Skip to content

fix: estimate GROUP BY rows from integer bounds - #27574

Open
kumarUjjawal wants to merge 3 commits into
apache:mainfrom
kumarUjjawal:kumarUjjawal/issue-25611-review
Open

kumarUjjawal wants to merge 3 commits into
apache:mainfrom
kumarUjjawal:kumarUjjawal/issue-25611-review

Conversation

@kumarUjjawal

@kumarUjjawal kumarUjjawal commented Nov 19, 2026 •

Copy link
Copy Markdown
Contributor

Which issue does this PR close?

Rationale for this change

In the issue’s TPC-H SF1 example, grouping by shipping year returns seven groups but the planner estimates 6,001,215.

What changes are included in this PR?

Derive year bounds for Date32 inputs, fix the radians bounds, and propagate inexact bounds through projections and partial aggregates. Partial aggregates keep whether a group key has NULLs. A final aggregate caps the distinct count of an integer key by its range. If the key has no distinct count, the range caps a known input row count.

What is the testing strategy for this PR?

The two-partition cases in parquet_statistics.slt check the final estimates, a nullable key without NULLs, radians bounds, and query results. Rust tests cover bounds propagation, NULL groups, and fallbacks.

Are there any user-facing changes?

Tighter group estimates can change join plans.

@github-actions github-actions Bot added physical-expr Changes to the physical-expr crates core Core DataFusion crate sqllogictest SQL Logic Tests (.slt) functions Changes to functions implementation physical-plan Changes to the physical-plan crate labels Nov 19, 2026
@kumarUjjawal

Copy link
Copy Markdown
Contributor Author

run benchmark sql_planner

@kumarUjjawal

Copy link
Copy Markdown
Contributor Author

run benchmark tpch

@kumarUjjawal

Copy link
Copy Markdown
Contributor Author

run benchmark tpcds

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6999920490-5467-sgznq 6.12.105+ #1 SMP Sat Sep 19 07:55:25 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 kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

Run configuration
run benchmark sql_planner

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6999922668-5468-fqfxg 6.12.105+ #1 SMP Sat Sep 19 07:55:25 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 kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

Run configuration
run benchmark tpch

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c6999924060-5469-8hvff 6.12.105+ #1 SMP Sat Sep 19 07:55:25 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 kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

Run configuration
run benchmark tpcds

Results will be posted here when complete


File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

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

Comparing kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

Run configuration
run benchmark tpch
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 kumarUjjawal_issue-25611-review
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Query     ┃     HEAD ┃ kumarUjjawal_issue-25611-review ┃    Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ QQuery 1  │ 40.77 ms │                        40.70 ms │ no change │
│ QQuery 2  │ 18.56 ms │                        18.32 ms │ no change │
│ QQuery 3  │ 28.47 ms │                        28.39 ms │ no change │
│ QQuery 4  │ 17.49 ms │                        17.42 ms │ no change │
│ QQuery 5  │ 35.03 ms │                        34.92 ms │ no change │
│ QQuery 6  │ 13.72 ms │                        13.65 ms │ no change │
│ QQuery 7  │ 41.22 ms │                        40.37 ms │ no change │
│ QQuery 8  │ 40.29 ms │                        40.30 ms │ no change │
│ QQuery 9  │ 49.67 ms │                        47.99 ms │ no change │
│ QQuery 10 │ 40.28 ms │                        40.19 ms │ no change │
│ QQuery 11 │ 13.25 ms │                        13.04 ms │ no change │
│ QQuery 12 │ 18.07 ms │                        17.94 ms │ no change │
│ QQuery 13 │ 39.32 ms │                        39.39 ms │ no change │
│ QQuery 14 │ 24.20 ms │                        24.19 ms │ no change │
│ QQuery 15 │ 31.32 ms │                        31.16 ms │ no change │
│ QQuery 16 │ 13.46 ms │                        13.34 ms │ no change │
│ QQuery 17 │ 66.92 ms │                        64.61 ms │ no change │
│ QQuery 18 │ 40.73 ms │                        40.19 ms │ no change │
│ QQuery 19 │ 31.78 ms │                        32.01 ms │ no change │
│ QQuery 20 │ 31.54 ms │                        31.67 ms │ no change │
│ QQuery 21 │ 55.39 ms │                        55.43 ms │ no change │
│ QQuery 22 │ 13.38 ms │                        13.35 ms │ no change │
└───────────┴──────────┴─────────────────────────────────┴───────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                              │ 704.85ms │
│ Total Time (kumarUjjawal_issue-25611-review)   │ 698.56ms │
│ Average Time (HEAD)                            │  32.04ms │
│ Average Time (kumarUjjawal_issue-25611-review) │  31.75ms │
│ Queries Faster                                 │        0 │
│ Queries Slower                                 │        0 │
│ Queries with No Change                         │       22 │
│ Queries with Failure                           │        0 │
└────────────────────────────────────────────────┴──────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and kumarUjjawal_issue-25611-review
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ kumarUjjawal_issue-25611-review ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │ 40.77 / 41.90 ±1.04 / 43.16 ms │  40.70 / 45.22 ±3.53 / 49.48 ms │ 1.08x slower │
│ QQuery 2  │ 18.56 / 18.78 ±0.31 / 19.39 ms │  18.32 / 18.92 ±0.77 / 20.39 ms │    no change │
│ QQuery 3  │ 28.47 / 28.68 ±0.14 / 28.86 ms │  28.39 / 28.66 ±0.22 / 29.04 ms │    no change │
│ QQuery 4  │ 17.49 / 17.73 ±0.39 / 18.50 ms │  17.42 / 18.03 ±0.67 / 19.16 ms │    no change │
│ QQuery 5  │ 35.03 / 35.44 ±0.57 / 36.55 ms │  34.92 / 35.41 ±0.67 / 36.73 ms │    no change │
│ QQuery 6  │ 13.72 / 14.43 ±0.86 / 15.86 ms │  13.65 / 13.76 ±0.08 / 13.88 ms │    no change │
│ QQuery 7  │ 41.22 / 43.15 ±1.93 / 46.84 ms │  40.37 / 41.99 ±0.98 / 43.28 ms │    no change │
│ QQuery 8  │ 40.29 / 41.02 ±0.64 / 41.83 ms │  40.30 / 40.89 ±0.82 / 42.51 ms │    no change │
│ QQuery 9  │ 49.67 / 50.42 ±0.72 / 51.64 ms │  47.99 / 48.94 ±0.61 / 49.66 ms │    no change │
│ QQuery 10 │ 40.28 / 41.84 ±1.79 / 44.63 ms │  40.19 / 40.34 ±0.14 / 40.60 ms │    no change │
│ QQuery 11 │ 13.25 / 13.76 ±0.64 / 14.91 ms │  13.04 / 13.36 ±0.23 / 13.76 ms │    no change │
│ QQuery 12 │ 18.07 / 18.37 ±0.19 / 18.57 ms │  17.94 / 18.33 ±0.44 / 19.15 ms │    no change │
│ QQuery 13 │ 39.32 / 39.74 ±0.56 / 40.85 ms │  39.39 / 40.56 ±1.10 / 41.96 ms │    no change │
│ QQuery 14 │ 24.20 / 24.99 ±0.95 / 26.75 ms │  24.19 / 24.40 ±0.21 / 24.79 ms │    no change │
│ QQuery 15 │ 31.32 / 32.51 ±1.07 / 34.23 ms │  31.16 / 31.89 ±0.78 / 33.10 ms │    no change │
│ QQuery 16 │ 13.46 / 13.66 ±0.10 / 13.76 ms │  13.34 / 13.57 ±0.14 / 13.76 ms │    no change │
│ QQuery 17 │ 66.92 / 68.02 ±0.62 / 68.62 ms │  64.61 / 65.36 ±0.64 / 66.31 ms │    no change │
│ QQuery 18 │ 40.73 / 42.00 ±1.16 / 43.82 ms │  40.19 / 41.41 ±1.03 / 43.29 ms │    no change │
│ QQuery 19 │ 31.78 / 32.66 ±1.62 / 35.90 ms │  32.01 / 33.13 ±1.24 / 35.30 ms │    no change │
│ QQuery 20 │ 31.54 / 32.29 ±0.74 / 33.67 ms │  31.67 / 31.97 ±0.25 / 32.30 ms │    no change │
│ QQuery 21 │ 55.39 / 55.67 ±0.30 / 56.17 ms │  55.43 / 56.23 ±0.63 / 57.22 ms │    no change │
│ QQuery 22 │ 13.38 / 13.67 ±0.22 / 14.07 ms │  13.35 / 13.67 ±0.30 / 14.22 ms │    no change │
└───────────┴────────────────────────────────┴─────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                              ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                              │ 720.71ms │
│ Total Time (kumarUjjawal_issue-25611-review)   │ 716.06ms │
│ Average Time (HEAD)                            │  32.76ms │
│ Average Time (kumarUjjawal_issue-25611-review) │  32.55ms │
│ Queries Faster                                 │        0 │
│ Queries Slower                                 │        1 │
│ 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 482.5 MiB
CPU user 19.2s
CPU sys 1.7s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 492.5 MiB
CPU user 19.3s
CPU sys 1.6s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

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

Comparing kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

Run configuration
run benchmark tpcds
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 kumarUjjawal_issue-25611-review
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃       HEAD ┃ kumarUjjawal_issue-25611-review ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │    5.38 ms │                         5.38 ms │     no change │
│ QQuery 2  │   80.70 ms │                        79.18 ms │     no change │
│ QQuery 3  │   28.69 ms │                        28.67 ms │     no change │
│ QQuery 4  │  458.23 ms │                       464.58 ms │     no change │
│ QQuery 5  │   50.41 ms │                        49.73 ms │     no change │
│ QQuery 6  │   34.79 ms │                        34.92 ms │     no change │
│ QQuery 7  │   71.93 ms │                        71.82 ms │     no change │
│ QQuery 8  │   36.18 ms │                        38.22 ms │  1.06x slower │
│ QQuery 9  │   54.65 ms │                        52.07 ms │     no change │
│ QQuery 10 │   60.53 ms │                        60.31 ms │     no change │
│ QQuery 11 │  284.67 ms │                       285.19 ms │     no change │
│ QQuery 12 │   29.26 ms │                        29.41 ms │     no change │
│ QQuery 13 │  112.97 ms │                       115.53 ms │     no change │
│ QQuery 14 │  375.56 ms │                       379.72 ms │     no change │
│ QQuery 15 │   54.27 ms │                        55.82 ms │     no change │
│ QQuery 16 │    5.93 ms │                         6.25 ms │  1.05x slower │
│ QQuery 17 │   77.89 ms │                        75.09 ms │     no change │
│ QQuery 18 │   93.70 ms │                        95.27 ms │     no change │
│ QQuery 19 │   40.91 ms │                        41.46 ms │     no change │
│ QQuery 20 │   38.66 ms │                        35.57 ms │ +1.09x faster │
│ QQuery 21 │   17.35 ms │                        17.03 ms │     no change │
│ QQuery 22 │   67.21 ms │                        63.57 ms │ +1.06x faster │
│ QQuery 23 │  301.40 ms │                       296.04 ms │     no change │
│ QQuery 24 │  195.74 ms │                       199.24 ms │     no change │
│ QQuery 25 │  102.76 ms │                       104.17 ms │     no change │
│ QQuery 26 │   46.14 ms │                        46.98 ms │     no change │
│ QQuery 27 │    5.88 ms │                         4.97 ms │ +1.18x faster │
│ QQuery 28 │   56.80 ms │                        56.62 ms │     no change │
│ QQuery 29 │   91.47 ms │                        93.29 ms │     no change │
│ QQuery 30 │   33.24 ms │                        32.84 ms │     no change │
│ QQuery 31 │  107.09 ms │                       107.10 ms │     no change │
│ QQuery 32 │   20.20 ms │                        20.49 ms │     no change │
│ QQuery 33 │   36.66 ms │                        37.69 ms │     no change │
│ QQuery 34 │    5.57 ms │                         6.12 ms │  1.10x slower │
│ QQuery 35 │   73.46 ms │                        72.35 ms │     no change │
│ QQuery 36 │    5.79 ms │                         5.46 ms │ +1.06x faster │
│ QQuery 37 │    7.01 ms │                         7.01 ms │     no change │
│ QQuery 38 │   61.59 ms │                        61.02 ms │     no change │
│ QQuery 39 │   73.28 ms │                        72.55 ms │     no change │
│ QQuery 40 │   24.67 ms │                        23.82 ms │     no change │
│ QQuery 41 │   13.16 ms │                        12.86 ms │     no change │
│ QQuery 42 │   23.68 ms │                        23.56 ms │     no change │
│ QQuery 43 │    4.63 ms │                         4.54 ms │     no change │
│ QQuery 44 │    5.55 ms │                         5.57 ms │     no change │
│ QQuery 45 │   37.69 ms │                        37.39 ms │     no change │
│ QQuery 46 │   11.92 ms │                        11.88 ms │     no change │
│ QQuery 47 │  230.72 ms │                       223.68 ms │     no change │
│ QQuery 48 │   91.77 ms │                        91.17 ms │     no change │
│ QQuery 49 │   68.76 ms │                        68.94 ms │     no change │
│ QQuery 50 │   56.68 ms │                        58.02 ms │     no change │
│ QQuery 51 │   92.32 ms │                        93.07 ms │     no change │
│ QQuery 52 │   24.22 ms │                        23.89 ms │     no change │
│ QQuery 53 │   29.73 ms │                        29.91 ms │     no change │
│ QQuery 54 │   55.13 ms │                        55.02 ms │     no change │
│ QQuery 55 │   24.59 ms │                        23.92 ms │     no change │
│ QQuery 56 │   38.38 ms │                        38.84 ms │     no change │
│ QQuery 57 │  169.34 ms │                       169.08 ms │     no change │
│ QQuery 58 │  106.68 ms │                       106.26 ms │     no change │
│ QQuery 59 │  115.03 ms │                       114.78 ms │     no change │
│ QQuery 60 │   38.36 ms │                        37.69 ms │     no change │
│ QQuery 61 │    9.46 ms │                         8.90 ms │ +1.06x faster │
│ QQuery 62 │   43.94 ms │                        45.44 ms │     no change │
│ QQuery 63 │   29.48 ms │                        29.51 ms │     no change │
│ QQuery 64 │  349.84 ms │                       255.69 ms │ +1.37x faster │
│ QQuery 65 │   73.40 ms │                        66.89 ms │ +1.10x faster │
│ QQuery 66 │   77.89 ms │                        74.67 ms │     no change │
│ QQuery 67 │  245.76 ms │                       227.90 ms │ +1.08x faster │
│ QQuery 68 │    6.88 ms │                         6.57 ms │     no change │
│ QQuery 69 │   55.96 ms │                        56.02 ms │     no change │
│ QQuery 70 │  102.93 ms │                       104.46 ms │     no change │
│ QQuery 71 │   35.21 ms │                        34.79 ms │     no change │
│ QQuery 72 │ 1730.77 ms │                      1569.60 ms │ +1.10x faster │
│ QQuery 73 │    5.21 ms │                         5.70 ms │  1.09x slower │
│ QQuery 74 │  201.59 ms │                       222.59 ms │  1.10x slower │
│ QQuery 75 │  136.85 ms │                       142.43 ms │     no change │
│ QQuery 76 │   34.05 ms │                        35.01 ms │     no change │
│ QQuery 77 │   59.75 ms │                        59.86 ms │     no change │
│ QQuery 78 │  282.10 ms │                       274.93 ms │     no change │
│ QQuery 79 │   67.16 ms │                        66.21 ms │     no change │
│ QQuery 80 │   93.60 ms │                        94.63 ms │     no change │
│ QQuery 81 │   25.89 ms │                        26.16 ms │     no change │
│ QQuery 82 │   16.65 ms │                        16.56 ms │     no change │
│ QQuery 83 │   32.79 ms │                        33.09 ms │     no change │
│ QQuery 84 │   29.95 ms │                        29.45 ms │     no change │
│ QQuery 85 │   97.32 ms │                        98.34 ms │     no change │
│ QQuery 86 │   25.17 ms │                        27.26 ms │  1.08x slower │
│ QQuery 87 │   61.29 ms │                        65.04 ms │  1.06x slower │
│ QQuery 88 │   59.89 ms │                        61.24 ms │     no change │
│ QQuery 89 │   34.92 ms │                        36.61 ms │     no change │
│ QQuery 90 │   16.30 ms │                        17.35 ms │  1.06x slower │
│ QQuery 91 │   40.58 ms │                        42.22 ms │     no change │
│ QQuery 92 │   29.64 ms │                        30.84 ms │     no change │
│ QQuery 93 │   48.91 ms │                        49.37 ms │     no change │
│ QQuery 94 │   38.87 ms │                        39.12 ms │     no change │
│ QQuery 95 │   78.63 ms │                        78.89 ms │     no change │
│ QQuery 96 │   23.92 ms │                        24.00 ms │     no change │
│ QQuery 97 │   50.14 ms │                        51.09 ms │     no change │
│ QQuery 98 │   42.75 ms │                        42.00 ms │     no change │
│ QQuery 99 │   64.87 ms │                        65.65 ms │     no change │
└───────────┴────────────┴─────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                              │ 9027.26ms │
│ Total Time (kumarUjjawal_issue-25611-review)   │ 8778.69ms │
│ Average Time (HEAD)                            │   91.18ms │
│ Average Time (kumarUjjawal_issue-25611-review) │   88.67ms │
│ Queries Faster                                 │         9 │
│ Queries Slower                                 │         8 │
│ Queries with No Change                         │        82 │
│ Queries with Failure                           │         0 │
└────────────────────────────────────────────────┴───────────┘

Distribution per query (min / mean ±stddev / max):

Comparing HEAD and kumarUjjawal_issue-25611-review
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃       kumarUjjawal_issue-25611-review ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.38 / 5.99 ±1.01 / 7.99 ms │           5.38 / 5.93 ±0.92 / 7.76 ms │     no change │
│ QQuery 2  │        80.70 / 81.94 ±0.96 / 83.29 ms │        79.18 / 79.43 ±0.25 / 79.90 ms │     no change │
│ QQuery 3  │        28.69 / 28.92 ±0.18 / 29.14 ms │        28.67 / 28.88 ±0.17 / 29.12 ms │     no change │
│ QQuery 4  │    458.23 / 478.88 ±24.42 / 522.61 ms │    464.58 / 483.51 ±13.44 / 503.91 ms │     no change │
│ QQuery 5  │        50.41 / 51.87 ±0.87 / 52.61 ms │        49.73 / 52.01 ±3.57 / 59.12 ms │     no change │
│ QQuery 6  │        34.79 / 36.07 ±1.15 / 37.99 ms │        34.92 / 35.39 ±0.43 / 36.03 ms │     no change │
│ QQuery 7  │        71.93 / 72.11 ±0.12 / 72.26 ms │        71.82 / 72.54 ±0.71 / 73.75 ms │     no change │
│ QQuery 8  │        36.18 / 36.55 ±0.29 / 36.98 ms │        38.22 / 38.63 ±0.29 / 38.99 ms │  1.06x slower │
│ QQuery 9  │        54.65 / 56.91 ±1.76 / 59.64 ms │        52.07 / 54.32 ±1.24 / 55.58 ms │     no change │
│ QQuery 10 │        60.53 / 61.15 ±0.69 / 62.25 ms │        60.31 / 60.56 ±0.25 / 60.97 ms │     no change │
│ QQuery 11 │     284.67 / 292.60 ±5.95 / 302.84 ms │    285.19 / 301.93 ±16.84 / 332.50 ms │     no change │
│ QQuery 12 │        29.26 / 31.20 ±1.37 / 32.67 ms │        29.41 / 30.17 ±0.52 / 30.81 ms │     no change │
│ QQuery 13 │     112.97 / 114.30 ±2.22 / 118.73 ms │     115.53 / 116.84 ±0.94 / 118.41 ms │     no change │
│ QQuery 14 │     375.56 / 378.97 ±3.80 / 385.48 ms │     379.72 / 384.86 ±5.08 / 393.64 ms │     no change │
│ QQuery 15 │        54.27 / 55.89 ±1.03 / 56.95 ms │        55.82 / 62.87 ±3.85 / 67.41 ms │  1.12x slower │
│ QQuery 16 │           5.93 / 6.07 ±0.13 / 6.31 ms │           6.25 / 6.45 ±0.18 / 6.78 ms │  1.06x slower │
│ QQuery 17 │        77.89 / 79.01 ±0.81 / 80.36 ms │        75.09 / 76.36 ±0.91 / 77.72 ms │     no change │
│ QQuery 18 │        93.70 / 95.93 ±1.42 / 97.74 ms │      95.27 / 100.40 ±4.16 / 105.53 ms │     no change │
│ QQuery 19 │        40.91 / 41.53 ±0.63 / 42.68 ms │        41.46 / 43.33 ±1.95 / 46.81 ms │     no change │
│ QQuery 20 │        38.66 / 39.51 ±1.12 / 41.70 ms │        35.57 / 36.27 ±0.50 / 36.97 ms │ +1.09x faster │
│ QQuery 21 │        17.35 / 17.82 ±0.42 / 18.45 ms │        17.03 / 17.20 ±0.13 / 17.41 ms │     no change │
│ QQuery 22 │        67.21 / 68.69 ±1.05 / 70.12 ms │        63.57 / 64.92 ±1.37 / 67.43 ms │ +1.06x faster │
│ QQuery 23 │     301.40 / 312.07 ±9.27 / 323.75 ms │    296.04 / 310.93 ±10.44 / 323.67 ms │     no change │
│ QQuery 24 │     195.74 / 200.98 ±4.59 / 206.59 ms │     199.24 / 205.08 ±4.87 / 210.24 ms │     no change │
│ QQuery 25 │     102.76 / 105.36 ±2.15 / 108.92 ms │     104.17 / 107.46 ±3.82 / 113.95 ms │     no change │
│ QQuery 26 │        46.14 / 47.07 ±1.00 / 48.69 ms │        46.98 / 47.89 ±0.66 / 48.87 ms │     no change │
│ QQuery 27 │           5.88 / 5.91 ±0.03 / 5.95 ms │           4.97 / 5.17 ±0.24 / 5.63 ms │ +1.14x faster │
│ QQuery 28 │        56.80 / 61.43 ±2.40 / 63.37 ms │        56.62 / 59.88 ±2.48 / 62.17 ms │     no change │
│ QQuery 29 │        91.47 / 93.18 ±1.51 / 95.81 ms │       93.29 / 96.13 ±2.31 / 100.20 ms │     no change │
│ QQuery 30 │        33.24 / 33.94 ±0.58 / 34.76 ms │        32.84 / 34.22 ±1.02 / 35.46 ms │     no change │
│ QQuery 31 │     107.09 / 109.37 ±2.91 / 114.90 ms │     107.10 / 108.74 ±1.45 / 111.37 ms │     no change │
│ QQuery 32 │        20.20 / 20.29 ±0.12 / 20.51 ms │        20.49 / 20.66 ±0.12 / 20.86 ms │     no change │
│ QQuery 33 │        36.66 / 38.11 ±2.08 / 42.23 ms │        37.69 / 38.57 ±0.84 / 40.12 ms │     no change │
│ QQuery 34 │           5.57 / 5.70 ±0.17 / 6.03 ms │           6.12 / 6.28 ±0.11 / 6.42 ms │  1.10x slower │
│ QQuery 35 │        73.46 / 78.16 ±3.76 / 83.20 ms │        72.35 / 73.07 ±0.87 / 74.77 ms │ +1.07x faster │
│ QQuery 36 │           5.79 / 5.91 ±0.16 / 6.22 ms │           5.46 / 5.68 ±0.22 / 6.10 ms │     no change │
│ QQuery 37 │           7.01 / 7.23 ±0.12 / 7.37 ms │           7.01 / 7.06 ±0.04 / 7.12 ms │     no change │
│ QQuery 38 │        61.59 / 62.21 ±0.38 / 62.77 ms │        61.02 / 62.60 ±1.95 / 66.42 ms │     no change │
│ QQuery 39 │        73.28 / 74.90 ±1.51 / 77.13 ms │        72.55 / 76.71 ±2.42 / 79.12 ms │     no change │
│ QQuery 40 │        24.67 / 25.58 ±0.83 / 27.01 ms │        23.82 / 24.58 ±0.49 / 25.01 ms │     no change │
│ QQuery 41 │        13.16 / 13.24 ±0.07 / 13.35 ms │        12.86 / 12.99 ±0.08 / 13.10 ms │     no change │
│ QQuery 42 │        23.68 / 24.11 ±0.37 / 24.75 ms │        23.56 / 24.28 ±0.59 / 25.31 ms │     no change │
│ QQuery 43 │           4.63 / 4.78 ±0.20 / 5.18 ms │           4.54 / 4.75 ±0.21 / 5.11 ms │     no change │
│ QQuery 44 │           5.55 / 5.69 ±0.10 / 5.87 ms │           5.57 / 5.70 ±0.17 / 6.04 ms │     no change │
│ QQuery 45 │        37.69 / 38.40 ±0.47 / 39.02 ms │        37.39 / 38.75 ±1.66 / 41.81 ms │     no change │
│ QQuery 46 │        11.92 / 12.07 ±0.24 / 12.55 ms │        11.88 / 12.29 ±0.65 / 13.58 ms │     no change │
│ QQuery 47 │     230.72 / 240.10 ±9.93 / 259.09 ms │    223.68 / 236.40 ±12.23 / 255.92 ms │     no change │
│ QQuery 48 │        91.77 / 93.94 ±1.59 / 95.91 ms │        91.17 / 92.72 ±1.54 / 95.28 ms │     no change │
│ QQuery 49 │        68.76 / 69.22 ±0.40 / 69.94 ms │        68.94 / 70.62 ±2.25 / 75.07 ms │     no change │
│ QQuery 50 │        56.68 / 58.42 ±2.31 / 62.96 ms │        58.02 / 58.64 ±0.44 / 59.31 ms │     no change │
│ QQuery 51 │       92.32 / 97.59 ±4.10 / 102.86 ms │        93.07 / 95.15 ±2.07 / 99.02 ms │     no change │
│ QQuery 52 │        24.22 / 24.45 ±0.18 / 24.70 ms │        23.89 / 24.17 ±0.21 / 24.47 ms │     no change │
│ QQuery 53 │        29.73 / 29.89 ±0.11 / 30.06 ms │        29.91 / 31.79 ±3.07 / 37.91 ms │  1.06x slower │
│ QQuery 54 │        55.13 / 58.30 ±4.12 / 65.86 ms │        55.02 / 56.34 ±0.78 / 57.29 ms │     no change │
│ QQuery 55 │        24.59 / 25.66 ±1.34 / 28.23 ms │        23.92 / 24.61 ±1.12 / 26.83 ms │     no change │
│ QQuery 56 │        38.38 / 39.93 ±0.87 / 41.06 ms │        38.84 / 39.48 ±0.38 / 39.81 ms │     no change │
│ QQuery 57 │     169.34 / 173.55 ±3.38 / 179.03 ms │     169.08 / 176.41 ±6.23 / 184.54 ms │     no change │
│ QQuery 58 │     106.68 / 107.27 ±0.40 / 107.72 ms │     106.26 / 110.08 ±2.66 / 114.06 ms │     no change │
│ QQuery 59 │     115.03 / 118.39 ±2.32 / 121.03 ms │     114.78 / 115.74 ±0.65 / 116.58 ms │     no change │
│ QQuery 60 │        38.36 / 38.79 ±0.41 / 39.36 ms │        37.69 / 37.97 ±0.23 / 38.30 ms │     no change │
│ QQuery 61 │           9.46 / 9.49 ±0.05 / 9.60 ms │           8.90 / 9.03 ±0.14 / 9.30 ms │     no change │
│ QQuery 62 │        43.94 / 45.44 ±2.56 / 50.56 ms │        45.44 / 46.57 ±0.66 / 47.49 ms │     no change │
│ QQuery 63 │        29.48 / 30.05 ±0.35 / 30.54 ms │        29.51 / 30.70 ±0.82 / 31.95 ms │     no change │
│ QQuery 64 │    349.84 / 363.46 ±10.40 / 375.79 ms │     255.69 / 258.93 ±3.34 / 264.92 ms │ +1.40x faster │
│ QQuery 65 │        73.40 / 74.26 ±1.23 / 76.69 ms │        66.89 / 67.11 ±0.22 / 67.51 ms │ +1.11x faster │
│ QQuery 66 │        77.89 / 79.29 ±2.17 / 83.60 ms │        74.67 / 77.67 ±1.88 / 79.92 ms │     no change │
│ QQuery 67 │     245.76 / 254.18 ±6.69 / 264.93 ms │     227.90 / 234.89 ±5.42 / 244.31 ms │ +1.08x faster │
│ QQuery 68 │           6.88 / 7.16 ±0.27 / 7.63 ms │          6.57 / 8.21 ±2.95 / 14.10 ms │  1.15x slower │
│ QQuery 69 │        55.96 / 56.30 ±0.39 / 57.01 ms │        56.02 / 56.49 ±0.30 / 56.89 ms │     no change │
│ QQuery 70 │     102.93 / 106.93 ±2.64 / 110.35 ms │     104.46 / 107.65 ±2.86 / 112.10 ms │     no change │
│ QQuery 71 │        35.21 / 38.18 ±4.82 / 47.78 ms │        34.79 / 35.90 ±0.75 / 36.82 ms │ +1.06x faster │
│ QQuery 72 │ 1730.77 / 1809.79 ±46.01 / 1856.65 ms │ 1569.60 / 1650.23 ±65.81 / 1762.86 ms │ +1.10x faster │
│ QQuery 73 │           5.21 / 5.35 ±0.19 / 5.72 ms │           5.70 / 5.76 ±0.06 / 5.85 ms │  1.08x slower │
│ QQuery 74 │     201.59 / 209.57 ±5.63 / 218.78 ms │     222.59 / 224.89 ±1.72 / 227.81 ms │  1.07x slower │
│ QQuery 75 │     136.85 / 140.20 ±2.72 / 144.28 ms │     142.43 / 144.71 ±1.40 / 146.09 ms │     no change │
│ QQuery 76 │        34.05 / 36.14 ±3.71 / 43.55 ms │        35.01 / 35.93 ±0.47 / 36.22 ms │     no change │
│ QQuery 77 │        59.75 / 60.38 ±0.42 / 60.93 ms │        59.86 / 61.36 ±1.76 / 64.71 ms │     no change │
│ QQuery 78 │    282.10 / 295.17 ±10.20 / 311.54 ms │    274.93 / 293.94 ±15.89 / 319.60 ms │     no change │
│ QQuery 79 │        67.16 / 71.43 ±3.49 / 75.45 ms │        66.21 / 67.00 ±0.61 / 67.65 ms │ +1.07x faster │
│ QQuery 80 │       93.60 / 95.38 ±2.43 / 100.13 ms │        94.63 / 97.05 ±1.89 / 99.95 ms │     no change │
│ QQuery 81 │        25.89 / 26.13 ±0.21 / 26.49 ms │        26.16 / 27.11 ±1.03 / 29.02 ms │     no change │
│ QQuery 82 │        16.65 / 16.75 ±0.08 / 16.88 ms │        16.56 / 16.83 ±0.33 / 17.47 ms │     no change │
│ QQuery 83 │        32.79 / 34.66 ±3.30 / 41.26 ms │        33.09 / 33.23 ±0.13 / 33.45 ms │     no change │
│ QQuery 84 │        29.95 / 30.71 ±0.43 / 31.07 ms │        29.45 / 29.52 ±0.07 / 29.64 ms │     no change │
│ QQuery 85 │        97.32 / 98.69 ±0.83 / 99.46 ms │      98.34 / 101.79 ±1.89 / 103.68 ms │     no change │
│ QQuery 86 │        25.17 / 25.39 ±0.24 / 25.69 ms │        27.26 / 27.82 ±0.69 / 29.13 ms │  1.10x slower │
│ QQuery 87 │        61.29 / 64.40 ±3.62 / 70.43 ms │        65.04 / 66.03 ±0.96 / 67.26 ms │     no change │
│ QQuery 88 │        59.89 / 60.88 ±1.03 / 62.32 ms │        61.24 / 63.07 ±2.66 / 68.32 ms │     no change │
│ QQuery 89 │        34.92 / 35.07 ±0.15 / 35.33 ms │        36.61 / 37.71 ±0.85 / 39.23 ms │  1.08x slower │
│ QQuery 90 │        16.30 / 16.49 ±0.14 / 16.73 ms │        17.35 / 17.97 ±0.97 / 19.90 ms │  1.09x slower │
│ QQuery 91 │        40.58 / 40.65 ±0.07 / 40.78 ms │        42.22 / 42.90 ±0.52 / 43.81 ms │  1.06x slower │
│ QQuery 92 │        29.64 / 31.90 ±3.27 / 38.32 ms │        30.84 / 31.24 ±0.41 / 31.93 ms │     no change │
│ QQuery 93 │        48.91 / 52.24 ±2.34 / 55.61 ms │        49.37 / 50.40 ±1.31 / 52.94 ms │     no change │
│ QQuery 94 │        38.87 / 39.72 ±0.69 / 40.76 ms │        39.12 / 39.84 ±0.80 / 41.10 ms │     no change │
│ QQuery 95 │        78.63 / 78.82 ±0.14 / 79.04 ms │        78.89 / 80.42 ±1.72 / 83.47 ms │     no change │
│ QQuery 96 │        23.92 / 24.20 ±0.33 / 24.84 ms │        24.00 / 24.08 ±0.07 / 24.17 ms │     no change │
│ QQuery 97 │        50.14 / 53.12 ±3.67 / 60.28 ms │        51.09 / 51.79 ±0.69 / 53.10 ms │     no change │
│ QQuery 98 │        42.75 / 43.38 ±0.64 / 44.62 ms │        42.00 / 44.34 ±2.42 / 48.60 ms │     no change │
│ QQuery 99 │        64.87 / 65.28 ±0.29 / 65.71 ms │        65.65 / 67.19 ±1.13 / 68.65 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Benchmark Summary                              ┃           ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Time (HEAD)                              │ 9319.73ms │
│ Total Time (kumarUjjawal_issue-25611-review)   │ 9079.72ms │
│ Average Time (HEAD)                            │   94.14ms │
│ Average Time (kumarUjjawal_issue-25611-review) │   91.71ms │
│ Queries Faster                                 │        10 │
│ Queries Slower                                 │        12 │
│ Queries with No Change                         │        77 │
│ Queries with Failure                           │         0 │
└────────────────────────────────────────────────┴───────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 50.0s
Peak memory 2.1 GiB
Avg memory 1.5 GiB
CPU user 192.5s
CPU sys 5.3s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 50.0s
Peak memory 2.3 GiB
Avg memory 1.5 GiB
CPU user 188.8s
CPU sys 5.5s
Peak spill 0 B

File an issue against this benchmark runner

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark completed (GKE) | trigger

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

Comparing kumarUjjawal/issue-25611-review (c5857fb) to 5ea948b (merge-base) diff

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

group                                                 HEAD                                   kumarUjjawal_issue-25611-review
-----                                                 ----                                   -------------------------------
logical_aggregate_with_join                           1.00    315.0±1.73µs        ? ?/sec    1.00    316.5±1.86µs        ? ?/sec
logical_correlated_subquery_exists                    1.00    210.5±0.69µs        ? ?/sec    1.01    212.0±0.63µs        ? ?/sec
logical_correlated_subquery_in                        1.00    212.2±1.36µs        ? ?/sec    1.01    214.7±1.41µs        ? ?/sec
logical_distinct_many_columns                         1.01    260.8±0.91µs        ? ?/sec    1.00    258.8±1.27µs        ? ?/sec
logical_join_4_with_agg_and_filter                    1.00    194.3±0.98µs        ? ?/sec    1.02    198.2±1.61µs        ? ?/sec
logical_join_8_with_agg_sort_limit                    1.00    335.8±2.81µs        ? ?/sec    1.01    340.1±2.06µs        ? ?/sec
logical_join_chain_16                                 1.00    614.8±2.68µs        ? ?/sec    1.02    627.1±7.28µs        ? ?/sec
logical_join_chain_4                                  1.00     73.5±0.33µs        ? ?/sec    1.00     73.7±0.32µs        ? ?/sec
logical_join_chain_8                                  1.00    197.2±0.86µs        ? ?/sec    1.02    200.2±1.17µs        ? ?/sec
logical_multiple_subqueries                           1.00    398.4±1.80µs        ? ?/sec    1.01    400.8±1.31µs        ? ?/sec
logical_nested_cte_4_levels                           1.00    197.5±1.98µs        ? ?/sec    1.02    200.5±1.19µs        ? ?/sec
logical_plan_struct_join_agg_sort                     1.00    123.6±1.52µs        ? ?/sec    1.00    123.8±0.85µs        ? ?/sec
logical_plan_tpcds_all                                1.00     70.1±0.35ms        ? ?/sec    1.00     70.3±0.15ms        ? ?/sec
logical_plan_tpch_all                                 1.00      5.1±0.02ms        ? ?/sec    1.00      5.1±0.02ms        ? ?/sec
logical_scalar_subquery                               1.00    217.8±0.89µs        ? ?/sec    1.02    221.4±3.25µs        ? ?/sec
logical_select_all_from_1000                          1.01      2.3±0.01ms        ? ?/sec    1.00      2.3±0.01ms        ? ?/sec
logical_select_one_from_700                           1.00    235.5±1.81µs        ? ?/sec    1.00    234.6±2.05µs        ? ?/sec
logical_trivial_join_high_numbered_columns            1.00    195.4±1.52µs        ? ?/sec    1.01    196.5±1.35µs        ? ?/sec
logical_trivial_join_low_numbered_columns             1.00    194.1±1.33µs        ? ?/sec    1.01    195.7±1.35µs        ? ?/sec
logical_union_4_branches                              1.00    315.8±1.41µs        ? ?/sec    1.00    317.0±1.28µs        ? ?/sec
logical_union_8_branches                              1.00    639.8±1.87µs        ? ?/sec    1.00    639.0±2.53µs        ? ?/sec
logical_wide_aggregate_1000_exprs                     1.01     21.2±0.06ms        ? ?/sec    1.00     21.0±0.06ms        ? ?/sec
logical_wide_aggregate_100_exprs                      1.00   1458.9±3.78µs        ? ?/sec    1.00   1459.2±3.13µs        ? ?/sec
logical_wide_case_50_exprs                            1.00   1425.7±3.29µs        ? ?/sec    1.00   1423.4±2.83µs        ? ?/sec
logical_wide_filter_200_predicates                    1.00   1081.3±6.37µs        ? ?/sec    1.01   1090.0±5.03µs        ? ?/sec
logical_wide_filter_50_predicates                     1.00    318.5±1.09µs        ? ?/sec    1.01    322.0±1.18µs        ? ?/sec
optimizer_correlated_exists                           1.00    220.3±0.97µs        ? ?/sec    1.00    219.6±1.71µs        ? ?/sec
optimizer_join_4_with_agg_filter                      1.00    383.6±3.32µs        ? ?/sec    1.03    394.7±3.41µs        ? ?/sec
optimizer_join_chain_4                                1.00    136.1±0.28µs        ? ?/sec    1.01    137.8±0.60µs        ? ?/sec
optimizer_join_chain_8                                1.00    369.7±0.76µs        ? ?/sec    1.00    370.4±1.72µs        ? ?/sec
optimizer_select_all_from_1000                        1.00      2.4±0.00ms        ? ?/sec    1.01      2.4±0.02ms        ? ?/sec
optimizer_select_one_from_700                         1.00    238.4±0.76µs        ? ?/sec    1.01    239.9±1.53µs        ? ?/sec
optimizer_tpcds_all                                   1.00    229.6±0.35ms        ? ?/sec    1.01    231.7±1.28ms        ? ?/sec
optimizer_tpch_all                                    1.00     14.0±0.02ms        ? ?/sec    1.02     14.3±0.07ms        ? ?/sec
optimizer_wide_aggregate_100                          1.00   1753.6±2.42µs        ? ?/sec    1.01  1769.3±14.65µs        ? ?/sec
optimizer_wide_filter_200                             1.00      3.1±0.01ms        ? ?/sec    1.00      3.1±0.01ms        ? ?/sec
optimizer_wide_projection_stack_all                   1.00      5.9±0.01ms        ? ?/sec    1.00      5.8±0.04ms        ? ?/sec
optimizer_wide_projection_stack_half                  1.00      4.1±0.00ms        ? ?/sec    1.00      4.0±0.03ms        ? ?/sec
physical_cte_materialized_chain_6                     1.00      2.4±0.01ms        ? ?/sec    1.00      2.4±0.00ms        ? ?/sec
physical_cte_materialized_one_ref                     1.01    757.7±2.39µs        ? ?/sec    1.00    750.1±2.38µs        ? ?/sec
physical_cte_materialized_two_refs                    1.00   1000.1±1.52µs        ? ?/sec    1.01   1007.6±2.62µs        ? ?/sec
physical_cte_mixed                                    1.00      3.1±0.01ms        ? ?/sec    1.01      3.1±0.01ms        ? ?/sec
physical_cte_not_materialized_chain_6                 1.00     17.7±0.06ms        ? ?/sec    1.01     17.8±0.09ms        ? ?/sec
physical_cte_not_materialized_one_ref                 1.00    686.2±2.92µs        ? ?/sec    1.02    699.1±1.91µs        ? ?/sec
physical_cte_not_materialized_two_refs                1.00   1219.9±2.77µs        ? ?/sec    1.02   1239.6±2.11µs        ? ?/sec
physical_cte_unmarked_chain_6                         1.01     18.0±0.14ms        ? ?/sec    1.00     17.8±0.14ms        ? ?/sec
physical_cte_unmarked_one_ref                         1.03    692.8±2.17µs        ? ?/sec    1.00    674.5±2.37µs        ? ?/sec
physical_cte_unmarked_two_refs                        1.01   1238.6±4.18µs        ? ?/sec    1.00   1230.8±4.70µs        ? ?/sec
physical_intersection                                 1.00    525.6±2.87µs        ? ?/sec    1.00    527.2±2.19µs        ? ?/sec
physical_join_consider_sort                           1.02    994.5±4.07µs        ? ?/sec    1.00    976.4±2.20µs        ? ?/sec
physical_join_distinct                                1.00    191.4±1.30µs        ? ?/sec    1.00    190.5±1.49µs        ? ?/sec
physical_many_self_joins                              1.01      5.5±0.02ms        ? ?/sec    1.00      5.5±0.01ms        ? ?/sec
physical_plan_clickbench_all                          1.00    151.9±0.74ms        ? ?/sec    1.01    152.8±1.03ms        ? ?/sec
physical_plan_clickbench_q1                           1.02  1305.9±10.86µs        ? ?/sec    1.00   1284.4±9.81µs        ? ?/sec
physical_plan_clickbench_q10                          1.01   1895.0±7.55µs        ? ?/sec    1.00  1882.6±12.48µs        ? ?/sec
physical_plan_clickbench_q11                          1.02      2.7±0.02ms        ? ?/sec    1.00      2.7±0.02ms        ? ?/sec
physical_plan_clickbench_q12                          1.01      2.8±0.02ms        ? ?/sec    1.00      2.7±0.02ms        ? ?/sec
physical_plan_clickbench_q13                          1.02      2.6±0.02ms        ? ?/sec    1.00      2.6±0.02ms        ? ?/sec
physical_plan_clickbench_q14                          1.00      2.7±0.02ms        ? ?/sec    1.00      2.7±0.02ms        ? ?/sec
physical_plan_clickbench_q15                          1.00      2.6±0.02ms        ? ?/sec    1.01      2.6±0.02ms        ? ?/sec
physical_plan_clickbench_q16                          1.04  1705.0±10.01µs        ? ?/sec    1.00  1643.8±12.04µs        ? ?/sec
physical_plan_clickbench_q17                          1.04   1747.0±8.60µs        ? ?/sec    1.00  1686.4±11.15µs        ? ?/sec
physical_plan_clickbench_q18                          1.03   1579.4±9.03µs        ? ?/sec    1.00   1527.3±9.93µs        ? ?/sec
physical_plan_clickbench_q19                          1.01  1929.9±10.07µs        ? ?/sec    1.00  1901.4±12.92µs        ? ?/sec
physical_plan_clickbench_q2                           1.03      2.2±0.02ms        ? ?/sec    1.00      2.1±0.02ms        ? ?/sec
physical_plan_clickbench_q20                          1.01  1864.6±15.89µs        ? ?/sec    1.00  1845.1±13.46µs        ? ?/sec
physical_plan_clickbench_q21                          1.01  1709.4±23.17µs        ? ?/sec    1.00  1699.3±21.03µs        ? ?/sec
physical_plan_clickbench_q22                          1.01      2.9±0.02ms        ? ?/sec    1.00      2.9±0.02ms        ? ?/sec
physical_plan_clickbench_q23                          1.00      3.2±0.03ms        ? ?/sec    1.00      3.3±0.02ms        ? ?/sec
physical_plan_clickbench_q24                          1.01      4.8±0.03ms        ? ?/sec    1.00      4.8±0.02ms        ? ?/sec
physical_plan_clickbench_q25                          1.03      2.5±0.02ms        ? ?/sec    1.00      2.4±0.02ms        ? ?/sec
physical_plan_clickbench_q26                          1.03      2.3±0.02ms        ? ?/sec    1.00      2.2±0.02ms        ? ?/sec
physical_plan_clickbench_q27                          1.01      2.4±0.02ms        ? ?/sec    1.00      2.4±0.01ms        ? ?/sec
physical_plan_clickbench_q28                          1.03      2.8±0.02ms        ? ?/sec    1.00      2.8±0.02ms        ? ?/sec
physical_plan_clickbench_q29                          1.00      2.9±0.02ms        ? ?/sec    1.00      2.9±0.02ms        ? ?/sec
physical_plan_clickbench_q3                           1.03   1571.8±7.47µs        ? ?/sec    1.00  1521.4±12.27µs        ? ?/sec
physical_plan_clickbench_q30                          1.01     11.7±0.04ms        ? ?/sec    1.00     11.6±0.03ms        ? ?/sec
physical_plan_clickbench_q31                          1.03      3.0±0.03ms        ? ?/sec    1.00      2.9±0.02ms        ? ?/sec
physical_plan_clickbench_q32                          1.02      3.0±0.02ms        ? ?/sec    1.00      2.9±0.02ms        ? ?/sec
physical_plan_clickbench_q33                          1.00   1896.4±7.76µs        ? ?/sec    1.00  1892.1±13.25µs        ? ?/sec
physical_plan_clickbench_q34                          1.01   1694.0±8.51µs        ? ?/sec    1.00   1674.8±9.99µs        ? ?/sec
physical_plan_clickbench_q35                          1.00   1709.2±7.52µs        ? ?/sec    1.00   1713.4±8.51µs        ? ?/sec
physical_plan_clickbench_q36                          1.02   1993.4±8.38µs        ? ?/sec    1.00  1948.1±11.47µs        ? ?/sec
physical_plan_clickbench_q37                          1.02      3.2±0.01ms        ? ?/sec    1.00      3.1±0.01ms        ? ?/sec
physical_plan_clickbench_q38                          1.02      3.1±0.01ms        ? ?/sec    1.00      3.1±0.02ms        ? ?/sec
physical_plan_clickbench_q39                          1.02      3.1±0.02ms        ? ?/sec    1.00      3.0±0.01ms        ? ?/sec
physical_plan_clickbench_q4                           1.03   1402.0±9.23µs        ? ?/sec    1.00   1357.7±9.51µs        ? ?/sec
physical_plan_clickbench_q40                          1.01      3.5±0.02ms        ? ?/sec    1.00      3.4±0.01ms        ? ?/sec
physical_plan_clickbench_q41                          1.00      3.4±0.01ms        ? ?/sec    1.00      3.4±0.02ms        ? ?/sec
physical_plan_clickbench_q42                          1.00      3.4±0.01ms        ? ?/sec    1.02      3.5±0.13ms        ? ?/sec
physical_plan_clickbench_q43                          1.00      3.4±0.01ms        ? ?/sec    1.00      3.4±0.02ms        ? ?/sec
physical_plan_clickbench_q44                          1.00   1474.0±9.45µs        ? ?/sec    1.00  1476.6±14.81µs        ? ?/sec
physical_plan_clickbench_q45                          1.01  1492.0±17.18µs        ? ?/sec    1.00  1476.0±13.13µs        ? ?/sec
physical_plan_clickbench_q46                          1.01   1753.0±7.42µs        ? ?/sec    1.00   1734.5±7.64µs        ? ?/sec
physical_plan_clickbench_q47                          1.00      2.3±0.01ms        ? ?/sec    1.00      2.3±0.01ms        ? ?/sec
physical_plan_clickbench_q48                          1.00      3.0±0.02ms        ? ?/sec    1.03      3.1±0.06ms        ? ?/sec
physical_plan_clickbench_q49                          1.00      3.0±0.02ms        ? ?/sec    1.06      3.2±0.05ms        ? ?/sec
physical_plan_clickbench_q5                           1.03   1514.0±9.21µs        ? ?/sec    1.00   1476.1±9.42µs        ? ?/sec
physical_plan_clickbench_q50                          1.00      5.0±0.02ms        ? ?/sec    1.03      5.1±0.06ms        ? ?/sec
physical_plan_clickbench_q51                          1.01  1887.5±14.20µs        ? ?/sec    1.00  1863.4±11.21µs        ? ?/sec
physical_plan_clickbench_q52                          1.00      2.3±0.02ms        ? ?/sec    1.01      2.3±0.01ms        ? ?/sec
physical_plan_clickbench_q53                          1.00   1677.8±9.81µs        ? ?/sec    1.00  1680.1±13.40µs        ? ?/sec
physical_plan_clickbench_q54                          1.01   1689.9±7.30µs        ? ?/sec    1.00  1667.8±14.13µs        ? ?/sec
physical_plan_clickbench_q55                          1.00   1626.4±9.27µs        ? ?/sec    1.02  1653.0±11.62µs        ? ?/sec
physical_plan_clickbench_q56                          1.00   1636.3±8.38µs        ? ?/sec    1.02  1665.8±15.81µs        ? ?/sec
physical_plan_clickbench_q57                          1.00      2.2±0.02ms        ? ?/sec    1.02      2.2±0.03ms        ? ?/sec
physical_plan_clickbench_q58                          1.00      2.8±0.03ms        ? ?/sec    1.01      2.9±0.03ms        ? ?/sec
physical_plan_clickbench_q59                          1.00   1618.8±9.92µs        ? ?/sec    1.03  1666.3±19.01µs        ? ?/sec
physical_plan_clickbench_q6                           1.03  1538.5±10.99µs        ? ?/sec    1.00   1497.1±7.77µs        ? ?/sec
physical_plan_clickbench_q60                          1.03   1641.7±7.60µs        ? ?/sec    1.00  1596.3±11.33µs        ? ?/sec
physical_plan_clickbench_q7                           1.00   1330.6±7.23µs        ? ?/sec    1.00   1331.7±8.88µs        ? ?/sec
physical_plan_clickbench_q8                           1.02      2.3±0.02ms        ? ?/sec    1.00      2.3±0.02ms        ? ?/sec
physical_plan_clickbench_q9                           1.02   1815.8±7.95µs        ? ?/sec    1.00  1782.6±10.77µs        ? ?/sec
physical_plan_struct_join_agg_sort                    1.01   1237.3±8.39µs        ? ?/sec    1.00   1229.9±7.88µs        ? ?/sec
physical_plan_tpcds_all                               1.00    554.4±3.59ms        ? ?/sec    1.00    556.0±2.32ms        ? ?/sec
physical_plan_tpch_all                                1.02     40.5±1.63ms        ? ?/sec    1.00     39.5±0.06ms        ? ?/sec
physical_plan_tpch_q1                                 1.00   1355.3±4.19µs        ? ?/sec    1.03   1393.3±2.48µs        ? ?/sec
physical_plan_tpch_q10                                1.00      2.1±0.00ms        ? ?/sec    1.01      2.1±0.00ms        ? ?/sec
physical_plan_tpch_q11                                1.00      2.1±0.00ms        ? ?/sec    1.01      2.1±0.00ms        ? ?/sec
physical_plan_tpch_q12                                1.00   1234.8±3.24µs        ? ?/sec    1.00   1238.9±2.53µs        ? ?/sec
physical_plan_tpch_q13                                1.01   1023.9±2.73µs        ? ?/sec    1.00   1012.2±3.02µs        ? ?/sec
physical_plan_tpch_q14                                1.00   1173.2±3.03µs        ? ?/sec    1.02   1191.0±2.35µs        ? ?/sec
physical_plan_tpch_q16                                1.00   1597.6±2.20µs        ? ?/sec    1.01   1620.4±3.05µs        ? ?/sec
physical_plan_tpch_q17                                1.00   1495.2±1.90µs        ? ?/sec    1.01   1511.8±3.12µs        ? ?/sec
physical_plan_tpch_q18                                1.00   1826.0±2.47µs        ? ?/sec    1.00   1828.6±3.08µs        ? ?/sec
physical_plan_tpch_q19                                1.00   1935.3±3.20µs        ? ?/sec    1.01   1945.6±3.51µs        ? ?/sec
physical_plan_tpch_q2                                 1.00      3.2±0.00ms        ? ?/sec    1.00      3.2±0.00ms        ? ?/sec
physical_plan_tpch_q20                                1.00   1993.2±4.56µs        ? ?/sec    1.00   1996.5±3.68µs        ? ?/sec
physical_plan_tpch_q21                                1.00      2.4±0.00ms        ? ?/sec    1.00      2.4±0.00ms        ? ?/sec
physical_plan_tpch_q22                                1.00   1495.7±4.16µs        ? ?/sec    1.00   1489.6±2.45µs        ? ?/sec
physical_plan_tpch_q3                                 1.00   1627.1±5.26µs        ? ?/sec    1.00   1623.6±8.83µs        ? ?/sec
physical_plan_tpch_q4                                 1.00   1077.6±2.44µs        ? ?/sec    1.01   1093.6±2.47µs        ? ?/sec
physical_plan_tpch_q5                                 1.00      2.1±0.00ms        ? ?/sec    1.02      2.1±0.00ms        ? ?/sec
physical_plan_tpch_q6                                 1.00    593.7±2.85µs        ? ?/sec    1.02    604.3±2.39µs        ? ?/sec
physical_plan_tpch_q7                                 1.00      2.4±0.00ms        ? ?/sec    1.01      2.5±0.00ms        ? ?/sec
physical_plan_tpch_q8                                 1.00      3.1±0.01ms        ? ?/sec    1.01      3.1±0.01ms        ? ?/sec
physical_plan_tpch_q9                                 1.00      2.4±0.00ms        ? ?/sec    1.00      2.4±0.01ms        ? ?/sec
physical_select_aggregates_from_200                   1.00      5.9±0.01ms        ? ?/sec    1.00      5.8±0.01ms        ? ?/sec
physical_select_all_from_1000                         1.00      5.7±0.01ms        ? ?/sec    1.00      5.7±0.01ms        ? ?/sec
physical_select_one_from_700                          1.00    647.8±3.50µs        ? ?/sec    1.00    650.7±3.26µs        ? ?/sec
physical_sorted_union_order_by_10_int64               1.00      3.4±0.00ms        ? ?/sec    1.01      3.5±0.02ms        ? ?/sec
physical_sorted_union_order_by_10_uint64              1.00      7.4±0.06ms        ? ?/sec    1.00      7.4±0.01ms        ? ?/sec
physical_sorted_union_order_by_50_int64               1.00     67.0±0.34ms        ? ?/sec    1.00     66.9±0.24ms        ? ?/sec
physical_sorted_union_order_by_50_uint64              1.02    312.7±6.50ms        ? ?/sec    1.00    305.8±1.05ms        ? ?/sec
physical_theta_join_consider_sort                     1.00   1007.8±2.92µs        ? ?/sec    1.00   1008.8±4.05µs        ? ?/sec
physical_unnest_to_join                               1.00    558.5±6.02µs        ? ?/sec    1.01    562.2±2.22µs        ? ?/sec
physical_window_function_partition_by_12_on_values    1.00    599.5±2.02µs        ? ?/sec    1.00    598.6±1.70µs        ? ?/sec
physical_window_function_partition_by_30_on_values    1.01   1032.3±2.86µs        ? ?/sec    1.00   1025.6±3.38µs        ? ?/sec
physical_window_function_partition_by_4_on_values     1.00    386.1±1.18µs        ? ?/sec    1.04    399.7±1.64µs        ? ?/sec
physical_window_function_partition_by_7_on_values     1.00    466.8±1.28µs        ? ?/sec    1.00    467.5±2.11µs        ? ?/sec
physical_window_function_partition_by_8_on_values     1.00    496.0±2.46µs        ? ?/sec    1.00    494.8±2.03µs        ? ?/sec
with_param_values_many_columns                        1.00    241.8±1.36µs        ? ?/sec    1.00    241.6±2.06µs        ? ?/sec

Resource Usage

sql_planner — base (merge-base)

Metric Value
Wall time 2090.5s
Peak memory 144.8 MiB
Avg memory 99.7 MiB
CPU user 2096.0s
CPU sys 1.6s
Peak spill 0 B

sql_planner — branch

Metric Value
Wall time 2165.5s
Peak memory 144.8 MiB
Avg memory 98.1 MiB
CPU user 2112.5s
CPU sys 1.6s
Peak spill 0 B

File an issue against this benchmark runner

@asolimando asolimando left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the PR @kumarUjjawal! The radians comment can change plans so I consider it "blocking". The others are minor estimate issues, but they are easy to fix.

} else if expr.is::<NegativeExpr>() {
supported_inputs()
} else {
expr.is::<ScalarFunctionExpr>()

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

radians_bounds returns a fixed (-pi, pi), which is wrong (radians(720) is about 12.57). Since this PR uses these bounds for statistics, the join below is estimated at 0 rows but returns 301. The merge base estimates 301.

statement ok
COPY (SELECT CAST(value AS DOUBLE) AS a FROM range(0, 721))
TO 'test_files/scratch/parquet_statistics/deg.parquet' STORED AS PARQUET;

statement ok
CREATE EXTERNAL TABLE deg STORED AS PARQUET
LOCATION 'test_files/scratch/parquet_statistics/deg.parquet';

statement ok
COPY (SELECT radians(CAST(value AS DOUBLE)) AS x FROM range(300, 601))
TO 'test_files/scratch/parquet_statistics/rad.parquet' STORED AS PARQUET;

statement ok
CREATE EXTERNAL TABLE rad STORED AS PARQUET
LOCATION 'test_files/scratch/parquet_statistics/rad.parquet';

statement ok
SET datafusion.explain.show_statistics = true;

query TT
EXPLAIN SELECT * FROM (SELECT radians(a) AS r FROM deg) s JOIN rad ON s.r = rad.x;
HashJoinExec: ..., statistics=[Rows=Inexact(0), ...
  DataSourceExec: ..., projection=[radians(a@0) as r], ... Min=Inexact(Float64(-3.1415926535897936)) Max=Inexact(Float64(3.1415926535897936))

Computing x * pi / 180 from the input interval in radians_bounds should fix it. The other functions in bounds.rs look fine.

});
column_statistics[idx].max_value = key_stats.max_value;
column_statistics[idx].min_value = key_stats.min_value;
column_statistics[idx].distinct_count = key_stats.distinct_count;

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The Partial output does not keep null_count, so the Final aggregate adds a NULL group that Single mode does not add. A nullable key with no NULLs in [1, 7] gets 8 rows with 2 partitions and 7 rows with 1 partition. The query returns 7.

statement ok
COPY (SELECT CASE WHEN value < 0 THEN NULL ELSE CAST(value % 7 + 1 AS INT) END AS k FROM range(0, 10000))
TO 'test_files/scratch/parquet_statistics/k.parquet' STORED AS PARQUET;

statement ok
CREATE EXTERNAL TABLE kt STORED AS PARQUET
LOCATION 'test_files/scratch/parquet_statistics/k.parquet';

statement ok
SET datafusion.explain.show_statistics = true;

statement ok
SET datafusion.execution.target_partitions = 2;

# FinalPartitioned: Rows=Inexact(8)
query TT
EXPLAIN SELECT k, count(*) FROM kt GROUP BY k;

statement ok
SET datafusion.execution.target_partitions = 1;

# Single: Rows=Inexact(7)
query TT
EXPLAIN SELECT k, count(*) FROM kt GROUP BY k;

Copying null_count for a plain GROUP BY (not for grouping sets) should fix it.

DROP TABLE literal_ndv_empty;

# The years of the date bounds limit the groups of the final aggregate. The
# key is nullable and its NULL count is unknown, so one NULL group is added to

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

d has Null=Exact(0), so the NULL count is known here. The extra group comes from expression_bounds_statistics dropping null_count. Could date_part keep the input NULL count? Otherwise, maybe update this comment.

stats: &ColumnStatistics,
input_schema: &Schema,
) -> Option<usize> {
if let Some(&ndv) = stats.distinct_count.get_value() {

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For an integer key, the range size is an upper bound for the NDV. With bounds [1, 7] and an Inexact NDV of 1000, this returns 1000. Maybe use the smaller of the two?

match (ndv, limit) {
match (group_count, limit) {
(Some(n), Some(l)) => Precision::Inexact(n.min(l)),
(Some(n), None) => Precision::Inexact(n),

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

When the input row count is unknown, this now returns the range size as the estimate, for example Inexact(2147483647) for bounds [1, i32::MAX]. Before, it returned Absent. The range size is only an upper bound, so I would use it only to cap a known row count.

- Compute radians bounds from the input bounds instead of (-pi, pi).
- Keep whether a plain GROUP BY key has NULLs in the aggregate output.
- Cap the NDV of an integer key by its range.
- Use a range without NDV only to cap a known input row count.

@asolimando asolimando left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @kumarUjjawal, all my comments have been addressed, LGTM! (cc: @gabotechs)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

core Core DataFusion crate functions Changes to functions implementation physical-expr Changes to the physical-expr crates physical-plan Changes to the physical-plan crate sqllogictest SQL Logic Tests (.slt)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants