	0 B

tpcds — branch

Metric	Value
Wall time	55.0s
Peak memory	6.9 GiB
Avg memory	6.2 GiB
CPU user	231.5s
CPU sys	5.2s
Peak spill	0 B

---

File an issue against this benchmark runner

[adriangbot]

🤖 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 xudong963_issue-22189
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                 xudong963_issue-22189 ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.26 / 4.71 ±6.87 / 18.45 ms │          1.24 / 4.59 ±6.65 / 17.88 ms │     no change │
│ QQuery 1  │        12.48 / 12.69 ±0.15 / 12.89 ms │        10.91 / 11.10 ±0.14 / 11.34 ms │ +1.14x faster │
│ QQuery 2  │        36.21 / 36.67 ±0.27 / 36.93 ms │        34.13 / 34.50 ±0.44 / 35.29 ms │ +1.06x faster │
│ QQuery 3  │        30.55 / 31.19 ±0.60 / 32.14 ms │        29.22 / 29.81 ±0.51 / 30.65 ms │     no change │
│ QQuery 4  │     217.32 / 223.39 ±3.74 / 228.60 ms │     217.09 / 221.96 ±2.78 / 224.81 ms │     no change │
│ QQuery 5  │     266.10 / 271.09 ±3.70 / 276.98 ms │     268.20 / 271.66 ±2.54 / 274.57 ms │     no change │
│ QQuery 6  │           1.30 / 1.45 ±0.23 / 1.89 ms │           1.28 / 1.43 ±0.23 / 1.88 ms │     no change │
│ QQuery 7  │        13.57 / 13.78 ±0.20 / 14.05 ms │        12.04 / 12.32 ±0.18 / 12.50 ms │ +1.12x faster │
│ QQuery 8  │     315.95 / 318.99 ±3.14 / 323.79 ms │     315.37 / 320.05 ±2.38 / 321.80 ms │     no change │
│ QQuery 9  │    435.07 / 454.79 ±12.95 / 475.75 ms │     445.11 / 451.20 ±4.44 / 456.27 ms │     no change │
│ QQuery 10 │        70.17 / 72.13 ±1.89 / 75.73 ms │        66.34 / 67.32 ±0.76 / 68.57 ms │ +1.07x faster │
│ QQuery 11 │        81.40 / 83.54 ±2.59 / 88.62 ms │        77.79 / 79.04 ±0.84 / 79.99 ms │ +1.06x faster │
│ QQuery 12 │     264.69 / 270.34 ±4.20 / 277.31 ms │    263.14 / 271.95 ±10.14 / 289.81 ms │     no change │
│ QQuery 13 │     365.77 / 374.15 ±8.92 / 387.17 ms │    366.86 / 381.80 ±13.58 / 404.91 ms │     no change │
│ QQuery 14 │     277.41 / 279.43 ±2.86 / 285.05 ms │     276.05 / 279.18 ±2.24 / 282.35 ms │     no change │
│ QQuery 15 │     258.63 / 265.60 ±5.71 / 273.67 ms │     262.90 / 266.40 ±2.70 / 271.17 ms │     no change │
│ QQuery 16 │     602.10 / 609.59 ±6.44 / 621.44 ms │     603.22 / 611.22 ±4.93 / 615.91 ms │     no change │
│ QQuery 17 │     600.93 / 609.63 ±4.89 / 615.76 ms │     605.77 / 617.94 ±8.38 / 626.90 ms │     no change │
│ QQuery 18 │ 1213.43 / 1242.15 ±17.00 / 1260.20 ms │ 1224.16 / 1255.73 ±22.80 / 1284.53 ms │     no change │
│ QQuery 19 │        27.97 / 28.50 ±0.48 / 29.18 ms │        25.89 / 28.28 ±3.63 / 35.47 ms │     no change │
│ QQuery 20 │     516.27 / 523.06 ±9.59 / 541.99 ms │     510.82 / 518.92 ±7.33 / 532.56 ms │     no change │
│ QQuery 21 │     593.14 / 596.64 ±4.47 / 604.99 ms │     588.67 / 593.71 ±4.38 / 599.62 ms │     no change │
│ QQuery 22 │  1053.86 / 1060.16 ±5.46 / 1067.48 ms │ 1042.37 / 1050.63 ±10.19 / 1070.16 ms │     no change │
│ QQuery 23 │ 3152.07 / 3178.23 ±23.88 / 3222.90 ms │ 3132.00 / 3153.31 ±17.74 / 3181.13 ms │     no change │
│ QQuery 24 │        41.64 / 43.26 ±2.04 / 47.22 ms │        39.62 / 44.23 ±5.20 / 54.02 ms │     no change │
│ QQuery 25 │     110.80 / 114.06 ±4.48 / 122.90 ms │     108.54 / 110.45 ±2.66 / 115.70 ms │     no change │
│ QQuery 26 │        42.05 / 43.64 ±1.97 / 46.80 ms │        39.95 / 41.70 ±2.57 / 46.81 ms │     no change │
│ QQuery 27 │     670.94 / 673.73 ±3.65 / 680.81 ms │     663.16 / 667.61 ±3.19 / 670.77 ms │     no change │
│ QQuery 28 │ 2999.44 / 3013.38 ±14.93 / 3039.95 ms │ 3005.47 / 3033.30 ±23.77 / 3069.80 ms │     no change │
│ QQuery 29 │        41.46 / 47.88 ±5.09 / 52.42 ms │        39.90 / 42.24 ±3.48 / 49.10 ms │ +1.13x faster │
│ QQuery 30 │     296.10 / 301.12 ±4.30 / 308.06 ms │     295.21 / 299.34 ±2.29 / 301.64 ms │     no change │
│ QQuery 31 │     279.66 / 287.45 ±7.73 / 302.19 ms │    278.38 / 293.04 ±21.91 / 336.05 ms │     no change │
│ QQuery 32 │     916.29 / 925.77 ±9.33 / 942.69 ms │    905.03 / 925.18 ±17.57 / 948.83 ms │     no change │
│ QQuery 33 │ 1422.75 / 1436.07 ±10.28 / 1447.74 ms │  1410.02 / 1424.50 ±9.42 / 1433.99 ms │     no change │
│ QQuery 34 │ 1413.39 / 1441.25 ±24.14 / 1482.59 ms │ 1417.68 / 1437.61 ±29.28 / 1495.49 ms │     no change │
│ QQuery 35 │     272.72 / 282.36 ±9.68 / 300.80 ms │    274.21 / 289.62 ±18.13 / 324.00 ms │     no change │
│ QQuery 36 │        64.47 / 68.55 ±3.97 / 75.10 ms │        62.14 / 67.31 ±3.96 / 72.03 ms │     no change │
│ QQuery 37 │        35.08 / 36.57 ±0.86 / 37.62 ms │        34.04 / 35.62 ±2.47 / 40.53 ms │     no change │
│ QQuery 38 │        40.59 / 47.02 ±6.26 / 58.73 ms │        38.13 / 41.12 ±3.38 / 46.59 ms │ +1.14x faster │
│ QQuery 39 │     139.86 / 146.03 ±4.27 / 152.91 ms │     129.16 / 137.89 ±5.54 / 144.67 ms │ +1.06x faster │
│ QQuery 40 │        13.86 / 15.89 ±2.59 / 20.90 ms │        12.42 / 12.62 ±0.20 / 12.99 ms │ +1.26x faster │
│ QQuery 41 │        13.54 / 15.61 ±3.65 / 22.89 ms │        12.01 / 13.40 ±1.99 / 17.33 ms │ +1.16x faster │
│ QQuery 42 │        12.95 / 13.07 ±0.11 / 13.25 ms │        11.57 / 11.66 ±0.09 / 11.84 ms │ +1.12x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                    ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                    │ 19514.59ms │
│ Total Time (xudong963_issue-22189)   │ 19462.49ms │
│ Average Time (HEAD)                  │   453.83ms │
│ Average Time (xudong963_issue-22189) │   452.62ms │
│ Queries Faster                       │         11 │
│ Queries Slower                       │          0 │
│ Queries with No Change               │         32 │
│ Queries with Failure                 │          0 │
└──────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric	Value
Wall time	100.0s
Peak memory	29.9 GiB
Avg memory	23.3 GiB
CPU user	1018.2s
CPU sys	61.6s
Peak spill	0 B

clickbench_partitioned — branch

Metric	Value
Wall time	100.0s
Peak memory	30.3 GiB
Avg memory	23.0 GiB
CPU user	1014.5s
CPU sys	63.6s
Peak spill	0 B

---

File an issue against this benchmark runner