Skip to content

Split Parquet tail work into morsels#22924

Draft
Dandandan wants to merge 2 commits into
apache:mainfrom
Dandandan:codex/parquet-morsel-splitting
Draft

Split Parquet tail work into morsels#22924
Dandandan wants to merge 2 commits into
apache:mainfrom
Dandandan:codex/parquet-morsel-splitting

Conversation

@Dandandan

@Dandandan Dandandan commented Jun 12, 2026

Copy link
Copy Markdown
Contributor

Which issue does this PR close?

N/A.

Rationale for this change

This change lets Parquet split remaining large row groups into smaller, page-aligned morsels when sibling streams would otherwise go idle.

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

Yes. A new Parquet read configuration option, datafusion.execution.parquet.morsel_split_size, controls the target compressed byte size for tail-work morsel splitting. Setting it to NULL disables splitting.

@github-actions github-actions Bot added documentation Improvements or additions to documentation sqllogictest SQL Logic Tests (.slt) common Related to common crate proto Related to proto crate datasource Changes to the datasource crate labels Jun 12, 2026
@Dandandan Dandandan changed the title [codex] Split Parquet tail work into morsels Split Parquet tail work into morsels Jun 12, 2026
@Dandandan

Copy link
Copy Markdown
Contributor Author

run benchmarks

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4688328065-546-bvqff 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (740db64) to 883c38e (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 running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4688328065-544-xtgnf 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (740db64) to 883c38e (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-c4688328065-545-lqsqr 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (740db64) to 883c38e (merge-base) diff using: tpcds
Results will be posted here when complete


File an issue against this benchmark runner

@github-actions

github-actions Bot commented Jun 12, 2026

Copy link
Copy Markdown

Thank you for opening this pull request!

Reviewer note: cargo-semver-checks reported the current version number is not SemVer-compatible with the changes in this pull request (compared against the base branch).

Details
     Cloning apache/main
    Building datafusion-common v54.0.0 (current)
       Built [  31.306s] (current)
     Parsing datafusion-common v54.0.0 (current)
      Parsed [   0.065s] (current)
    Building datafusion-common v54.0.0 (baseline)
       Built [  31.291s] (baseline)
     Parsing datafusion-common v54.0.0 (baseline)
      Parsed [   0.056s] (baseline)
    Checking datafusion-common v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   0.807s] 223 checks: 222 pass, 1 fail, 0 warn, 30 skip

--- failure constructible_struct_adds_field: externally-constructible struct adds field ---

Description:
A pub struct constructible with a struct literal has a new pub field. Existing struct literals must be updated to include the new field.
        ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.48.0/src/lints/constructible_struct_adds_field.ron

Failed in:
  field ParquetOptions.morsel_split_size in /home/runner/work/datafusion/datafusion/datafusion/common/src/config.rs:1010

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [  64.966s] datafusion-common
    Building datafusion-datasource v54.0.0 (current)
       Built [  34.700s] (current)
     Parsing datafusion-datasource v54.0.0 (current)
      Parsed [   0.030s] (current)
    Building datafusion-datasource v54.0.0 (baseline)
       Built [  34.422s] (baseline)
     Parsing datafusion-datasource v54.0.0 (baseline)
      Parsed [   0.029s] (baseline)
    Checking datafusion-datasource v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   0.306s] 223 checks: 223 pass, 30 skip
     Summary no semver update required
    Finished [  71.095s] datafusion-datasource
    Building datafusion-datasource-parquet v54.0.0 (current)
       Built [  40.541s] (current)
     Parsing datafusion-datasource-parquet v54.0.0 (current)
      Parsed [   0.027s] (current)
    Building datafusion-datasource-parquet v54.0.0 (baseline)
       Built [  40.305s] (baseline)
     Parsing datafusion-datasource-parquet v54.0.0 (baseline)
      Parsed [   0.027s] (baseline)
    Checking datafusion-datasource-parquet v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   0.166s] 223 checks: 223 pass, 30 skip
     Summary no semver update required
    Finished [  82.400s] datafusion-datasource-parquet
    Building datafusion-proto v54.0.0 (current)
       Built [  54.058s] (current)
     Parsing datafusion-proto v54.0.0 (current)
      Parsed [   0.017s] (current)
    Building datafusion-proto v54.0.0 (baseline)
       Built [  53.508s] (baseline)
     Parsing datafusion-proto v54.0.0 (baseline)
      Parsed [   0.018s] (baseline)
    Checking datafusion-proto v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   0.287s] 223 checks: 223 pass, 30 skip
     Summary no semver update required
    Finished [ 109.624s] datafusion-proto
    Building datafusion-proto-common v54.0.0 (current)
       Built [  20.018s] (current)
     Parsing datafusion-proto-common v54.0.0 (current)
      Parsed [   0.045s] (current)
    Building datafusion-proto-common v54.0.0 (baseline)
       Built [  19.872s] (baseline)
     Parsing datafusion-proto-common v54.0.0 (baseline)
      Parsed [   0.045s] (baseline)
    Checking datafusion-proto-common v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   1.282s] 223 checks: 222 pass, 1 fail, 0 warn, 30 skip

--- failure constructible_struct_adds_field: externally-constructible struct adds field ---

Description:
A pub struct constructible with a struct literal has a new pub field. Existing struct literals must be updated to include the new field.
        ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.48.0/src/lints/constructible_struct_adds_field.ron

Failed in:
  field ParquetOptions.morsel_split_size_opt in /home/runner/work/datafusion/datafusion/datafusion/proto-common/src/generated/prost.rs:906
  field ParquetOptions.morsel_split_size_opt in /home/runner/work/datafusion/datafusion/datafusion/proto-common/src/generated/prost.rs:906
  field ParquetOptions.morsel_split_size_opt in /home/runner/work/datafusion/datafusion/datafusion/proto-common/src/generated/prost.rs:906

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [  42.681s] datafusion-proto-common
    Building datafusion-proto-models v54.0.0 (current)
       Built [  22.558s] (current)
     Parsing datafusion-proto-models v54.0.0 (current)
      Parsed [   0.118s] (current)
    Building datafusion-proto-models v54.0.0 (baseline)
       Built [  22.349s] (baseline)
     Parsing datafusion-proto-models v54.0.0 (baseline)
      Parsed [   0.117s] (baseline)
    Checking datafusion-proto-models v54.0.0 -> v54.0.0 (no change; assume patch)
     Checked [   1.845s] 223 checks: 222 pass, 1 fail, 0 warn, 30 skip

--- failure constructible_struct_adds_field: externally-constructible struct adds field ---

Description:
A pub struct constructible with a struct literal has a new pub field. Existing struct literals must be updated to include the new field.
        ref: https://doc.rust-lang.org/reference/expressions/struct-expr.html
       impl: https://github.com/obi1kenobi/cargo-semver-checks/tree/v0.48.0/src/lints/constructible_struct_adds_field.ron

Failed in:
  field ParquetOptions.morsel_split_size_opt in /home/runner/work/datafusion/datafusion/datafusion/proto-models/src/generated/datafusion_proto_common.rs:906
  field ParquetOptions.morsel_split_size_opt in /home/runner/work/datafusion/datafusion/datafusion/proto-models/src/generated/datafusion_proto_common.rs:906

     Summary semver requires new major version: 1 major and 0 minor checks failed
    Finished [  48.495s] datafusion-proto-models
    Building datafusion-sqllogictest v54.0.0 (current)
error: running cargo-doc on crate 'datafusion-sqllogictest' failed with output:
-----
   Compiling proc-macro2 v1.0.106
   Compiling quote v1.0.45
   Compiling unicode-ident v1.0.24
   Compiling libc v0.2.186
    Checking cfg-if v1.0.4
    Checking bytes v1.11.1
    Checking memchr v2.8.2
   Compiling serde_core v1.0.228
   Compiling syn v2.0.117
   Compiling autocfg v1.5.1
   Compiling jobserver v0.1.34
   Compiling shlex v2.0.1
   Compiling find-msvc-tools v0.1.9
   Compiling cc v1.2.64
    Checking itoa v1.0.18
    Checking foldhash v0.2.0
    Checking equivalent v1.0.2
    Checking once_cell v1.21.4
    Checking allocator-api2 v0.2.21
   Compiling libm v0.2.16
   Compiling num-traits v0.2.19
    Checking hashbrown v0.17.1
    Checking pin-project-lite v0.2.17
   Compiling zmij v1.0.21
    Checking indexmap v2.14.0
    Checking futures-core v0.3.32
   Compiling zerocopy v0.8.52
    Checking futures-sink v0.3.32
    Checking errno v0.3.14
    Checking socket2 v0.6.4
    Checking signal-hook-registry v1.4.8
    Checking mio v1.2.1
   Compiling version_check v0.9.5
   Compiling serde v1.0.228
   Compiling getrandom v0.3.4
    Checking num-integer v0.1.46
    Checking slab v0.4.12
   Compiling serde_json v1.0.150
    Checking futures-channel v0.3.32
    Checking smallvec v1.15.2
    Checking num-bigint v0.4.6
   Compiling synstructure v0.13.2
    Checking http v1.4.2
    Checking futures-task v0.3.32
    Checking futures-io v0.3.32
    Checking base64 v0.22.1
    Checking iana-time-zone v0.1.65
    Checking chrono v0.4.45
    Checking tracing-core v0.1.36
   Compiling zerocopy-derive v0.8.52
   Compiling tokio-macros v2.7.0
   Compiling serde_derive v1.0.228
    Checking tokio v1.52.3
   Compiling futures-macro v0.3.32
    Checking futures-util v0.3.32
   Compiling zerofrom-derive v0.1.7
   Compiling tracing-attributes v0.1.31
    Checking siphasher v1.0.3
    Checking zerofrom v0.1.8
   Compiling yoke-derive v0.8.2
    Checking stable_deref_trait v1.2.1
    Checking num-complex v0.4.6
   Compiling zerovec-derive v0.11.3
    Checking getrandom v0.2.17
    Checking tracing v0.1.44
   Compiling pkg-config v0.3.33
    Checking half v2.7.1
    Checking yoke v0.8.3
    Checking phf_shared v0.12.1
   Compiling displaydoc v0.2.6
   Compiling ahash v0.8.12
    Checking percent-encoding v2.3.2
   Compiling thiserror v2.0.18
   Compiling chrono-tz v0.10.4
    Checking phf v0.12.1
    Checking zerovec v0.11.6
    Checking arrow-buffer v59.0.0
    Checking ppv-lite86 v0.2.21
   Compiling thiserror-impl v2.0.18
    Checking arrow-schema v59.0.0
   Compiling ring v0.17.14
    Checking arrow-data v59.0.0
   Compiling semver v1.0.28
   Compiling getrandom v0.4.2
    Checking rand_core v0.10.1
    Checking log v0.4.32
    Checking tinystr v0.8.3
    Checking untrusted v0.9.0
    Checking bitflags v2.13.0
    Checking writeable v0.6.3
    Checking litemap v0.8.2
    Checking icu_locale_core v2.2.0
    Checking potential_utf v0.1.5
    Checking zerotrie v0.2.4
   Compiling zstd-sys v2.0.16+zstd.1.5.7
   Compiling async-trait v0.1.89
    Checking rand_core v0.9.5
   Compiling icu_properties_data v2.2.0
    Checking utf8_iter v1.0.4
   Compiling icu_normalizer_data v2.2.0
    Checking icu_collections v2.2.0
    Checking rand_chacha v0.9.0
    Checking icu_provider v2.2.0
    Checking rand v0.9.4
    Checking aho-corasick v1.1.4
    Checking regex-syntax v0.8.11
    Checking arrow-array v59.0.0
   Compiling zstd-safe v7.2.4
    Checking lexical-util v1.0.7
    Checking ryu v1.0.23
   Compiling object v0.37.3
    Checking regex-automata v0.4.14
    Checking arrow-select v59.0.0
    Checking icu_normalizer v2.2.0
    Checking icu_properties v2.2.0
   Compiling rustix v1.1.4
    Checking regex v1.12.4
    Checking idna_adapter v1.2.2
    Checking form_urlencoded v1.2.2
    Checking unicode-width v0.2.2
   Compiling crc32fast v1.5.0
   Compiling parking_lot_core v0.9.12
    Checking typenum v1.20.1
    Checking either v1.16.0
    Checking idna v1.1.0
    Checking lexical-write-integer v1.0.6
    Checking lexical-parse-integer v1.0.6
   Compiling rustc_version v0.4.1
    Checking futures-executor v0.3.32
   Compiling pin-project-internal v1.1.13
    Checking simd-adler32 v0.3.9
    Checking adler2 v2.0.1
    Checking scopeguard v1.2.0
    Checking lock_api v0.4.14
    Checking miniz_oxide v0.8.9
   Compiling flatbuffers v25.12.19
    Checking futures v0.3.32
    Checking lexical-parse-float v1.0.6
    Checking pin-project v1.1.13
    Checking lexical-write-float v1.0.6
    Checking url v2.5.8
    Checking unicode-segmentation v1.13.3
   Compiling ar_archive_writer v0.5.2
    Checking zlib-rs v0.6.3
    Checking comfy-table v7.2.2
    Checking lexical-core v1.0.6
    Checking itertools v0.14.0
   Compiling psm v0.1.31
    Checking flate2 v1.1.9
    Checking arrow-ord v59.0.0
    Checking twox-hash v2.1.2
    Checking atoi v2.0.0
   Compiling stacker v0.1.24
    Checking hex v0.4.3
    Checking alloc-no-stdlib v2.0.4
   Compiling snap v1.1.1
    Checking alloc-stdlib v0.2.2
    Checking arrow-cast v59.0.0
    Checking lz4_flex v0.13.1
    Checking parking_lot v0.12.5
    Checking csv-core v0.1.13
    Checking same-file v1.0.6
    Checking simdutf8 v0.1.5
    Checking humantime v2.3.0
   Compiling paste v1.0.15
    Checking csv v1.4.0
    Checking walkdir v2.5.0
    Checking brotli-decompressor v5.0.1
   Compiling recursive-proc-macro-impl v0.1.1
    Checking subtle v2.6.1
    Checking arrow-csv v59.0.0
    Checking recursive v0.1.1
    Checking arrow-json v59.0.0
    Checking object_store v0.13.2
    Checking brotli v8.0.3
    Checking arrow-string v59.0.0
    Checking arrow-arith v59.0.0
    Checking arrow-row v59.0.0
    Checking zstd v0.13.3
    Checking uuid v1.23.3
    Checking arrow-ipc v59.0.0
    Checking tokio-util v0.7.18
   Compiling sqlparser_derive v0.5.0
   Compiling seq-macro v0.3.6
    Checking sqlparser v0.62.0
    Checking cpufeatures v0.3.0
    Checking hybrid-array v0.4.12
    Checking arrow v59.0.0
    Checking parquet v59.0.0
    Checking linux-raw-sys v0.12.1
    Checking cmov v0.5.4
    Checking crypto-common v0.2.2
    Checking ctutils v0.4.2
    Checking block-buffer v0.12.1
   Compiling generic-array v0.14.7
    Checking const-oid v0.10.2
    Checking digest v0.11.3
    Checking datafusion-doc v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/doc)
    Checking foldhash v0.1.5
   Compiling crossbeam-utils v0.8.21
    Checking hashbrown v0.15.5
    Checking block-buffer v0.10.4
    Checking crypto-common v0.1.7
    Checking fixedbitset v0.5.7
   Compiling heck v0.5.0
    Checking fastrand v2.4.1
    Checking petgraph v0.8.3
    Checking tempfile v3.27.0
    Checking digest v0.10.7
    Checking md-5 v0.11.0
    Checking sha2 v0.11.0
    Checking hashbrown v0.14.5
    Checking dashmap v6.2.1
   Compiling datafusion-macros v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/macros)
   Compiling blake3 v1.8.5
   Compiling anyhow v1.0.102
    Checking constant_time_eq v0.4.2
    Checking arrayvec v0.7.6
    Checking arrayref v0.3.9
    Checking blake2 v0.10.6
   Compiling liblzma-sys v0.4.6
    Checking libbz2-rs-sys v0.2.5
    Checking bzip2 v0.6.1
    Checking datafusion-common-runtime v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/common-runtime)
    Checking compression-core v0.4.32
   Compiling httparse v1.10.1
    Checking http-body v1.0.1
   Compiling prost-derive v0.14.4
    Checking glob v0.3.3
    Checking tower-service v0.3.3
    Checking fnv v1.0.7
    Checking atomic-waker v1.1.2
    Checking try-lock v0.2.5
    Checking want v0.3.1
    Checking h2 v0.4.14
    Checking tokio-stream v0.1.18
    Checking httpdate v1.0.3
    Checking zeroize v1.8.2
    Checking rustls-pki-types v1.14.1
    Checking hyper v1.10.1
    Checking datafusion-common v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/common)
   Compiling prost v0.14.4
    Checking hyper-util v0.1.20
    Checking http-body-util v0.1.3
   Compiling prettyplease v0.2.37
   Compiling rustls v0.23.40
    Checking sync_wrapper v1.0.2
    Checking tower-layer v0.3.3
   Compiling prost-types v0.14.4
    Checking datafusion-expr-common v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/expr-common)
    Checking liblzma v0.4.6
    Checking compression-codecs v0.4.38
    Checking async-compression v0.4.42
    Checking datafusion-physical-expr-common v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/physical-expr-common)
    Checking rustls-webpki v0.103.13
   Compiling serde_derive_internals v0.29.1
    Checking mime v0.3.17
    Checking datafusion-functions-window-common v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions-window-common)
    Checking datafusion-functions-aggregate-common v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions-aggregate-common)
   Compiling schemars v0.8.22
   Compiling schemars_derive v0.8.22
   Compiling hashbrown v0.16.1
    Checking axum-core v0.5.6
    Checking datafusion-expr v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/expr)
    Checking tower v0.5.3
    Checking matchit v0.8.4
   Compiling multimap v0.10.1
   Compiling dyn-clone v1.0.20
    Checking axum v0.8.9
    Checking datafusion-physical-expr v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/physical-expr)
    Checking datafusion-execution v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/execution)
   Compiling prost-build v0.14.4
    Checking datafusion-functions v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions)
   Compiling regress v0.10.5
   Compiling pbjson-build v0.8.0
    Checking hyper-timeout v0.5.2
    Checking chacha20 v0.10.0
   Compiling strsim v0.11.1
   Compiling portable-atomic v1.13.1
   Compiling ident_case v1.0.1
   Compiling darling_core v0.23.0
   Compiling typify-impl v0.5.0
    Checking rand v0.10.1
    Checking datafusion-physical-plan v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/physical-plan)
    Checking datafusion-physical-expr-adapter v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/physical-expr-adapter)
    Checking tonic v0.14.6
   Compiling serde_tokenstream v0.2.3
    Checking ureq-proto v0.6.0
   Compiling bigdecimal v0.4.10
   Compiling bollard-buildkit-proto v0.7.0
    Checking utf8parse v0.2.2
    Checking crc-catalog v2.5.0
    Checking utf8-zero v0.8.1
    Checking tinyvec_macros v0.1.1
    Checking tinyvec v1.11.0
    Checking ureq v3.3.0
    Checking crc v3.4.0
    Checking anstyle-parse v1.0.0
    Checking tonic-prost v0.14.6
   Compiling darling_macro v0.23.0
    Checking datafusion-functions-aggregate v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions-aggregate)
   Compiling strum_macros v0.28.0
   Compiling typify-macro v0.5.0
   Compiling structmeta-derive v0.3.0
    Checking deranged v0.5.8
    Checking anstyle-query v1.1.5
    Checking anstyle v1.0.14
    Checking time-core v0.1.9
    Checking colorchoice v1.0.5
    Checking num-conv v0.2.2
    Checking powerfmt v0.2.0
   Compiling unsafe-libyaml v0.2.11
    Checking is_terminal_polyfill v1.70.2
    Checking anstream v1.0.0
   Compiling structmeta v0.3.0
    Checking time v0.3.48
    Checking datafusion-functions-nested v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions-nested)
   Compiling serde_yaml v0.9.34+deprecated
    Checking datafusion-session v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/session)
    Checking datafusion-datasource v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource)
    Checking arrow-avro v59.0.0
   Compiling typify v0.5.0
   Compiling darling v0.23.0
    Checking datafusion-catalog v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/catalog)
    Checking datafusion-pruning v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/pruning)
    Checking unicode-normalization v0.1.25
   Compiling pbjson-types v0.8.0
    Checking tokio-rustls v0.26.4
   Compiling serde_repr v0.1.20
   Compiling async-stream-impl v0.3.6
    Checking num-rational v0.4.2
    Checking num-iter v0.1.45
    Checking unicode-bidi v0.3.18
    Checking openssl-probe v0.2.1
    Checking unicode-properties v0.1.4
    Checking clap_lex v1.1.0
    Checking clap_builder v4.6.0
    Checking rustls-native-certs v0.8.4
    Checking stringprep v0.1.5
    Checking bollard-stubs v1.52.1-rc.29.1.3
    Checking num v0.4.3
    Checking async-stream v0.3.6
    Checking hyper-rustls v0.27.9
    Checking datafusion-sql v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/sql)
    Checking datafusion-datasource-avro v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource-avro)
    Checking datafusion-datasource-parquet v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource-parquet)
    Checking datafusion-physical-optimizer v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/physical-optimizer)
    Checking datafusion-datasource-json v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource-json)
    Checking datafusion-datasource-arrow v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource-arrow)
    Checking datafusion-datasource-csv v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/datasource-csv)
   Compiling serde_with_macros v3.21.0
   Compiling substrait v0.63.0
   Compiling parse-display-derive v0.9.1
    Checking datafusion-optimizer v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/optimizer)
    Checking datafusion-functions-window v54.0.0 (/home/runner/work/datafusion/datafusion/datafusion/functions-window)
    Checking hyperlocal v0.9.1
   Compiling clap_derive v4.6.1
    Checking xattr v1.6.1
    Checking hmac v0.13.0
    Checking serde_urlencoded v0.7.1
    Checking pbjson v0.8.0
    Checking rand_core v0.6.4
   Compiling enum-ordinalize-derive v4.3.2
   Compiling fs-err v3.3.0
    Checking filetime v0.2.29
    Checking fallible-iterator v0.2.0
   Compiling owo-colors v4.3.0
    Checking byteorder v1.5.0
    Checking home v0.5.12
    Checking rustc-hash v2.1.2
    Checking postgres-protocol v0.6.12
    Checking bollard v0.20.2
    Checking astral-tokio-tar v0.6.2
    Checking clap v4.6.1
error[E0119]: conflicting implementations of trait `From<format_description::parse::format_item::HourBase>` for type `<format_description::parse::format_item::HourBase as format_description::parse::format_item::ModifierValue>::Type`
   --> /home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/bollard-0.20.2/src/docker.rs:209:1
    |
209 | impl<T: Into<String>> From<T> for MaybeClientVersion {
    | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    |
    = note: conflicting implementation in crate `time`:
            - impl From<format_description::parse::format_item::HourBase> for <format_description::parse::format_item::HourBase as format_description::parse::format_item::ModifierValue>::Type;
    = note: upstream crates may add a new impl of trait `std::convert::Into<std::string::String>` for type `time::format_description::parse::format_item::HourBase` in future versions

   Compiling enum-ordinalize v4.3.2
    Checking rand_chacha v0.3.1
For more information about this error, try `rustc --explain E0119`.
error: could not compile `bollard` (lib) due to 1 previous error
warning: build failed, waiting for other jobs to finish...

-----

error: failed to build rustdoc for crate datafusion-sqllogictest v54.0.0
note: this is usually due to a compilation error in the crate,
      and is unlikely to be a bug in cargo-semver-checks
note: the following command can be used to reproduce the error:
      cargo new --lib example &&
          cd example &&
          echo '[workspace]' >> Cargo.toml &&
          cargo add --path /home/runner/work/datafusion/datafusion/datafusion/sqllogictest --features avro,backtrace,bytes,chrono,datafusion-substrait,parquet_encryption,postgres,postgres-types,substrait,testcontainers-modules,tokio-postgres &&
          cargo check &&
          cargo doc

error: aborting due to failure to build rustdoc for crate datafusion-sqllogictest v54.0.0

@github-actions github-actions Bot added the auto detected api change Auto detected API change label Jun 12, 2026
@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 codex_parquet-morsel-splitting
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.05 / 38.86 ±1.06 / 40.86 ms │ 38.08 / 39.40 ±1.79 / 42.81 ms │     no change │
│ QQuery 2  │ 18.46 / 18.68 ±0.14 / 18.88 ms │ 18.43 / 18.67 ±0.27 / 19.18 ms │     no change │
│ QQuery 3  │ 33.26 / 34.03 ±0.99 / 35.89 ms │ 30.33 / 31.88 ±1.38 / 33.54 ms │ +1.07x faster │
│ QQuery 4  │ 16.81 / 17.10 ±0.16 / 17.32 ms │ 17.03 / 17.67 ±0.52 / 18.57 ms │     no change │
│ QQuery 5  │ 36.97 / 38.66 ±1.13 / 40.04 ms │ 39.25 / 39.84 ±0.46 / 40.54 ms │     no change │
│ QQuery 6  │ 15.90 / 16.91 ±1.14 / 18.91 ms │ 16.08 / 16.22 ±0.11 / 16.36 ms │     no change │
│ QQuery 7  │ 43.25 / 45.15 ±1.73 / 47.23 ms │ 44.60 / 45.52 ±1.03 / 47.54 ms │     no change │
│ QQuery 8  │ 42.54 / 42.77 ±0.21 / 43.03 ms │ 42.65 / 42.97 ±0.30 / 43.43 ms │     no change │
│ QQuery 9  │ 48.49 / 49.27 ±0.48 / 49.81 ms │ 51.49 / 52.69 ±1.19 / 54.96 ms │  1.07x slower │
│ QQuery 10 │ 41.58 / 42.28 ±1.06 / 44.39 ms │ 42.27 / 42.78 ±0.51 / 43.69 ms │     no change │
│ QQuery 11 │ 12.67 / 13.21 ±0.77 / 14.74 ms │ 13.06 / 13.27 ±0.17 / 13.58 ms │     no change │
│ QQuery 12 │ 23.82 / 24.17 ±0.30 / 24.58 ms │ 24.59 / 25.20 ±0.55 / 26.16 ms │     no change │
│ QQuery 13 │ 31.92 / 34.56 ±2.50 / 38.59 ms │ 33.08 / 34.79 ±1.24 / 35.91 ms │     no change │
│ QQuery 14 │ 23.55 / 23.84 ±0.15 / 23.98 ms │ 23.48 / 23.66 ±0.20 / 23.97 ms │     no change │
│ QQuery 15 │ 30.95 / 31.39 ±0.44 / 32.06 ms │ 31.57 / 31.83 ±0.21 / 32.20 ms │     no change │
│ QQuery 16 │ 13.71 / 13.82 ±0.10 / 14.02 ms │ 14.02 / 14.13 ±0.09 / 14.25 ms │     no change │
│ QQuery 17 │ 72.33 / 73.87 ±1.51 / 76.65 ms │ 73.92 / 76.68 ±2.90 / 81.88 ms │     no change │
│ QQuery 18 │ 55.93 / 58.48 ±1.38 / 60.08 ms │ 67.06 / 70.08 ±1.96 / 73.14 ms │  1.20x slower │
│ QQuery 19 │ 32.57 / 32.85 ±0.36 / 33.56 ms │ 32.93 / 35.24 ±3.56 / 42.21 ms │  1.07x slower │
│ QQuery 20 │ 31.97 / 32.08 ±0.07 / 32.18 ms │ 32.06 / 32.55 ±0.26 / 32.86 ms │     no change │
│ QQuery 21 │ 54.82 / 57.18 ±1.58 / 58.60 ms │ 60.02 / 61.53 ±1.25 / 63.59 ms │  1.08x slower │
│ QQuery 22 │ 13.68 / 14.14 ±0.47 / 15.00 ms │ 13.49 / 13.72 ±0.15 / 13.91 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                             ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 753.29ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 780.30ms │
│ Average Time (HEAD)                           │  34.24ms │
│ Average Time (codex_parquet-morsel-splitting) │  35.47ms │
│ Queries Faster                                │        1 │
│ Queries Slower                                │        4 │
│ Queries with No Change                        │       17 │
│ Queries with Failure                          │        0 │
└───────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.3 GiB
Avg memory 540.4 MiB
CPU user 22.0s
CPU sys 1.6s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.3 GiB
Avg memory 537.3 MiB
CPU user 23.6s
CPU sys 1.8s
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 codex_parquet-morsel-splitting
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃        codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.74 / 6.25 ±0.88 / 8.01 ms │           5.77 / 6.31 ±0.92 / 8.14 ms │     no change │
│ QQuery 2  │        80.24 / 80.57 ±0.37 / 81.21 ms │        81.61 / 82.01 ±0.32 / 82.58 ms │     no change │
│ QQuery 3  │        29.05 / 29.38 ±0.22 / 29.61 ms │        28.86 / 29.05 ±0.12 / 29.21 ms │     no change │
│ QQuery 4  │     486.85 / 490.97 ±2.74 / 494.05 ms │     495.87 / 503.99 ±5.77 / 512.42 ms │     no change │
│ QQuery 5  │        51.32 / 51.59 ±0.25 / 52.03 ms │        51.84 / 52.04 ±0.16 / 52.30 ms │     no change │
│ QQuery 6  │        36.30 / 36.63 ±0.25 / 36.93 ms │        36.65 / 36.91 ±0.20 / 37.26 ms │     no change │
│ QQuery 7  │        95.22 / 96.12 ±0.47 / 96.57 ms │        95.46 / 95.71 ±0.20 / 96.01 ms │     no change │
│ QQuery 8  │        37.01 / 38.28 ±1.97 / 42.18 ms │        37.25 / 37.50 ±0.24 / 37.95 ms │     no change │
│ QQuery 9  │        52.44 / 53.92 ±0.92 / 55.05 ms │        52.29 / 55.76 ±2.95 / 60.78 ms │     no change │
│ QQuery 10 │        63.00 / 63.33 ±0.18 / 63.54 ms │        63.05 / 63.35 ±0.26 / 63.84 ms │     no change │
│ QQuery 11 │     295.26 / 301.29 ±3.93 / 307.42 ms │     298.64 / 302.39 ±2.60 / 305.37 ms │     no change │
│ QQuery 12 │        28.46 / 28.90 ±0.22 / 29.07 ms │        28.77 / 29.07 ±0.43 / 29.93 ms │     no change │
│ QQuery 13 │     119.27 / 120.68 ±1.01 / 122.37 ms │     115.71 / 116.54 ±0.66 / 117.35 ms │     no change │
│ QQuery 14 │     412.03 / 415.05 ±2.48 / 419.23 ms │     411.21 / 413.56 ±1.30 / 415.19 ms │     no change │
│ QQuery 15 │        58.57 / 59.88 ±0.68 / 60.50 ms │        58.22 / 59.08 ±0.73 / 59.99 ms │     no change │
│ QQuery 16 │           6.68 / 6.83 ±0.22 / 7.27 ms │           6.69 / 6.86 ±0.19 / 7.23 ms │     no change │
│ QQuery 17 │        81.02 / 82.21 ±1.40 / 84.87 ms │        80.11 / 80.69 ±0.31 / 80.93 ms │     no change │
│ QQuery 18 │     123.91 / 126.16 ±1.43 / 128.14 ms │     122.03 / 124.15 ±2.37 / 128.62 ms │     no change │
│ QQuery 19 │        41.61 / 41.95 ±0.32 / 42.53 ms │        41.84 / 42.07 ±0.19 / 42.42 ms │     no change │
│ QQuery 20 │        35.36 / 36.47 ±1.42 / 39.26 ms │        35.28 / 36.51 ±1.22 / 38.86 ms │     no change │
│ QQuery 21 │        17.08 / 17.52 ±0.35 / 18.11 ms │        17.12 / 17.32 ±0.11 / 17.44 ms │     no change │
│ QQuery 22 │        62.35 / 63.55 ±0.96 / 65.16 ms │        62.64 / 63.90 ±1.28 / 66.33 ms │     no change │
│ QQuery 23 │     318.64 / 320.85 ±1.83 / 324.18 ms │     317.92 / 322.84 ±2.90 / 327.03 ms │     no change │
│ QQuery 24 │     225.66 / 228.92 ±3.18 / 234.84 ms │     216.71 / 220.81 ±2.29 / 223.68 ms │     no change │
│ QQuery 25 │     111.47 / 112.98 ±0.84 / 113.67 ms │     110.43 / 112.79 ±2.96 / 118.56 ms │     no change │
│ QQuery 26 │        58.07 / 58.39 ±0.26 / 58.77 ms │        58.10 / 58.44 ±0.28 / 58.83 ms │     no change │
│ QQuery 27 │           6.68 / 6.78 ±0.16 / 7.09 ms │          6.68 / 8.25 ±2.72 / 13.67 ms │  1.22x slower │
│ QQuery 28 │        56.65 / 59.77 ±2.38 / 62.34 ms │        60.79 / 62.95 ±3.43 / 69.77 ms │  1.05x slower │
│ QQuery 29 │       98.36 / 99.03 ±0.96 / 100.88 ms │        97.17 / 97.99 ±0.59 / 98.83 ms │     no change │
│ QQuery 30 │        32.87 / 33.73 ±1.53 / 36.78 ms │        32.58 / 32.92 ±0.22 / 33.18 ms │     no change │
│ QQuery 31 │     111.82 / 113.27 ±1.48 / 115.68 ms │     112.24 / 113.95 ±2.62 / 119.13 ms │     no change │
│ QQuery 32 │        20.00 / 20.40 ±0.24 / 20.70 ms │        20.05 / 20.64 ±0.52 / 21.50 ms │     no change │
│ QQuery 33 │        38.61 / 38.82 ±0.16 / 39.00 ms │        38.67 / 39.06 ±0.30 / 39.56 ms │     no change │
│ QQuery 34 │           9.33 / 9.66 ±0.23 / 9.96 ms │           8.48 / 8.69 ±0.26 / 9.19 ms │ +1.11x faster │
│ QQuery 35 │        71.65 / 74.55 ±2.19 / 78.01 ms │        73.31 / 73.80 ±0.32 / 74.26 ms │     no change │
│ QQuery 36 │           6.12 / 6.23 ±0.15 / 6.53 ms │           6.01 / 6.11 ±0.16 / 6.44 ms │     no change │
│ QQuery 37 │           6.87 / 7.01 ±0.10 / 7.17 ms │           6.81 / 6.97 ±0.12 / 7.17 ms │     no change │
│ QQuery 38 │        63.26 / 64.03 ±0.51 / 64.75 ms │        62.79 / 63.56 ±0.64 / 64.51 ms │     no change │
│ QQuery 39 │     450.49 / 458.91 ±4.48 / 463.62 ms │     453.33 / 461.43 ±8.98 / 478.43 ms │     no change │
│ QQuery 40 │        22.98 / 24.33 ±2.28 / 28.89 ms │        21.27 / 22.62 ±1.24 / 24.67 ms │ +1.08x faster │
│ QQuery 41 │        11.11 / 11.31 ±0.19 / 11.64 ms │        11.28 / 11.46 ±0.25 / 11.94 ms │     no change │
│ QQuery 42 │        23.58 / 24.43 ±0.82 / 25.96 ms │        24.10 / 24.62 ±0.46 / 25.43 ms │     no change │
│ QQuery 43 │           4.80 / 4.95 ±0.20 / 5.33 ms │           4.80 / 4.90 ±0.18 / 5.27 ms │     no change │
│ QQuery 44 │           9.04 / 9.19 ±0.13 / 9.39 ms │           8.76 / 8.93 ±0.12 / 9.12 ms │     no change │
│ QQuery 45 │        38.49 / 39.71 ±1.40 / 42.43 ms │        38.10 / 38.98 ±0.65 / 39.95 ms │     no change │
│ QQuery 46 │        11.58 / 11.69 ±0.06 / 11.75 ms │        10.62 / 10.81 ±0.16 / 11.04 ms │ +1.08x faster │
│ QQuery 47 │     231.06 / 236.68 ±4.96 / 244.84 ms │     229.04 / 232.21 ±2.58 / 236.48 ms │     no change │
│ QQuery 48 │       96.90 / 99.37 ±2.38 / 103.79 ms │        92.18 / 94.24 ±1.37 / 96.43 ms │ +1.05x faster │
│ QQuery 49 │        77.63 / 78.56 ±1.23 / 80.96 ms │        77.56 / 78.65 ±1.75 / 82.12 ms │     no change │
│ QQuery 50 │        59.83 / 60.18 ±0.18 / 60.32 ms │        59.16 / 59.44 ±0.24 / 59.80 ms │     no change │
│ QQuery 51 │        92.79 / 95.04 ±2.69 / 99.85 ms │        91.58 / 93.81 ±1.61 / 96.48 ms │     no change │
│ QQuery 52 │        24.08 / 24.32 ±0.16 / 24.57 ms │        24.26 / 24.71 ±0.41 / 25.49 ms │     no change │
│ QQuery 53 │        29.37 / 29.68 ±0.21 / 29.94 ms │        29.60 / 29.75 ±0.14 / 29.98 ms │     no change │
│ QQuery 54 │        55.79 / 56.08 ±0.22 / 56.31 ms │        55.19 / 55.49 ±0.33 / 56.11 ms │     no change │
│ QQuery 55 │        23.30 / 24.67 ±1.98 / 28.60 ms │        23.58 / 23.82 ±0.16 / 24.00 ms │     no change │
│ QQuery 56 │        39.22 / 41.07 ±2.92 / 46.88 ms │        38.46 / 39.45 ±0.94 / 41.13 ms │     no change │
│ QQuery 57 │     176.37 / 178.39 ±2.08 / 182.24 ms │     175.04 / 176.59 ±1.56 / 179.43 ms │     no change │
│ QQuery 58 │     114.18 / 115.05 ±0.95 / 116.39 ms │     114.45 / 116.45 ±2.90 / 122.12 ms │     no change │
│ QQuery 59 │     117.09 / 118.28 ±0.86 / 119.25 ms │     117.52 / 119.07 ±2.18 / 123.39 ms │     no change │
│ QQuery 60 │        39.25 / 39.96 ±0.61 / 40.75 ms │        39.58 / 40.52 ±0.74 / 41.48 ms │     no change │
│ QQuery 61 │        13.13 / 13.24 ±0.08 / 13.38 ms │        12.79 / 13.00 ±0.26 / 13.50 ms │     no change │
│ QQuery 62 │        46.39 / 46.51 ±0.09 / 46.66 ms │        46.08 / 46.56 ±0.33 / 47.01 ms │     no change │
│ QQuery 63 │        29.73 / 29.95 ±0.14 / 30.10 ms │        29.56 / 30.46 ±0.75 / 31.79 ms │     no change │
│ QQuery 64 │     396.20 / 400.05 ±4.88 / 409.67 ms │    412.30 / 423.42 ±12.43 / 445.92 ms │  1.06x slower │
│ QQuery 65 │     146.12 / 153.63 ±5.07 / 161.82 ms │     142.95 / 150.28 ±4.86 / 157.03 ms │     no change │
│ QQuery 66 │        78.81 / 81.16 ±3.44 / 87.99 ms │        79.50 / 80.22 ±0.73 / 81.14 ms │     no change │
│ QQuery 67 │     253.46 / 258.01 ±3.67 / 262.36 ms │     248.80 / 256.49 ±7.03 / 265.35 ms │     no change │
│ QQuery 68 │        11.79 / 12.07 ±0.24 / 12.50 ms │        10.96 / 11.12 ±0.20 / 11.49 ms │ +1.09x faster │
│ QQuery 69 │        57.37 / 57.69 ±0.25 / 58.05 ms │        57.53 / 57.67 ±0.15 / 57.91 ms │     no change │
│ QQuery 70 │     104.46 / 111.09 ±4.86 / 116.38 ms │     105.68 / 112.74 ±9.71 / 131.11 ms │     no change │
│ QQuery 71 │        35.50 / 36.22 ±0.59 / 37.22 ms │        35.59 / 36.40 ±0.85 / 37.96 ms │     no change │
│ QQuery 72 │ 2103.92 / 2173.10 ±44.51 / 2229.02 ms │ 2126.08 / 2171.68 ±47.34 / 2261.85 ms │     no change │
│ QQuery 73 │          9.14 / 9.61 ±0.40 / 10.20 ms │           8.07 / 8.32 ±0.21 / 8.70 ms │ +1.16x faster │
│ QQuery 74 │     171.04 / 176.37 ±4.31 / 183.83 ms │     173.95 / 177.36 ±6.07 / 189.48 ms │     no change │
│ QQuery 75 │     148.00 / 148.93 ±1.09 / 150.99 ms │     136.06 / 136.98 ±1.28 / 139.52 ms │ +1.09x faster │
│ QQuery 76 │        35.57 / 38.63 ±4.86 / 48.30 ms │        35.74 / 36.22 ±0.34 / 36.75 ms │ +1.07x faster │
│ QQuery 77 │        61.70 / 64.26 ±3.20 / 70.49 ms │        61.29 / 63.93 ±3.34 / 70.51 ms │     no change │
│ QQuery 78 │     187.70 / 193.12 ±4.19 / 200.11 ms │     184.79 / 189.35 ±4.21 / 196.86 ms │     no change │
│ QQuery 79 │        67.32 / 68.07 ±0.46 / 68.59 ms │        65.98 / 66.72 ±0.51 / 67.50 ms │     no change │
│ QQuery 80 │     100.24 / 102.92 ±2.77 / 108.26 ms │        94.00 / 96.45 ±1.36 / 97.66 ms │ +1.07x faster │
│ QQuery 81 │        26.00 / 26.40 ±0.38 / 27.00 ms │        25.90 / 28.20 ±3.72 / 35.59 ms │  1.07x slower │
│ QQuery 82 │        16.42 / 16.96 ±0.52 / 17.84 ms │        16.60 / 17.22 ±0.42 / 17.72 ms │     no change │
│ QQuery 83 │        39.58 / 40.09 ±0.37 / 40.50 ms │        40.09 / 40.43 ±0.21 / 40.69 ms │     no change │
│ QQuery 84 │        34.69 / 34.87 ±0.14 / 35.04 ms │        34.53 / 34.99 ±0.38 / 35.65 ms │     no change │
│ QQuery 85 │     108.14 / 111.95 ±3.64 / 118.61 ms │      99.56 / 101.66 ±2.80 / 107.20 ms │ +1.10x faster │
│ QQuery 86 │        25.22 / 25.59 ±0.21 / 25.84 ms │        25.38 / 25.74 ±0.24 / 26.10 ms │     no change │
│ QQuery 87 │        64.21 / 65.08 ±0.71 / 66.02 ms │        63.66 / 64.73 ±0.76 / 65.76 ms │     no change │
│ QQuery 88 │        62.47 / 64.12 ±1.92 / 67.77 ms │        62.60 / 63.45 ±0.51 / 63.96 ms │     no change │
│ QQuery 89 │        35.94 / 37.89 ±2.54 / 42.87 ms │        36.30 / 36.77 ±0.44 / 37.49 ms │     no change │
│ QQuery 90 │        17.01 / 17.29 ±0.38 / 18.04 ms │        17.18 / 19.23 ±3.67 / 26.58 ms │  1.11x slower │
│ QQuery 91 │        45.28 / 45.56 ±0.35 / 46.21 ms │        46.34 / 46.56 ±0.19 / 46.82 ms │     no change │
│ QQuery 92 │        30.25 / 30.61 ±0.33 / 31.22 ms │        30.41 / 30.71 ±0.25 / 31.14 ms │     no change │
│ QQuery 93 │        50.45 / 52.12 ±1.95 / 55.90 ms │        77.75 / 79.32 ±1.44 / 81.97 ms │  1.52x slower │
│ QQuery 94 │        38.79 / 40.33 ±2.27 / 44.83 ms │        38.47 / 40.29 ±1.79 / 43.45 ms │     no change │
│ QQuery 95 │        84.02 / 86.04 ±1.02 / 86.70 ms │        83.05 / 84.64 ±1.04 / 85.71 ms │     no change │
│ QQuery 96 │        24.10 / 24.34 ±0.28 / 24.86 ms │        24.12 / 24.60 ±0.33 / 24.97 ms │     no change │
│ QQuery 97 │        46.30 / 47.69 ±1.80 / 51.18 ms │        45.99 / 46.47 ±0.48 / 47.13 ms │     no change │
│ QQuery 98 │        43.17 / 43.91 ±0.54 / 44.50 ms │        43.08 / 44.93 ±1.98 / 48.36 ms │     no change │
│ QQuery 99 │        69.17 / 69.65 ±0.50 / 70.60 ms │        70.41 / 70.58 ±0.18 / 70.92 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                             ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 10338.98ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 10338.45ms │
│ Average Time (HEAD)                           │   104.43ms │
│ Average Time (codex_parquet-morsel-splitting) │   104.43ms │
│ Queries Faster                                │         10 │
│ Queries Slower                                │          6 │
│ Queries with No Change                        │         83 │
│ Queries with Failure                          │          0 │
└───────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 2.0 GiB
Avg memory 1.3 GiB
CPU user 232.0s
CPU sys 6.1s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 1.9 GiB
Avg memory 1.3 GiB
CPU user 239.1s
CPU sys 6.6s
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)

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 codex_parquet-morsel-splitting
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃        codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.16 / 3.38 ±4.34 / 12.06 ms │          1.13 / 3.27 ±4.16 / 11.60 ms │     no change │
│ QQuery 1  │        12.34 / 12.96 ±0.36 / 13.42 ms │        12.44 / 12.96 ±0.61 / 14.14 ms │     no change │
│ QQuery 2  │        35.16 / 35.70 ±0.41 / 36.27 ms │        34.55 / 34.81 ±0.36 / 35.49 ms │     no change │
│ QQuery 3  │        30.05 / 30.84 ±0.64 / 31.93 ms │        29.59 / 29.79 ±0.21 / 30.18 ms │     no change │
│ QQuery 4  │     223.44 / 228.87 ±3.29 / 232.66 ms │     220.25 / 224.47 ±3.06 / 229.19 ms │     no change │
│ QQuery 5  │     272.43 / 274.52 ±2.56 / 279.20 ms │     270.84 / 274.80 ±5.13 / 284.73 ms │     no change │
│ QQuery 6  │           1.19 / 1.34 ±0.22 / 1.77 ms │           1.18 / 1.32 ±0.22 / 1.76 ms │     no change │
│ QQuery 7  │        14.07 / 14.20 ±0.09 / 14.32 ms │        13.69 / 13.88 ±0.13 / 14.09 ms │     no change │
│ QQuery 8  │     321.45 / 324.93 ±4.28 / 333.11 ms │    321.71 / 333.83 ±10.40 / 349.36 ms │     no change │
│ QQuery 9  │     463.07 / 466.99 ±3.32 / 471.72 ms │     436.98 / 445.83 ±7.82 / 460.12 ms │     no change │
│ QQuery 10 │        68.15 / 69.27 ±0.80 / 70.37 ms │        67.77 / 68.84 ±1.20 / 71.06 ms │     no change │
│ QQuery 11 │        80.65 / 81.43 ±0.95 / 83.19 ms │        79.65 / 80.63 ±0.56 / 81.39 ms │     no change │
│ QQuery 12 │     263.45 / 268.99 ±5.50 / 278.00 ms │     266.07 / 271.20 ±4.29 / 277.17 ms │     no change │
│ QQuery 13 │     363.73 / 373.90 ±9.28 / 389.30 ms │     368.89 / 377.77 ±6.09 / 387.91 ms │     no change │
│ QQuery 14 │     276.98 / 282.85 ±6.57 / 295.10 ms │     283.20 / 287.22 ±2.92 / 290.75 ms │     no change │
│ QQuery 15 │     268.30 / 277.02 ±6.89 / 284.47 ms │     270.19 / 283.69 ±8.73 / 295.24 ms │     no change │
│ QQuery 16 │     614.48 / 622.60 ±6.08 / 630.77 ms │     617.79 / 628.22 ±6.38 / 636.75 ms │     no change │
│ QQuery 17 │     613.08 / 625.72 ±8.20 / 634.64 ms │     625.09 / 634.63 ±9.08 / 647.07 ms │     no change │
│ QQuery 18 │ 1280.42 / 1297.52 ±17.83 / 1328.41 ms │ 1257.32 / 1305.13 ±26.76 / 1331.41 ms │     no change │
│ QQuery 19 │       27.16 / 34.44 ±13.72 / 61.87 ms │       27.28 / 38.18 ±13.27 / 55.92 ms │  1.11x slower │
│ QQuery 20 │    514.20 / 534.24 ±22.94 / 575.99 ms │    514.95 / 530.27 ±14.81 / 557.95 ms │     no change │
│ QQuery 21 │     511.29 / 515.66 ±4.82 / 522.97 ms │     514.82 / 518.69 ±2.81 / 522.05 ms │     no change │
│ QQuery 22 │     978.58 / 992.33 ±7.66 / 999.65 ms │     953.27 / 958.87 ±4.39 / 964.85 ms │     no change │
│ QQuery 23 │ 3132.05 / 3179.92 ±28.64 / 3214.66 ms │  3067.22 / 3077.64 ±5.40 / 3082.51 ms │     no change │
│ QQuery 24 │        41.00 / 46.91 ±9.89 / 66.60 ms │        40.80 / 44.07 ±5.73 / 55.50 ms │ +1.06x faster │
│ QQuery 25 │     110.23 / 111.48 ±0.84 / 112.77 ms │     110.75 / 114.71 ±3.19 / 118.35 ms │     no change │
│ QQuery 26 │        41.53 / 44.45 ±4.04 / 52.42 ms │        41.24 / 44.45 ±4.21 / 52.55 ms │     no change │
│ QQuery 27 │     665.76 / 677.14 ±7.08 / 684.11 ms │    664.55 / 679.51 ±14.38 / 705.83 ms │     no change │
│ QQuery 28 │ 3028.19 / 3060.92 ±28.53 / 3108.97 ms │ 2865.42 / 2892.38 ±13.50 / 2900.16 ms │ +1.06x faster │
│ QQuery 29 │       39.90 / 51.87 ±16.18 / 81.36 ms │        39.30 / 41.78 ±3.88 / 49.44 ms │ +1.24x faster │
│ QQuery 30 │     299.42 / 308.16 ±6.91 / 320.41 ms │     301.33 / 308.75 ±8.27 / 324.47 ms │     no change │
│ QQuery 31 │     282.10 / 288.35 ±5.37 / 298.20 ms │     284.74 / 290.59 ±6.12 / 301.67 ms │     no change │
│ QQuery 32 │   934.15 / 985.78 ±29.07 / 1015.56 ms │    940.96 / 960.25 ±17.57 / 989.59 ms │     no change │
│ QQuery 33 │ 1460.64 / 1497.21 ±22.77 / 1522.32 ms │ 1462.60 / 1498.73 ±29.74 / 1540.54 ms │     no change │
│ QQuery 34 │ 1479.69 / 1506.53 ±19.57 / 1531.10 ms │ 1477.57 / 1513.85 ±27.73 / 1546.64 ms │     no change │
│ QQuery 35 │    282.48 / 293.70 ±17.80 / 329.18 ms │    281.73 / 332.25 ±54.49 / 430.89 ms │  1.13x slower │
│ QQuery 36 │        64.57 / 68.92 ±5.36 / 79.41 ms │        66.57 / 71.82 ±5.06 / 80.90 ms │     no change │
│ QQuery 37 │        34.92 / 36.49 ±1.67 / 38.85 ms │        36.15 / 40.48 ±5.08 / 49.48 ms │  1.11x slower │
│ QQuery 38 │        39.92 / 43.69 ±3.67 / 49.90 ms │        40.91 / 43.36 ±2.67 / 48.12 ms │     no change │
│ QQuery 39 │     139.16 / 148.91 ±8.81 / 162.66 ms │    131.29 / 154.11 ±11.43 / 160.88 ms │     no change │
│ QQuery 40 │        13.84 / 14.85 ±1.51 / 17.83 ms │        13.87 / 15.07 ±1.38 / 17.64 ms │     no change │
│ QQuery 41 │        13.47 / 14.35 ±1.51 / 17.36 ms │        13.74 / 13.90 ±0.14 / 14.10 ms │     no change │
│ QQuery 42 │        12.79 / 13.65 ±1.13 / 15.85 ms │        13.02 / 13.17 ±0.13 / 13.39 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                             ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 19762.97ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 19509.19ms │
│ Average Time (HEAD)                           │   459.60ms │
│ Average Time (codex_parquet-morsel-splitting) │   453.70ms │
│ Queries Faster                                │          3 │
│ Queries Slower                                │          3 │
│ Queries with No Change                        │         37 │
│ Queries with Failure                          │          0 │
└───────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 11.1 GiB
Avg memory 4.5 GiB
CPU user 1012.4s
CPU sys 70.6s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 100.0s
Peak memory 10.9 GiB
Avg memory 4.3 GiB
CPU user 1013.2s
CPU sys 72.6s
Peak spill 0 B

File an issue against this benchmark runner

A stream that found the shared work queue empty terminated immediately,
so by the time a donor finished planning a large file and donated its
surplus morsels, no sibling was left to steal them (and the split hint
often reported a single active worker, suppressing the split entirely).

Popping a file now registers a FileLease on the shared work source;
while any lease is alive, an idle sibling yields Poll::Pending and
re-polls instead of finishing, then steals donated morsels once the
donor publishes them. The lease is released when the donation window
closes (the first morsel starts streaming) or the file is abandoned.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@Dandandan

Copy link
Copy Markdown
Contributor Author

run benchmarks

@adriangbot

Copy link
Copy Markdown

🤖 Benchmark running (GKE) | trigger
Instance: c4a-highmem-16 (12 vCPU / 65 GiB) | Linux bench-c4693734572-561-ll495 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (7bb5a20) to 883c38e (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-c4693734572-562-9bdkm 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (7bb5a20) to 883c38e (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-c4693734572-563-zlgc9 6.12.68+ #1 SMP Sat May 2 07:49:07 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 codex/parquet-morsel-splitting (7bb5a20) to 883c38e (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 codex_parquet-morsel-splitting
--------------------
Benchmark tpch_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                           HEAD ┃ codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │ 38.76 / 39.31 ±1.03 / 41.37 ms │ 38.28 / 39.05 ±1.07 / 41.14 ms │     no change │
│ QQuery 2  │ 18.79 / 18.92 ±0.11 / 19.09 ms │ 18.94 / 19.07 ±0.11 / 19.20 ms │     no change │
│ QQuery 3  │ 32.35 / 34.54 ±1.56 / 37.22 ms │ 30.99 / 31.47 ±0.34 / 32.01 ms │ +1.10x faster │
│ QQuery 4  │ 17.29 / 18.07 ±1.28 / 20.63 ms │ 17.79 / 18.28 ±0.70 / 19.64 ms │     no change │
│ QQuery 5  │ 38.54 / 40.69 ±1.39 / 42.69 ms │ 41.07 / 43.77 ±4.36 / 52.48 ms │  1.08x slower │
│ QQuery 6  │ 16.24 / 16.44 ±0.13 / 16.58 ms │ 16.41 / 17.12 ±0.95 / 19.00 ms │     no change │
│ QQuery 7  │ 44.46 / 45.54 ±0.87 / 46.85 ms │ 44.96 / 46.64 ±1.64 / 49.46 ms │     no change │
│ QQuery 8  │ 43.12 / 43.63 ±0.47 / 44.50 ms │ 43.40 / 45.10 ±2.05 / 48.84 ms │     no change │
│ QQuery 9  │ 49.80 / 50.74 ±0.78 / 51.99 ms │ 52.44 / 53.22 ±0.70 / 54.53 ms │     no change │
│ QQuery 10 │ 42.74 / 42.89 ±0.12 / 43.10 ms │ 42.81 / 43.42 ±0.70 / 44.80 ms │     no change │
│ QQuery 11 │ 13.56 / 13.71 ±0.11 / 13.85 ms │ 13.68 / 13.83 ±0.16 / 14.12 ms │     no change │
│ QQuery 12 │ 23.88 / 24.31 ±0.40 / 24.90 ms │ 24.33 / 25.02 ±0.43 / 25.62 ms │     no change │
│ QQuery 13 │ 31.91 / 33.45 ±1.28 / 34.83 ms │ 32.78 / 33.71 ±0.61 / 34.47 ms │     no change │
│ QQuery 14 │ 24.06 / 24.26 ±0.10 / 24.35 ms │ 23.96 / 24.15 ±0.14 / 24.31 ms │     no change │
│ QQuery 15 │ 31.84 / 32.73 ±0.62 / 33.48 ms │ 31.78 / 32.01 ±0.20 / 32.38 ms │     no change │
│ QQuery 16 │ 14.19 / 14.35 ±0.17 / 14.65 ms │ 14.28 / 14.71 ±0.24 / 14.94 ms │     no change │
│ QQuery 17 │ 76.81 / 78.09 ±1.27 / 80.03 ms │ 76.27 / 76.99 ±0.54 / 77.67 ms │     no change │
│ QQuery 18 │ 60.34 / 61.74 ±1.62 / 64.31 ms │ 67.28 / 69.80 ±1.53 / 71.85 ms │  1.13x slower │
│ QQuery 19 │ 33.29 / 34.05 ±1.16 / 36.35 ms │ 33.45 / 34.18 ±0.49 / 34.97 ms │     no change │
│ QQuery 20 │ 32.49 / 32.64 ±0.12 / 32.84 ms │ 32.50 / 33.19 ±0.63 / 34.32 ms │     no change │
│ QQuery 21 │ 56.02 / 57.74 ±1.19 / 59.13 ms │ 62.41 / 63.81 ±1.39 / 66.34 ms │  1.11x slower │
│ QQuery 22 │ 14.31 / 14.55 ±0.17 / 14.80 ms │ 14.42 / 14.55 ±0.10 / 14.68 ms │     no change │
└───────────┴────────────────────────────────┴────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓
┃ Benchmark Summary                             ┃          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 772.40ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 793.08ms │
│ Average Time (HEAD)                           │  35.11ms │
│ Average Time (codex_parquet-morsel-splitting) │  36.05ms │
│ Queries Faster                                │        1 │
│ Queries Slower                                │        3 │
│ Queries with No Change                        │       18 │
│ Queries with Failure                          │        0 │
└───────────────────────────────────────────────┴──────────┘

Resource Usage

tpch — base (merge-base)

Metric Value
Wall time 5.0s
Peak memory 1.1 GiB
Avg memory 507.2 MiB
CPU user 23.0s
CPU sys 1.7s
Peak spill 0 B

tpch — branch

Metric Value
Wall time 5.0s
Peak memory 1.2 GiB
Avg memory 515.6 MiB
CPU user 24.5s
CPU sys 1.9s
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 codex_parquet-morsel-splitting
--------------------
Benchmark tpcds_sf1.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃        codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 1  │           5.82 / 6.36 ±0.84 / 8.03 ms │           5.80 / 6.36 ±0.90 / 8.15 ms │     no change │
│ QQuery 2  │        80.35 / 80.53 ±0.21 / 80.95 ms │        81.44 / 81.76 ±0.20 / 82.05 ms │     no change │
│ QQuery 3  │        28.78 / 28.93 ±0.12 / 29.08 ms │        28.56 / 28.80 ±0.18 / 29.06 ms │     no change │
│ QQuery 4  │     494.33 / 497.08 ±2.86 / 502.02 ms │     503.93 / 508.30 ±3.98 / 514.50 ms │     no change │
│ QQuery 5  │        51.33 / 51.65 ±0.32 / 52.24 ms │        51.46 / 52.14 ±0.44 / 52.75 ms │     no change │
│ QQuery 6  │        36.73 / 36.89 ±0.12 / 37.07 ms │        36.23 / 36.60 ±0.29 / 37.12 ms │     no change │
│ QQuery 7  │        95.40 / 95.72 ±0.43 / 96.57 ms │        95.04 / 96.71 ±1.95 / 99.51 ms │     no change │
│ QQuery 8  │        36.82 / 37.14 ±0.19 / 37.37 ms │        37.25 / 37.62 ±0.29 / 37.98 ms │     no change │
│ QQuery 9  │        51.37 / 54.06 ±1.57 / 56.18 ms │        52.26 / 54.41 ±1.68 / 56.04 ms │     no change │
│ QQuery 10 │        62.75 / 63.16 ±0.44 / 64.01 ms │        63.32 / 63.57 ±0.17 / 63.78 ms │     no change │
│ QQuery 11 │     299.44 / 308.33 ±6.62 / 317.71 ms │     312.09 / 319.13 ±4.99 / 327.71 ms │     no change │
│ QQuery 12 │        28.31 / 28.92 ±0.41 / 29.55 ms │        29.00 / 29.24 ±0.17 / 29.48 ms │     no change │
│ QQuery 13 │     118.87 / 120.25 ±1.17 / 122.43 ms │     115.86 / 117.26 ±1.08 / 118.77 ms │     no change │
│ QQuery 14 │     410.39 / 413.37 ±2.06 / 415.70 ms │     412.51 / 421.91 ±8.14 / 434.29 ms │     no change │
│ QQuery 15 │        58.72 / 58.99 ±0.23 / 59.27 ms │        59.19 / 60.27 ±0.56 / 60.76 ms │     no change │
│ QQuery 16 │           6.57 / 6.74 ±0.16 / 7.02 ms │           6.80 / 7.00 ±0.26 / 7.50 ms │     no change │
│ QQuery 17 │        80.43 / 81.12 ±0.53 / 82.07 ms │        80.75 / 81.81 ±1.14 / 83.99 ms │     no change │
│ QQuery 18 │     124.33 / 125.60 ±1.37 / 128.14 ms │     122.18 / 125.88 ±4.33 / 134.06 ms │     no change │
│ QQuery 19 │        41.40 / 41.91 ±0.46 / 42.74 ms │        42.02 / 42.42 ±0.35 / 42.88 ms │     no change │
│ QQuery 20 │        34.68 / 34.98 ±0.25 / 35.29 ms │        35.78 / 36.35 ±0.56 / 37.38 ms │     no change │
│ QQuery 21 │        16.79 / 17.27 ±0.41 / 17.90 ms │        17.39 / 18.41 ±1.67 / 21.75 ms │  1.07x slower │
│ QQuery 22 │        61.92 / 63.61 ±1.59 / 66.56 ms │        62.97 / 64.14 ±1.06 / 65.65 ms │     no change │
│ QQuery 23 │     317.99 / 319.49 ±1.86 / 323.11 ms │     323.43 / 326.69 ±2.25 / 329.39 ms │     no change │
│ QQuery 24 │     223.11 / 226.98 ±3.48 / 233.48 ms │     210.12 / 214.68 ±6.05 / 226.56 ms │ +1.06x faster │
│ QQuery 25 │     109.96 / 113.26 ±3.96 / 120.92 ms │     110.03 / 110.91 ±0.73 / 111.94 ms │     no change │
│ QQuery 26 │        57.70 / 58.32 ±0.54 / 59.06 ms │        57.75 / 59.93 ±3.07 / 66.02 ms │     no change │
│ QQuery 27 │           6.67 / 6.81 ±0.18 / 7.16 ms │           6.68 / 6.86 ±0.14 / 7.10 ms │     no change │
│ QQuery 28 │        57.62 / 60.37 ±1.37 / 61.09 ms │        61.30 / 62.79 ±2.20 / 67.16 ms │     no change │
│ QQuery 29 │        97.82 / 98.35 ±0.38 / 98.77 ms │        98.12 / 98.40 ±0.27 / 98.76 ms │     no change │
│ QQuery 30 │        32.53 / 32.80 ±0.29 / 33.21 ms │        32.77 / 33.91 ±1.74 / 37.37 ms │     no change │
│ QQuery 31 │     110.85 / 113.78 ±2.94 / 119.43 ms │     111.64 / 113.96 ±2.25 / 118.26 ms │     no change │
│ QQuery 32 │        20.00 / 20.22 ±0.19 / 20.54 ms │        20.63 / 20.81 ±0.12 / 21.00 ms │     no change │
│ QQuery 33 │        38.36 / 38.50 ±0.13 / 38.71 ms │        38.36 / 38.79 ±0.53 / 39.74 ms │     no change │
│ QQuery 34 │           9.36 / 9.53 ±0.23 / 9.98 ms │           8.27 / 8.39 ±0.11 / 8.59 ms │ +1.14x faster │
│ QQuery 35 │        71.32 / 73.15 ±2.09 / 76.89 ms │        72.99 / 74.60 ±2.36 / 79.29 ms │     no change │
│ QQuery 36 │           5.87 / 6.01 ±0.16 / 6.32 ms │           5.85 / 6.02 ±0.16 / 6.32 ms │     no change │
│ QQuery 37 │           6.80 / 6.96 ±0.15 / 7.19 ms │           6.88 / 6.94 ±0.07 / 7.05 ms │     no change │
│ QQuery 38 │        63.33 / 63.68 ±0.39 / 64.30 ms │        63.38 / 63.80 ±0.48 / 64.72 ms │     no change │
│ QQuery 39 │     451.19 / 456.44 ±4.07 / 461.93 ms │     461.20 / 470.42 ±5.32 / 476.17 ms │     no change │
│ QQuery 40 │        23.20 / 23.92 ±0.98 / 25.87 ms │        21.14 / 22.99 ±2.97 / 28.92 ms │     no change │
│ QQuery 41 │        11.11 / 11.31 ±0.24 / 11.74 ms │        11.29 / 11.52 ±0.23 / 11.95 ms │     no change │
│ QQuery 42 │        23.50 / 24.21 ±0.75 / 25.61 ms │        24.32 / 24.74 ±0.66 / 26.05 ms │     no change │
│ QQuery 43 │           4.74 / 4.85 ±0.14 / 5.13 ms │           4.80 / 4.95 ±0.19 / 5.32 ms │     no change │
│ QQuery 44 │           8.84 / 8.91 ±0.09 / 9.10 ms │           9.02 / 9.15 ±0.12 / 9.36 ms │     no change │
│ QQuery 45 │        38.21 / 38.64 ±0.43 / 39.37 ms │        40.00 / 40.56 ±0.34 / 40.95 ms │     no change │
│ QQuery 46 │        11.74 / 12.00 ±0.29 / 12.45 ms │        10.74 / 10.92 ±0.19 / 11.19 ms │ +1.10x faster │
│ QQuery 47 │     228.39 / 230.93 ±2.34 / 234.72 ms │     242.47 / 246.10 ±3.52 / 251.50 ms │  1.07x slower │
│ QQuery 48 │        96.28 / 96.93 ±0.41 / 97.53 ms │       93.97 / 95.68 ±2.28 / 100.07 ms │     no change │
│ QQuery 49 │        76.96 / 79.62 ±4.45 / 88.49 ms │        77.63 / 78.53 ±0.71 / 79.48 ms │     no change │
│ QQuery 50 │        59.15 / 59.47 ±0.25 / 59.78 ms │        59.76 / 60.49 ±0.62 / 61.63 ms │     no change │
│ QQuery 51 │       92.70 / 95.99 ±3.02 / 101.68 ms │       91.80 / 95.99 ±5.11 / 106.00 ms │     no change │
│ QQuery 52 │        24.08 / 24.29 ±0.33 / 24.95 ms │        24.16 / 24.55 ±0.27 / 24.99 ms │     no change │
│ QQuery 53 │        29.44 / 29.67 ±0.17 / 29.93 ms │        29.66 / 29.93 ±0.24 / 30.28 ms │     no change │
│ QQuery 54 │        55.19 / 55.71 ±0.41 / 56.42 ms │        55.77 / 56.32 ±0.35 / 56.85 ms │     no change │
│ QQuery 55 │        23.56 / 23.85 ±0.25 / 24.21 ms │        23.87 / 24.46 ±0.52 / 25.25 ms │     no change │
│ QQuery 56 │        38.57 / 41.07 ±3.10 / 46.47 ms │        39.01 / 39.72 ±0.79 / 41.20 ms │     no change │
│ QQuery 57 │     175.98 / 177.01 ±0.53 / 177.46 ms │     177.14 / 179.96 ±3.14 / 186.04 ms │     no change │
│ QQuery 58 │     114.45 / 116.70 ±2.60 / 121.74 ms │     114.24 / 115.57 ±0.85 / 116.89 ms │     no change │
│ QQuery 59 │     116.45 / 119.45 ±2.48 / 122.66 ms │     118.69 / 119.67 ±0.94 / 121.28 ms │     no change │
│ QQuery 60 │        39.23 / 39.97 ±0.69 / 41.17 ms │        39.93 / 40.93 ±0.90 / 42.31 ms │     no change │
│ QQuery 61 │        12.96 / 13.09 ±0.15 / 13.38 ms │        13.33 / 13.49 ±0.14 / 13.74 ms │     no change │
│ QQuery 62 │        45.81 / 46.44 ±0.34 / 46.77 ms │        46.92 / 47.16 ±0.21 / 47.54 ms │     no change │
│ QQuery 63 │        29.46 / 29.63 ±0.11 / 29.79 ms │        30.41 / 31.38 ±1.50 / 34.35 ms │  1.06x slower │
│ QQuery 64 │    398.33 / 408.05 ±11.89 / 426.68 ms │     416.08 / 425.73 ±4.96 / 430.15 ms │     no change │
│ QQuery 65 │     144.63 / 152.16 ±5.68 / 160.40 ms │     149.10 / 151.92 ±2.62 / 156.84 ms │     no change │
│ QQuery 66 │        79.66 / 80.60 ±0.53 / 81.31 ms │        80.54 / 83.17 ±3.17 / 89.41 ms │     no change │
│ QQuery 67 │     256.43 / 263.89 ±7.32 / 277.52 ms │     255.72 / 259.83 ±4.09 / 264.98 ms │     no change │
│ QQuery 68 │        11.84 / 12.05 ±0.13 / 12.21 ms │        10.90 / 11.17 ±0.23 / 11.46 ms │ +1.08x faster │
│ QQuery 69 │        57.34 / 58.01 ±0.41 / 58.56 ms │        57.43 / 58.10 ±0.35 / 58.41 ms │     no change │
│ QQuery 70 │     104.78 / 109.23 ±6.85 / 122.88 ms │    106.23 / 113.95 ±12.26 / 138.32 ms │     no change │
│ QQuery 71 │        35.51 / 36.09 ±0.39 / 36.69 ms │        35.58 / 36.01 ±0.26 / 36.35 ms │     no change │
│ QQuery 72 │ 2146.08 / 2212.09 ±61.40 / 2291.05 ms │ 2162.03 / 2216.29 ±47.79 / 2293.42 ms │     no change │
│ QQuery 73 │           9.48 / 9.66 ±0.11 / 9.78 ms │           8.19 / 8.35 ±0.16 / 8.64 ms │ +1.16x faster │
│ QQuery 74 │     179.57 / 181.79 ±1.26 / 183.46 ms │     179.09 / 183.97 ±7.89 / 199.65 ms │     no change │
│ QQuery 75 │     147.95 / 149.12 ±0.76 / 150.08 ms │     136.73 / 138.59 ±1.83 / 141.98 ms │ +1.08x faster │
│ QQuery 76 │        35.56 / 37.54 ±3.29 / 44.09 ms │        35.42 / 35.94 ±0.35 / 36.38 ms │     no change │
│ QQuery 77 │        61.69 / 62.53 ±0.54 / 63.14 ms │        61.00 / 61.22 ±0.18 / 61.49 ms │     no change │
│ QQuery 78 │     186.00 / 190.60 ±4.65 / 199.27 ms │     184.02 / 188.82 ±6.05 / 200.73 ms │     no change │
│ QQuery 79 │        67.16 / 67.43 ±0.28 / 67.93 ms │        66.90 / 69.93 ±3.29 / 76.03 ms │     no change │
│ QQuery 80 │     101.20 / 102.98 ±1.14 / 104.41 ms │        94.08 / 95.04 ±1.22 / 97.44 ms │ +1.08x faster │
│ QQuery 81 │        26.25 / 26.62 ±0.32 / 27.10 ms │        26.03 / 26.32 ±0.17 / 26.56 ms │     no change │
│ QQuery 82 │        17.00 / 17.42 ±0.64 / 18.69 ms │        16.74 / 17.33 ±1.12 / 19.56 ms │     no change │
│ QQuery 83 │        40.50 / 41.05 ±0.52 / 41.69 ms │        39.79 / 43.59 ±5.96 / 55.44 ms │  1.06x slower │
│ QQuery 84 │        34.99 / 35.22 ±0.22 / 35.62 ms │        35.25 / 35.69 ±0.41 / 36.37 ms │     no change │
│ QQuery 85 │     109.08 / 113.01 ±3.68 / 118.81 ms │     100.18 / 101.48 ±1.04 / 103.30 ms │ +1.11x faster │
│ QQuery 86 │        25.18 / 25.49 ±0.21 / 25.78 ms │        25.28 / 25.63 ±0.21 / 25.85 ms │     no change │
│ QQuery 87 │        65.39 / 66.32 ±0.63 / 67.09 ms │        65.07 / 67.29 ±2.34 / 71.21 ms │     no change │
│ QQuery 88 │        63.41 / 63.93 ±0.37 / 64.42 ms │        62.88 / 63.90 ±0.58 / 64.63 ms │     no change │
│ QQuery 89 │        36.19 / 36.93 ±0.51 / 37.61 ms │        36.53 / 36.99 ±0.40 / 37.65 ms │     no change │
│ QQuery 90 │        17.13 / 17.46 ±0.21 / 17.69 ms │        17.57 / 17.67 ±0.09 / 17.80 ms │     no change │
│ QQuery 91 │        45.19 / 45.48 ±0.15 / 45.61 ms │        45.25 / 47.28 ±3.58 / 54.43 ms │     no change │
│ QQuery 92 │        30.10 / 30.37 ±0.25 / 30.85 ms │        30.68 / 31.12 ±0.27 / 31.52 ms │     no change │
│ QQuery 93 │        51.15 / 53.08 ±3.03 / 59.12 ms │        32.44 / 33.17 ±0.57 / 33.75 ms │ +1.60x faster │
│ QQuery 94 │        39.48 / 40.14 ±0.63 / 41.26 ms │        39.59 / 39.97 ±0.27 / 40.34 ms │     no change │
│ QQuery 95 │        84.65 / 85.56 ±0.76 / 86.86 ms │        80.18 / 86.76 ±6.47 / 98.25 ms │     no change │
│ QQuery 96 │        24.86 / 25.05 ±0.20 / 25.43 ms │        24.53 / 24.74 ±0.16 / 24.89 ms │     no change │
│ QQuery 97 │        46.31 / 48.78 ±4.06 / 56.88 ms │        46.89 / 47.21 ±0.25 / 47.53 ms │     no change │
│ QQuery 98 │        42.77 / 43.72 ±0.66 / 44.76 ms │        42.99 / 43.87 ±1.08 / 45.50 ms │     no change │
│ QQuery 99 │        69.70 / 70.01 ±0.25 / 70.26 ms │        70.60 / 70.86 ±0.16 / 71.11 ms │     no change │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                             ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 10380.42ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 10431.67ms │
│ Average Time (HEAD)                           │   104.85ms │
│ Average Time (codex_parquet-morsel-splitting) │   105.37ms │
│ Queries Faster                                │          9 │
│ Queries Slower                                │          4 │
│ Queries with No Change                        │         86 │
│ Queries with Failure                          │          0 │
└───────────────────────────────────────────────┴────────────┘

Resource Usage

tpcds — base (merge-base)

Metric Value
Wall time 55.0s
Peak memory 1.7 GiB
Avg memory 1.2 GiB
CPU user 241.8s
CPU sys 5.9s
Peak spill 0 B

tpcds — branch

Metric Value
Wall time 55.0s
Peak memory 1.7 GiB
Avg memory 1.2 GiB
CPU user 248.8s
CPU sys 6.1s
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 codex_parquet-morsel-splitting
--------------------
Benchmark clickbench_partitioned.json
--------------------
┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Query     ┃                                  HEAD ┃        codex_parquet-morsel-splitting ┃        Change ┃
┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ QQuery 0  │          1.17 / 3.26 ±4.14 / 11.54 ms │          1.20 / 3.31 ±4.17 / 11.65 ms │     no change │
│ QQuery 1  │        12.28 / 12.55 ±0.16 / 12.72 ms │        12.47 / 12.88 ±0.22 / 13.11 ms │     no change │
│ QQuery 2  │        35.42 / 35.59 ±0.16 / 35.83 ms │        35.04 / 35.61 ±0.59 / 36.66 ms │     no change │
│ QQuery 3  │        30.07 / 30.56 ±0.50 / 31.52 ms │        29.93 / 30.19 ±0.26 / 30.55 ms │     no change │
│ QQuery 4  │     223.66 / 229.11 ±4.88 / 237.84 ms │     218.20 / 224.44 ±5.07 / 231.83 ms │     no change │
│ QQuery 5  │     269.45 / 272.31 ±2.38 / 275.47 ms │     270.53 / 272.92 ±2.18 / 275.66 ms │     no change │
│ QQuery 6  │           1.18 / 1.33 ±0.21 / 1.74 ms │           1.20 / 1.35 ±0.23 / 1.79 ms │     no change │
│ QQuery 7  │        13.57 / 13.71 ±0.12 / 13.91 ms │        13.31 / 13.56 ±0.13 / 13.69 ms │     no change │
│ QQuery 8  │     318.19 / 323.89 ±4.94 / 332.15 ms │     320.29 / 332.59 ±8.94 / 345.22 ms │     no change │
│ QQuery 9  │     454.08 / 463.32 ±7.59 / 473.78 ms │     439.22 / 447.14 ±6.37 / 454.74 ms │     no change │
│ QQuery 10 │        67.16 / 68.72 ±1.37 / 71.16 ms │        67.00 / 68.16 ±0.89 / 69.34 ms │     no change │
│ QQuery 11 │        80.32 / 82.52 ±2.32 / 86.77 ms │        78.08 / 84.55 ±7.09 / 95.96 ms │     no change │
│ QQuery 12 │     269.52 / 274.60 ±4.73 / 282.77 ms │     259.73 / 268.41 ±6.39 / 276.20 ms │     no change │
│ QQuery 13 │     365.47 / 376.66 ±7.91 / 389.90 ms │    362.34 / 378.79 ±17.92 / 411.85 ms │     no change │
│ QQuery 14 │    283.76 / 307.37 ±18.97 / 332.14 ms │     277.53 / 284.86 ±7.59 / 299.29 ms │ +1.08x faster │
│ QQuery 15 │    268.72 / 280.44 ±15.36 / 310.47 ms │     261.73 / 274.40 ±7.84 / 285.34 ms │     no change │
│ QQuery 16 │    606.64 / 622.19 ±10.57 / 635.23 ms │     616.14 / 621.89 ±4.61 / 627.91 ms │     no change │
│ QQuery 17 │     619.49 / 625.25 ±3.57 / 628.95 ms │     612.96 / 624.60 ±7.58 / 635.34 ms │     no change │
│ QQuery 18 │  1270.21 / 1279.61 ±9.02 / 1294.42 ms │ 1258.71 / 1275.44 ±14.45 / 1300.93 ms │     no change │
│ QQuery 19 │        26.72 / 26.96 ±0.15 / 27.13 ms │        26.91 / 27.43 ±0.45 / 28.27 ms │     no change │
│ QQuery 20 │    510.24 / 531.00 ±17.42 / 553.51 ms │     517.49 / 526.24 ±5.77 / 533.23 ms │     no change │
│ QQuery 21 │     511.54 / 517.89 ±4.38 / 525.05 ms │     502.32 / 518.24 ±9.14 / 528.80 ms │     no change │
│ QQuery 22 │     975.36 / 983.86 ±5.69 / 993.01 ms │     944.68 / 951.65 ±4.07 / 956.75 ms │     no change │
│ QQuery 23 │ 3167.27 / 3202.15 ±25.37 / 3241.13 ms │ 3036.56 / 3059.29 ±15.38 / 3077.80 ms │     no change │
│ QQuery 24 │        40.76 / 41.10 ±0.20 / 41.32 ms │       40.68 / 49.69 ±15.75 / 81.14 ms │  1.21x slower │
│ QQuery 25 │     109.97 / 111.15 ±0.87 / 112.69 ms │     109.69 / 111.77 ±2.24 / 116.02 ms │     no change │
│ QQuery 26 │        41.14 / 43.55 ±3.29 / 50.06 ms │        41.23 / 43.74 ±3.18 / 49.93 ms │     no change │
│ QQuery 27 │     663.05 / 668.16 ±4.73 / 674.19 ms │    646.30 / 665.35 ±12.34 / 684.15 ms │     no change │
│ QQuery 28 │ 3039.68 / 3068.41 ±20.52 / 3092.98 ms │ 2882.54 / 2890.41 ±12.08 / 2914.48 ms │ +1.06x faster │
│ QQuery 29 │       39.89 / 51.49 ±20.24 / 91.79 ms │       39.45 / 45.28 ±11.04 / 67.35 ms │ +1.14x faster │
│ QQuery 30 │     297.81 / 304.70 ±7.52 / 317.42 ms │     292.94 / 302.96 ±5.68 / 308.70 ms │     no change │
│ QQuery 31 │     284.87 / 290.29 ±4.24 / 297.70 ms │    278.58 / 293.95 ±12.31 / 311.58 ms │     no change │
│ QQuery 32 │    926.51 / 948.51 ±17.98 / 968.57 ms │  959.15 / 1015.88 ±36.18 / 1070.87 ms │  1.07x slower │
│ QQuery 33 │ 1437.80 / 1472.97 ±26.04 / 1515.22 ms │ 1459.80 / 1502.54 ±32.29 / 1537.55 ms │     no change │
│ QQuery 34 │ 1471.06 / 1497.78 ±39.33 / 1574.54 ms │ 1468.86 / 1531.27 ±45.98 / 1594.73 ms │     no change │
│ QQuery 35 │    284.71 / 324.79 ±65.48 / 454.05 ms │     273.05 / 283.15 ±7.82 / 296.73 ms │ +1.15x faster │
│ QQuery 36 │        65.06 / 71.66 ±4.95 / 77.02 ms │        66.38 / 71.53 ±5.71 / 81.62 ms │     no change │
│ QQuery 37 │        35.21 / 37.45 ±2.40 / 41.92 ms │        35.76 / 38.74 ±4.47 / 47.63 ms │     no change │
│ QQuery 38 │        42.02 / 47.67 ±6.71 / 60.47 ms │        43.06 / 46.32 ±3.64 / 53.19 ms │     no change │
│ QQuery 39 │     147.43 / 155.46 ±7.56 / 168.09 ms │     146.30 / 157.67 ±6.92 / 164.61 ms │     no change │
│ QQuery 40 │        13.48 / 14.20 ±0.72 / 15.33 ms │        14.02 / 19.36 ±9.97 / 39.29 ms │  1.36x slower │
│ QQuery 41 │        13.56 / 16.78 ±3.40 / 21.44 ms │        13.41 / 16.94 ±6.53 / 29.99 ms │     no change │
│ QQuery 42 │        12.98 / 14.85 ±3.21 / 21.25 ms │        12.68 / 13.34 ±0.66 / 14.56 ms │ +1.11x faster │
└───────────┴───────────────────────────────────────┴───────────────────────────────────────┴───────────────┘
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ Benchmark Summary                             ┃            ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━┩
│ Total Time (HEAD)                             │ 19745.83ms │
│ Total Time (codex_parquet-morsel-splitting)   │ 19437.81ms │
│ Average Time (HEAD)                           │   459.21ms │
│ Average Time (codex_parquet-morsel-splitting) │   452.04ms │
│ Queries Faster                                │          5 │
│ Queries Slower                                │          3 │
│ Queries with No Change                        │         35 │
│ Queries with Failure                          │          0 │
└───────────────────────────────────────────────┴────────────┘

Resource Usage

clickbench_partitioned — base (merge-base)

Metric Value
Wall time 100.0s
Peak memory 11.4 GiB
Avg memory 4.2 GiB
CPU user 1009.6s
CPU sys 70.2s
Peak spill 0 B

clickbench_partitioned — branch

Metric Value
Wall time 100.0s
Peak memory 11.3 GiB
Avg memory 4.8 GiB
CPU user 1007.8s
CPU sys 73.3s
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

auto detected api change Auto detected API change common Related to common crate datasource Changes to the datasource crate documentation Improvements or additions to documentation proto Related to proto crate sqllogictest SQL Logic Tests (.slt)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants