Skip to content

mem: Cleanup resources of done streams immediately#22064

Open
EmilyMatt wants to merge 6 commits intoapache:mainfrom
EmilyMatt:resource-cleanup
Open

mem: Cleanup resources of done streams immediately#22064
EmilyMatt wants to merge 6 commits intoapache:mainfrom
EmilyMatt:resource-cleanup

Conversation

@EmilyMatt
Copy link
Copy Markdown
Contributor

Which issue does this PR close?

Rationale for this change

Reduces memory pressure, cleans up resources eagerly, and makes pools aware that operators are done by dropping their MemoryReservation and MemoryConsumers.

What changes are included in this PR?

Whenever a stream is polled and returns None(is depleted), drops that stream, or replaces it with EmptyRecordBatchStream.

Are these changes tested?

This should have no effect on logic, as the streams are already depleted.

Are there any user-facing changes?

No, users implementing their own memory pool can expect to see the consumer count drop whenever a stream is released, but that is well within parameters and I don't think is considered a change at all.

@github-actions github-actions Bot added the physical-plan Changes to the physical-plan crate label May 7, 2026
Comment on lines 1621 to 1622
// No data at all — go directly to Done
self.left_exhausted = true;
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.

you forgot to update this as well

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

No bc above it can be converted back to false.
I only placed it there bc if at the point I placed it it is set to true, it means the input is actually depleted.

Comment on lines +1614 to +1618
// If the left stream is fully exhausted, release its resources so the
// upstream pipeline can be torn down before we move on to probing.
if self.left_exhausted {
active.left_stream = None;
}
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.

maybe there are cases that left exhausted is true and left stream is not None, because what is the reason to have both exhausted flag and Option

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Yes replied above

Comment on lines +285 to +287
// Release the input pipeline's resources.
this.inner =
Box::pin(EmptyRecordBatchStream::new(Arc::clone(&this.schema)));
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.

I think to avoid bugs the schema for inner should be kept the same and not use this.schema (even when they are the same schema) because you only want to release the stream

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Yeah fair point, applied

Copy link
Copy Markdown
Member

@rluvaton rluvaton left a comment

Choose a reason for hiding this comment

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

Can you please do the same for ObservedStream, BatchSplitStream, RecordBatchStreamAdapter, ReservationStream and more stream in that file

@EmilyMatt
Copy link
Copy Markdown
Contributor Author

Can you please do the same for ObservedStream, BatchSplitStream, RecordBatchStreamAdapter, ReservationStream and more stream in that file

Applied

Copy link
Copy Markdown
Member

@Weijun-H Weijun-H left a comment

Choose a reason for hiding this comment

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

LGTM! Thanks @EmilyMatt 👍

Comment on lines +1531 to +1533
// Release the right input pipeline's resources.
let right_schema = self.right_stream.schema();
self.right_stream = Box::pin(EmptyRecordBatchStream::new(right_schema));
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.

nit:
extract as a helper function

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

sg, applied

@rluvaton
Copy link
Copy Markdown
Member

rluvaton commented May 7, 2026

run benchmarks

@adriangbot
Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4399405101-2057-vj525 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 resource-cleanup (e2e4372) to 3b634aa (merge-base) diff using: tpcds
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-c4399405101-2056-86zhw 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 resource-cleanup (e2e4372) to 3b634aa (merge-base) diff using: clickbench_partitioned
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-c4399405101-2058-stjxc 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 resource-cleanup (e2e4372) to 3b634aa (merge-base) diff using: tpch
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)

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 resource-cleanup
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃               resource-cleanup ┃       Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.92 / 39.93 ±0.99 / 41.71 ms │ 39.36 / 40.58 ±1.08 / 42.43 ms │    no change │
│ QQuery 2  │ 20.81 / 21.47 ±0.63 / 22.44 ms │ 21.00 / 21.26 ±0.25 / 21.74 ms │    no change │
│ QQuery 3  │ 35.69 / 37.40 ±1.51 / 39.30 ms │ 36.15 / 38.12 ±1.60 / 39.78 ms │    no change │
│ QQuery 4  │ 17.86 / 18.56 ±0.65 / 19.80 ms │ 18.22 / 18.44 ±0.13 / 18.61 ms │    no change │
│ QQuery 5  │ 44.73 / 45.79 ±1.18 / 47.43 ms │ 44.22 / 46.94 ±2.22 / 49.58 ms │    no change │
│ QQuery 6  │ 16.88 / 17.38 ±0.85 / 19.07 ms │ 16.88 / 16.98 ±0.08 / 17.06 ms │    no change │
│ QQuery 7  │ 50.82 / 52.38 ±0.91 / 53.66 ms │ 51.41 / 53.16 ±1.66 / 55.97 ms │    no change │
│ QQuery 8  │ 46.53 / 47.04 ±0.36 / 47.50 ms │ 46.69 / 47.29 ±0.59 / 48.17 ms │    no change │
│ QQuery 9  │ 52.59 / 53.99 ±1.59 / 56.09 ms │ 52.56 / 53.25 ±0.49 / 53.91 ms │    no change │
│ QQuery 10 │ 65.31 / 65.81 ±0.60 / 66.90 ms │ 65.42 / 66.09 ±0.70 / 67.27 ms │    no change │
│ QQuery 11 │ 13.92 / 14.26 ±0.22 / 14.58 ms │ 13.95 / 14.23 ±0.21 / 14.53 ms │    no change │
│ QQuery 12 │ 25.76 / 26.02 ±0.22 / 26.32 ms │ 25.83 / 26.50 ±0.87 / 28.20 ms │    no change │
│ QQuery 13 │ 36.40 / 37.65 ±1.42 / 40.23 ms │ 36.14 / 36.81 ±0.65 / 38.01 ms │    no change │
│ QQuery 14 │ 26.41 / 26.57 ±0.10 / 26.69 ms │ 26.63 / 26.88 ±0.16 / 27.12 ms │    no change │
│ QQuery 15 │ 32.39 / 32.58 ±0.13 / 32.72 ms │ 32.62 / 32.99 ±0.51 / 33.96 ms │    no change │
│ QQuery 16 │ 15.14 / 15.31 ±0.15 / 15.55 ms │ 15.70 / 15.86 ±0.19 / 16.21 ms │    no change │
│ QQuery 17 │ 77.61 / 80.33 ±2.28 / 83.73 ms │ 79.15 / 80.57 ±1.08 / 82.21 ms │    no change │
│ QQuery 18 │ 70.32 / 71.56 ±1.35 / 73.57 ms │ 70.44 / 71.67 ±0.90 / 73.23 ms │    no change │
│ QQuery 19 │ 35.89 / 36.11 ±0.17 / 36.36 ms │ 36.50 / 37.92 ±1.55 / 40.83 ms │ 1.05x slower │
│ QQuery 20 │ 38.59 / 38.91 ±0.53 / 39.97 ms │ 38.96 / 39.61 ±0.51 / 40.49 ms │    no change │
│ QQuery 21 │ 61.67 / 63.74 ±2.15 / 66.57 ms │ 60.98 / 63.95 ±3.47 / 70.73 ms │    no change │
│ QQuery 22 │ 24.00 / 24.18 ±0.15 / 24.35 ms │ 24.44 / 24.61 ±0.11 / 24.78 ms │    no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴──────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary               ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)               │ 866.97ms │
│ Total Time (resource-cleanup)   │ 873.71ms │
│ Average Time (HEAD)             │  39.41ms │
│ Average Time (resource-cleanup) │  39.71ms │
│ 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 5.5 GiB
Avg memory 5.0 GiB
CPU user 32.2s
CPU sys 2.3s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 5.5 GiB
Avg memory 5.0 GiB
CPU user 32.2s
CPU sys 2.4s
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)

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 resource-cleanup
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                      resource-cleanup ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           6.21 / 6.81 ±0.87 / 8.53 ms │           6.52 / 7.00 ±0.86 / 8.72 ms │     no change │
│ QQuery 2  │        82.97 / 83.26 ±0.25 / 83.56 ms │        83.12 / 83.36 ±0.20 / 83.70 ms │     no change │
│ QQuery 3  │        29.47 / 29.84 ±0.21 / 30.13 ms │        29.36 / 29.72 ±0.24 / 30.00 ms │     no change │
│ QQuery 4  │     525.20 / 528.17 ±2.94 / 532.95 ms │     521.76 / 525.18 ±4.75 / 534.49 ms │     no change │
│ QQuery 5  │        52.66 / 53.17 ±0.44 / 53.99 ms │        52.72 / 53.39 ±0.48 / 54.18 ms │     no change │
│ QQuery 6  │        36.35 / 36.61 ±0.26 / 37.10 ms │        36.23 / 36.54 ±0.22 / 36.76 ms │     no change │
│ QQuery 7  │     111.43 / 115.10 ±4.43 / 122.68 ms │     110.73 / 114.08 ±4.04 / 121.92 ms │     no change │
│ QQuery 8  │        39.57 / 39.70 ±0.13 / 39.93 ms │        38.96 / 39.56 ±0.44 / 40.05 ms │     no change │
│ QQuery 9  │        55.38 / 57.24 ±2.55 / 62.27 ms │        53.06 / 55.30 ±1.39 / 57.26 ms │     no change │
│ QQuery 10 │        81.48 / 82.72 ±1.61 / 85.87 ms │        82.41 / 83.58 ±1.39 / 86.23 ms │     no change │
│ QQuery 11 │     324.86 / 327.50 ±2.60 / 332.25 ms │     323.14 / 325.62 ±2.89 / 330.88 ms │     no change │
│ QQuery 12 │        29.20 / 29.53 ±0.27 / 29.84 ms │        29.09 / 29.64 ±0.33 / 30.08 ms │     no change │
│ QQuery 13 │     129.08 / 129.53 ±0.52 / 130.32 ms │     128.95 / 130.22 ±0.75 / 131.03 ms │     no change │
│ QQuery 14 │     522.34 / 523.81 ±1.60 / 526.77 ms │    516.10 / 524.54 ±12.07 / 548.28 ms │     no change │
│ QQuery 15 │        61.57 / 62.40 ±1.49 / 65.37 ms │        60.56 / 62.30 ±1.34 / 64.59 ms │     no change │
│ QQuery 16 │           6.88 / 7.12 ±0.19 / 7.42 ms │           7.12 / 7.29 ±0.18 / 7.63 ms │     no change │
│ QQuery 17 │        84.96 / 85.29 ±0.26 / 85.69 ms │        84.37 / 85.47 ±0.95 / 87.23 ms │     no change │
│ QQuery 18 │     156.95 / 157.51 ±0.43 / 158.11 ms │     157.45 / 158.09 ±0.57 / 158.90 ms │     no change │
│ QQuery 19 │        41.80 / 42.11 ±0.19 / 42.28 ms │        41.57 / 41.83 ±0.25 / 42.25 ms │     no change │
│ QQuery 20 │        35.70 / 36.16 ±0.31 / 36.48 ms │        35.44 / 35.99 ±0.38 / 36.51 ms │     no change │
│ QQuery 21 │        18.27 / 18.56 ±0.26 / 18.96 ms │        18.29 / 18.50 ±0.26 / 19.00 ms │     no change │
│ QQuery 22 │        62.59 / 63.72 ±1.13 / 65.61 ms │        62.11 / 62.61 ±0.40 / 63.27 ms │     no change │
│ QQuery 23 │     486.31 / 489.90 ±2.73 / 492.89 ms │     487.46 / 491.76 ±6.31 / 504.23 ms │     no change │
│ QQuery 24 │     245.34 / 248.94 ±2.97 / 253.57 ms │     242.90 / 247.81 ±3.89 / 254.44 ms │     no change │
│ QQuery 25 │     119.37 / 121.36 ±2.16 / 124.72 ms │     117.72 / 119.44 ±1.32 / 121.78 ms │     no change │
│ QQuery 26 │        71.60 / 72.15 ±0.54 / 73.13 ms │        72.85 / 73.29 ±0.41 / 74.01 ms │     no change │
│ QQuery 27 │           7.03 / 7.17 ±0.20 / 7.55 ms │           7.07 / 7.19 ±0.14 / 7.46 ms │     no change │
│ QQuery 28 │        58.57 / 63.16 ±2.75 / 67.24 ms │        58.85 / 61.63 ±2.20 / 63.94 ms │     no change │
│ QQuery 29 │     103.14 / 105.77 ±1.87 / 108.61 ms │     102.77 / 104.07 ±1.17 / 105.52 ms │     no change │
│ QQuery 30 │                                  FAIL │                                  FAIL │  incomparable │
│ QQuery 31 │     114.05 / 115.28 ±1.55 / 118.32 ms │     112.30 / 113.49 ±0.66 / 114.14 ms │     no change │
│ QQuery 32 │        20.64 / 20.89 ±0.24 / 21.34 ms │        21.01 / 21.26 ±0.29 / 21.76 ms │     no change │
│ QQuery 33 │        39.63 / 40.04 ±0.33 / 40.47 ms │        39.14 / 39.49 ±0.33 / 40.08 ms │     no change │
│ QQuery 34 │         9.89 / 10.42 ±0.38 / 10.91 ms │         9.93 / 10.30 ±0.20 / 10.50 ms │     no change │
│ QQuery 35 │        82.01 / 83.05 ±1.04 / 84.87 ms │        81.66 / 82.82 ±1.18 / 85.06 ms │     no change │
│ QQuery 36 │           6.45 / 6.58 ±0.16 / 6.89 ms │           6.35 / 6.49 ±0.19 / 6.86 ms │     no change │
│ QQuery 37 │           7.18 / 7.31 ±0.14 / 7.51 ms │           7.22 / 7.37 ±0.09 / 7.45 ms │     no change │
│ QQuery 38 │        68.89 / 69.26 ±0.41 / 69.99 ms │        69.17 / 69.76 ±0.43 / 70.35 ms │     no change │
│ QQuery 39 │     102.07 / 103.36 ±1.12 / 104.95 ms │     100.61 / 101.56 ±0.70 / 102.42 ms │     no change │
│ QQuery 40 │        24.13 / 24.52 ±0.32 / 24.93 ms │        24.13 / 24.43 ±0.22 / 24.70 ms │     no change │
│ QQuery 41 │        14.22 / 14.35 ±0.20 / 14.75 ms │        14.23 / 14.39 ±0.16 / 14.68 ms │     no change │
│ QQuery 42 │        24.36 / 24.89 ±0.35 / 25.48 ms │        24.11 / 24.33 ±0.23 / 24.76 ms │     no change │
│ QQuery 43 │           5.34 / 5.42 ±0.14 / 5.69 ms │           5.37 / 5.48 ±0.16 / 5.78 ms │     no change │
│ QQuery 44 │        10.83 / 10.95 ±0.10 / 11.11 ms │        10.96 / 11.04 ±0.10 / 11.22 ms │     no change │
│ QQuery 45 │        41.07 / 41.91 ±0.45 / 42.36 ms │        40.77 / 41.67 ±0.68 / 42.38 ms │     no change │
│ QQuery 46 │        13.62 / 14.08 ±0.63 / 15.33 ms │        13.40 / 13.74 ±0.23 / 14.07 ms │     no change │
│ QQuery 47 │     237.91 / 240.06 ±1.57 / 242.73 ms │     232.36 / 234.48 ±1.91 / 237.34 ms │     no change │
│ QQuery 48 │     104.61 / 105.26 ±0.46 / 105.92 ms │     104.72 / 105.65 ±0.73 / 106.84 ms │     no change │
│ QQuery 49 │        82.90 / 84.45 ±1.48 / 87.25 ms │        82.74 / 83.85 ±1.29 / 86.35 ms │     no change │
│ QQuery 50 │        63.69 / 65.10 ±1.77 / 68.49 ms │        63.12 / 64.86 ±1.12 / 66.10 ms │     no change │
│ QQuery 51 │        94.91 / 96.57 ±1.23 / 98.15 ms │        91.37 / 94.25 ±2.36 / 98.34 ms │     no change │
│ QQuery 52 │        24.77 / 25.64 ±1.18 / 27.96 ms │        24.73 / 24.85 ±0.13 / 25.09 ms │     no change │
│ QQuery 53 │        30.85 / 31.26 ±0.29 / 31.75 ms │        30.49 / 31.28 ±1.25 / 33.75 ms │     no change │
│ QQuery 54 │        54.60 / 55.17 ±0.41 / 55.66 ms │        55.29 / 55.89 ±0.79 / 57.37 ms │     no change │
│ QQuery 55 │        24.11 / 24.27 ±0.20 / 24.65 ms │        23.97 / 24.19 ±0.19 / 24.42 ms │     no change │
│ QQuery 56 │        40.41 / 41.10 ±1.01 / 43.09 ms │        39.87 / 40.19 ±0.18 / 40.42 ms │     no change │
│ QQuery 57 │     180.12 / 181.66 ±0.95 / 182.81 ms │     178.38 / 178.75 ±0.38 / 179.45 ms │     no change │
│ QQuery 58 │     119.30 / 120.20 ±0.95 / 121.93 ms │     118.12 / 119.19 ±0.79 / 120.35 ms │     no change │
│ QQuery 59 │     120.01 / 120.61 ±0.41 / 121.24 ms │     119.22 / 119.91 ±0.40 / 120.43 ms │     no change │
│ QQuery 60 │        40.74 / 41.45 ±0.54 / 42.11 ms │        40.25 / 40.85 ±0.82 / 42.45 ms │     no change │
│ QQuery 61 │        14.03 / 14.25 ±0.21 / 14.57 ms │        13.93 / 14.08 ±0.16 / 14.36 ms │     no change │
│ QQuery 62 │        47.28 / 48.24 ±1.03 / 50.19 ms │        46.71 / 47.38 ±0.84 / 49.04 ms │     no change │
│ QQuery 63 │        30.76 / 31.10 ±0.31 / 31.50 ms │        30.98 / 31.27 ±0.28 / 31.75 ms │     no change │
│ QQuery 64 │     482.64 / 486.08 ±4.43 / 494.65 ms │     477.74 / 481.97 ±3.33 / 486.94 ms │     no change │
│ QQuery 65 │     147.17 / 151.55 ±3.00 / 156.24 ms │     145.67 / 149.19 ±2.44 / 152.09 ms │     no change │
│ QQuery 66 │        83.78 / 84.89 ±0.65 / 85.69 ms │        82.75 / 83.36 ±0.49 / 84.04 ms │     no change │
│ QQuery 67 │     248.80 / 254.12 ±5.23 / 262.93 ms │     246.31 / 251.73 ±4.12 / 258.91 ms │     no change │
│ QQuery 68 │        13.82 / 14.01 ±0.17 / 14.24 ms │        13.89 / 14.21 ±0.32 / 14.73 ms │     no change │
│ QQuery 69 │        76.91 / 77.40 ±0.41 / 78.06 ms │        77.87 / 78.27 ±0.32 / 78.66 ms │     no change │
│ QQuery 70 │     106.65 / 111.08 ±6.72 / 124.36 ms │     105.34 / 108.46 ±3.25 / 114.56 ms │     no change │
│ QQuery 71 │        36.38 / 36.69 ±0.19 / 36.96 ms │        35.97 / 36.44 ±0.51 / 37.42 ms │     no change │
│ QQuery 72 │ 2236.23 / 2294.56 ±59.78 / 2399.06 ms │ 2198.68 / 2245.66 ±52.04 / 2342.27 ms │     no change │
│ QQuery 73 │          9.61 / 9.77 ±0.17 / 10.07 ms │          9.61 / 9.88 ±0.18 / 10.12 ms │     no change │
│ QQuery 74 │     185.03 / 187.42 ±2.31 / 190.79 ms │     181.68 / 184.21 ±1.95 / 186.89 ms │     no change │
│ QQuery 75 │     151.66 / 152.98 ±1.68 / 156.18 ms │     150.31 / 152.30 ±1.52 / 155.01 ms │     no change │
│ QQuery 76 │        36.00 / 37.08 ±1.56 / 40.16 ms │        35.85 / 36.18 ±0.23 / 36.44 ms │     no change │
│ QQuery 77 │        63.56 / 64.43 ±1.12 / 66.45 ms │        62.63 / 62.92 ±0.27 / 63.39 ms │     no change │
│ QQuery 78 │     197.74 / 200.54 ±1.80 / 202.40 ms │     194.34 / 198.22 ±2.28 / 201.33 ms │     no change │
│ QQuery 79 │        67.95 / 68.25 ±0.26 / 68.68 ms │        68.06 / 68.90 ±1.15 / 71.14 ms │     no change │
│ QQuery 80 │     104.21 / 106.66 ±1.72 / 109.53 ms │     104.22 / 105.26 ±0.69 / 106.37 ms │     no change │
│ QQuery 81 │        24.65 / 26.14 ±1.97 / 30.03 ms │        24.58 / 24.79 ±0.16 / 25.05 ms │ +1.05x faster │
│ QQuery 82 │        16.95 / 17.37 ±0.38 / 17.89 ms │        16.89 / 17.71 ±1.42 / 20.53 ms │     no change │
│ QQuery 83 │        38.51 / 38.95 ±0.42 / 39.58 ms │        38.06 / 38.32 ±0.28 / 38.83 ms │     no change │
│ QQuery 84 │        43.50 / 43.78 ±0.19 / 44.09 ms │        43.26 / 44.40 ±1.40 / 47.12 ms │     no change │
│ QQuery 85 │     138.67 / 139.43 ±1.01 / 141.42 ms │     138.42 / 139.40 ±1.16 / 141.66 ms │     no change │
│ QQuery 86 │        25.59 / 26.27 ±0.61 / 27.41 ms │        24.98 / 25.49 ±0.54 / 26.44 ms │     no change │
│ QQuery 87 │        69.52 / 70.86 ±0.93 / 72.16 ms │        69.16 / 70.55 ±1.94 / 74.36 ms │     no change │
│ QQuery 88 │        65.60 / 67.09 ±2.09 / 71.24 ms │        65.38 / 65.49 ±0.08 / 65.58 ms │     no change │
│ QQuery 89 │        36.59 / 37.15 ±0.33 / 37.50 ms │        36.38 / 36.98 ±0.67 / 38.28 ms │     no change │
│ QQuery 90 │        17.90 / 17.99 ±0.07 / 18.10 ms │        17.65 / 17.87 ±0.13 / 18.00 ms │     no change │
│ QQuery 91 │        51.75 / 52.57 ±0.49 / 53.05 ms │        52.54 / 53.12 ±0.53 / 54.11 ms │     no change │
│ QQuery 92 │        30.06 / 30.73 ±1.00 / 32.65 ms │        29.49 / 29.69 ±0.33 / 30.34 ms │     no change │
│ QQuery 93 │        54.59 / 55.57 ±0.50 / 55.93 ms │        51.45 / 53.73 ±1.60 / 56.01 ms │     no change │
│ QQuery 94 │        39.20 / 39.51 ±0.20 / 39.73 ms │        37.90 / 38.45 ±0.31 / 38.81 ms │     no change │
│ QQuery 95 │        92.13 / 93.02 ±0.66 / 93.67 ms │        89.37 / 90.99 ±0.96 / 92.18 ms │     no change │
│ QQuery 96 │        24.73 / 24.88 ±0.14 / 25.10 ms │        24.36 / 24.63 ±0.33 / 25.29 ms │     no change │
│ QQuery 97 │        47.62 / 47.88 ±0.25 / 48.34 ms │        46.92 / 47.41 ±0.30 / 47.86 ms │     no change │
│ QQuery 98 │        42.86 / 43.74 ±0.83 / 45.21 ms │        42.07 / 42.68 ±0.36 / 43.15 ms │     no change │
│ QQuery 99 │        70.90 / 71.20 ±0.44 / 72.06 ms │        69.30 / 70.03 ±0.47 / 70.62 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary               ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)               │ 10824.63ms │
│ Total Time (resource-cleanup)   │ 10712.46ms │
│ Average Time (HEAD)             │   110.46ms │
│ Average Time (resource-cleanup) │   109.31ms │
│ Queries Faster                  │          1 │
│ Queries Slower                  │          0 │
│ Queries with No Change          │         97 │
│ Queries with Failure            │          1 │
└─────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 6.9 GiB
Avg memory 6.1 GiB
CPU user 250.9s
CPU sys 6.6s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 6.9 GiB
Avg memory 6.2 GiB
CPU user 246.5s
CPU sys 6.3s
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)

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 resource-cleanup
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃                      resource-cleanup ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.22 / 4.71 ±6.84 / 18.39 ms │          1.20 / 4.73 ±6.90 / 18.52 ms │     no change │
│ QQuery 1  │        13.25 / 13.35 ±0.08 / 13.45 ms │        12.88 / 13.33 ±0.23 / 13.52 ms │     no change │
│ QQuery 2  │        36.32 / 36.65 ±0.48 / 37.61 ms │        36.18 / 36.49 ±0.26 / 36.88 ms │     no change │
│ QQuery 3  │        30.72 / 31.33 ±0.73 / 32.72 ms │        31.05 / 31.69 ±0.38 / 32.18 ms │     no change │
│ QQuery 4  │     233.40 / 235.96 ±1.81 / 237.72 ms │     242.80 / 246.02 ±2.17 / 248.78 ms │     no change │
│ QQuery 5  │     278.83 / 281.37 ±2.18 / 284.68 ms │     280.69 / 287.63 ±5.97 / 296.47 ms │     no change │
│ QQuery 6  │           6.81 / 7.54 ±0.58 / 8.43 ms │          6.77 / 8.27 ±1.57 / 11.15 ms │  1.10x slower │
│ QQuery 7  │        14.42 / 14.54 ±0.11 / 14.73 ms │        14.59 / 14.74 ±0.10 / 14.90 ms │     no change │
│ QQuery 8  │     315.64 / 318.55 ±3.14 / 324.14 ms │     323.32 / 327.94 ±4.05 / 334.63 ms │     no change │
│ QQuery 9  │     455.23 / 458.04 ±1.48 / 459.37 ms │     460.05 / 464.65 ±5.40 / 474.37 ms │     no change │
│ QQuery 10 │        69.44 / 70.23 ±0.47 / 70.68 ms │        70.43 / 75.40 ±8.10 / 91.55 ms │  1.07x slower │
│ QQuery 11 │        80.64 / 81.46 ±0.54 / 82.28 ms │        81.10 / 82.71 ±2.14 / 86.92 ms │     no change │
│ QQuery 12 │     268.64 / 272.62 ±3.87 / 278.06 ms │     274.93 / 278.69 ±4.14 / 286.75 ms │     no change │
│ QQuery 13 │     375.90 / 383.75 ±7.09 / 393.76 ms │     387.07 / 392.65 ±7.35 / 406.95 ms │     no change │
│ QQuery 14 │     275.74 / 278.05 ±2.17 / 281.98 ms │     283.41 / 288.26 ±5.02 / 297.62 ms │     no change │
│ QQuery 15 │     273.71 / 278.78 ±7.48 / 293.60 ms │     276.98 / 280.99 ±3.34 / 285.78 ms │     no change │
│ QQuery 16 │     600.48 / 604.20 ±5.36 / 614.76 ms │     600.67 / 609.86 ±7.78 / 619.91 ms │     no change │
│ QQuery 17 │     594.74 / 603.87 ±5.85 / 610.77 ms │     599.53 / 611.66 ±8.26 / 624.04 ms │     no change │
│ QQuery 18 │ 1179.10 / 1199.01 ±13.17 / 1220.06 ms │  1205.85 / 1212.48 ±5.21 / 1218.14 ms │     no change │
│ QQuery 19 │        27.80 / 30.19 ±2.81 / 35.25 ms │       28.18 / 40.61 ±15.18 / 63.39 ms │  1.34x slower │
│ QQuery 20 │     517.17 / 526.14 ±7.50 / 536.97 ms │     517.38 / 522.16 ±4.39 / 528.87 ms │     no change │
│ QQuery 21 │     589.75 / 592.94 ±2.97 / 597.52 ms │     589.76 / 595.52 ±3.56 / 600.01 ms │     no change │
│ QQuery 22 │  1072.16 / 1075.62 ±3.89 / 1082.94 ms │ 1054.62 / 1073.87 ±11.57 / 1087.26 ms │     no change │
│ QQuery 23 │ 3155.05 / 3185.21 ±33.93 / 3249.97 ms │ 3198.35 / 3218.28 ±12.97 / 3238.82 ms │     no change │
│ QQuery 24 │       42.57 / 49.54 ±11.15 / 71.51 ms │        41.51 / 42.43 ±0.83 / 43.83 ms │ +1.17x faster │
│ QQuery 25 │     112.19 / 113.01 ±0.59 / 113.62 ms │     112.14 / 113.39 ±1.24 / 115.75 ms │     no change │
│ QQuery 26 │        41.93 / 44.03 ±2.17 / 48.23 ms │        42.14 / 42.68 ±0.53 / 43.69 ms │     no change │
│ QQuery 27 │     671.25 / 674.57 ±1.79 / 676.49 ms │     668.69 / 673.55 ±4.58 / 680.37 ms │     no change │
│ QQuery 28 │ 3037.84 / 3067.21 ±20.15 / 3100.71 ms │ 3012.34 / 3025.06 ±10.93 / 3041.40 ms │     no change │
│ QQuery 29 │        42.40 / 43.85 ±2.46 / 48.76 ms │        41.62 / 46.15 ±8.87 / 63.89 ms │  1.05x slower │
│ QQuery 30 │     313.44 / 318.81 ±5.09 / 326.76 ms │     299.65 / 304.14 ±5.39 / 314.49 ms │     no change │
│ QQuery 31 │     299.05 / 307.61 ±8.13 / 321.97 ms │     287.36 / 294.29 ±5.56 / 303.30 ms │     no change │
│ QQuery 32 │     970.06 / 976.44 ±6.58 / 988.32 ms │    918.26 / 929.32 ±10.57 / 943.15 ms │     no change │
│ QQuery 33 │ 1473.13 / 1497.04 ±14.77 / 1511.12 ms │  1426.95 / 1440.01 ±9.15 / 1450.10 ms │     no change │
│ QQuery 34 │ 1487.86 / 1512.24 ±17.04 / 1540.15 ms │ 1423.97 / 1451.69 ±15.01 / 1466.80 ms │     no change │
│ QQuery 35 │    300.31 / 321.78 ±31.78 / 384.53 ms │    283.14 / 316.45 ±37.46 / 386.12 ms │     no change │
│ QQuery 36 │        65.03 / 71.01 ±6.48 / 83.26 ms │        62.31 / 66.95 ±5.75 / 78.21 ms │ +1.06x faster │
│ QQuery 37 │        36.71 / 38.16 ±2.33 / 42.81 ms │        36.32 / 38.30 ±2.00 / 41.46 ms │     no change │
│ QQuery 38 │        42.38 / 46.01 ±4.38 / 54.08 ms │        39.88 / 42.11 ±1.29 / 43.35 ms │ +1.09x faster │
│ QQuery 39 │     124.58 / 134.84 ±8.90 / 146.93 ms │     125.99 / 130.03 ±2.94 / 134.24 ms │     no change │
│ QQuery 40 │        14.94 / 15.17 ±0.23 / 15.58 ms │        14.48 / 16.73 ±3.11 / 22.80 ms │  1.10x slower │
│ QQuery 41 │        14.55 / 14.80 ±0.14 / 14.98 ms │        14.26 / 16.41 ±4.17 / 24.74 ms │  1.11x slower │
│ QQuery 42 │        14.37 / 15.65 ±2.46 / 20.58 ms │        13.12 / 13.25 ±0.12 / 13.42 ms │ +1.18x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary               ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)               │ 19845.87ms │
│ Total Time (resource-cleanup)   │ 19721.57ms │
│ Average Time (HEAD)             │   461.53ms │
│ Average Time (resource-cleanup) │   458.64ms │
│ Queries Faster                  │          4 │
│ Queries Slower                  │          6 │
│ Queries with No Change          │         33 │
│ Queries with Failure            │          0 │
└─────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 30.7 GiB
Avg memory 23.3 GiB
CPU user 1048.4s
CPU sys 62.0s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 100.0s
Peak memory 30.5 GiB
Avg memory 23.3 GiB
CPU user 1037.8s
CPU sys 62.3s
Peak spill 0 B

File an issue against this benchmark runner

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

Labels

physical-plan Changes to the physical-plan crate

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Cleanup resources of finished streams immediately.

4 participants