You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@datafusion.apache.org by "friendlymatthew (via GitHub)" <gi...@apache.org> on 2026/03/13 15:27:44 UTC
[PR] Friendlymatthew/update projections for structs [datafusion]
friendlymatthew opened a new pull request, #20925:
URL: https://github.com/apache/datafusion/pull/20925
- Added on from https://github.com/apache/datafusion/pull/20913
Please review from the third commit
## Rationale for this change
This PR reuses the `ParquetReadPlan` (introduced for the row filter pushdown) to also resolve projection expressions to parquet leaf column indices
Previously, projecting a single field from a struct with many children would read all leaves of that struct. This aligns the projection path with the row filter path, which already had leaf-level struct pruning
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "friendlymatthew (via GitHub)" <gi...@apache.org>.
friendlymatthew commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143001255
run benchmark parquet_struct_projection
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143426543
Nice! these numbers look really good.
```
wide_struct/select_small_field 1.00 419.3±1.93µs ? ?/sec 19.48 8.2±0.12ms ? ?/sec
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4137629738
run benchmark parquet_struct_projection
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4137698911
run benchmark parquet_struct_projection
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143024473
🤖 Criterion benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4142992979)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4142992979-580-5pqnf 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
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 friendlymatthew/update-projections-for-structs (4e3eca988796d4747570111a4365b187c1ad406b) to 37cd3de (merge-base) [diff](https://github.com/apache/datafusion/compare/37cd3de82fcfa7619b04cfb9f19607ff55d44bc4..4e3eca988796d4747570111a4365b187c1ad406b)
BENCH_NAME=parquet_struct_projection
BENCH_COMMAND=cargo bench --features=parquet --bench parquet_struct_projection
BENCH_FILTER=
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4137665444
🤖 Criterion benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4137629738)
`Linux bench-c4137629738-566-5sqv4 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
Comparing friendlymatthew/update-projections-for-structs (63724faa17280cfe4ab2ccf8220406d8c4084cd2) to 1416ed4 (merge-base) [diff](https://github.com/apache/datafusion/compare/1416ed4d5007180136dae0aaeb921f0681ec001e..63724faa17280cfe4ab2ccf8220406d8c4084cd2)
BENCH_NAME=parquet_struct_projection
BENCH_COMMAND=cargo bench --features=parquet --bench parquet_struct_projection
BENCH_FILTER=
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143135945
🤖 Criterion benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4142992979)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
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>
<details><summary>Details</summary>
<p>
```
group friendlymatthew_update-projections-for-structs main
----- ---------------------------------------------- ----
narrow_struct/select_all 1.00 571.9±1.44µs ? ?/sec 1.01 576.7±17.32µs ? ?/sec
narrow_struct/select_id_and_small_field 1.00 444.5±5.65µs ? ?/sec 1.54 684.4±7.11µs ? ?/sec
narrow_struct/select_large_field 1.00 694.7±2.83µs ? ?/sec 1.00 692.6±18.49µs ? ?/sec
narrow_struct/select_small_field 1.00 422.2±3.92µs ? ?/sec 1.55 654.4±23.99µs ? ?/sec
narrow_struct/select_struct 1.00 528.1±1.73µs ? ?/sec 1.09 573.2±2.98µs ? ?/sec
narrow_struct/sum_small_field 1.00 633.5±2.89µs ? ?/sec 1.34 850.5±1.97µs ? ?/sec
nested_struct/select_all 1.00 4.1±0.05ms ? ?/sec 1.19 4.8±0.07ms ? ?/sec
nested_struct/select_extra_string 1.00 685.0±5.25µs ? ?/sec 4.89 3.4±0.06ms ? ?/sec
nested_struct/select_inner_small_field 1.00 449.7±1.02µs ? ?/sec 2.00 899.0±5.72µs ? ?/sec
nested_struct/select_inner_struct 1.00 3.0±0.01ms ? ?/sec 1.06 3.2±0.06ms ? ?/sec
nested_struct/select_struct 1.00 4.0±0.11ms ? ?/sec 1.14 4.6±0.42ms ? ?/sec
nested_struct/sum_inner_small_field 1.00 662.1±2.13µs ? ?/sec 1.75 1160.8±10.90µs ? ?/sec
wide_struct/select_all 1.05 10.0±0.14ms ? ?/sec 1.00 9.5±0.08ms ? ?/sec
wide_struct/select_one_string_field 1.00 670.6±2.60µs ? ?/sec 12.30 8.3±0.18ms ? ?/sec
wide_struct/select_small_field 1.00 419.3±1.93µs ? ?/sec 19.48 8.2±0.12ms ? ?/sec
wide_struct/select_struct 1.10 10.7±0.25ms ? ?/sec 1.00 9.7±0.18ms ? ?/sec
wide_struct/select_two_string_fields 1.00 4.7±0.10ms ? ?/sec 2.05 9.7±0.25ms ? ?/sec
wide_struct/sum_small_field 1.00 637.8±2.48µs ? ?/sec 12.83 8.2±0.05ms ? ?/sec
```
</p>
</details>
<details><summary>Resource Usage</summary>
**base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 67.6s |
| Peak memory | 4.3 GiB |
| Avg memory | 4.2 GiB |
| CPU user | 78.8s |
| CPU sys | 29.3s |
| Disk read | 0 B |
| Disk write | 611.5 MiB |
**branch**
| Metric | Value |
|--------|-------|
| Wall time | 66.5s |
| Peak memory | 4.3 GiB |
| Avg memory | 4.3 GiB |
| CPU user | 88.1s |
| CPU sys | 19.8s |
| Disk read | 0 B |
| Disk write | 4.7 MiB |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143133933
🤖 Criterion benchmark completed (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4143001255)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB)
<details><summary>CPU Details (lscpu)</summary>
```
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>
<details><summary>Details</summary>
<p>
```
group friendlymatthew_update-projections-for-structs main
----- ---------------------------------------------- ----
narrow_struct/select_all 1.00 551.5±2.00µs ? ?/sec 5.02 2.8±0.04ms ? ?/sec
narrow_struct/select_id_and_small_field 1.00 435.1±7.31µs ? ?/sec 1.84 801.7±7.38µs ? ?/sec
narrow_struct/select_large_field 1.00 631.1±4.48µs ? ?/sec 4.29 2.7±0.07ms ? ?/sec
narrow_struct/select_small_field 1.00 395.7±4.84µs ? ?/sec 1.88 744.2±5.04µs ? ?/sec
narrow_struct/select_struct 1.00 524.3±8.02µs ? ?/sec 5.06 2.7±0.05ms ? ?/sec
narrow_struct/sum_small_field 1.00 621.9±2.96µs ? ?/sec 1.54 957.3±3.68µs ? ?/sec
nested_struct/select_all 1.00 4.2±0.20ms ? ?/sec 1.29 5.5±0.09ms ? ?/sec
nested_struct/select_extra_string 1.00 700.1±6.72µs ? ?/sec 6.38 4.5±0.13ms ? ?/sec
nested_struct/select_inner_small_field 1.00 432.4±2.16µs ? ?/sec 12.43 5.4±0.12ms ? ?/sec
nested_struct/select_inner_struct 1.00 3.3±0.10ms ? ?/sec 1.37 4.5±0.19ms ? ?/sec
nested_struct/select_struct 1.00 4.2±0.09ms ? ?/sec 1.31 5.6±0.14ms ? ?/sec
nested_struct/sum_inner_small_field 1.00 638.0±3.33µs ? ?/sec 9.60 6.1±0.48ms ? ?/sec
wide_struct/select_all 1.00 9.2±0.25ms ? ?/sec 1.14 10.5±0.27ms ? ?/sec
wide_struct/select_one_string_field 1.00 687.5±2.23µs ? ?/sec 14.10 9.7±0.18ms ? ?/sec
wide_struct/select_small_field 1.00 400.9±2.16µs ? ?/sec 24.43 9.8±0.19ms ? ?/sec
wide_struct/select_struct 1.00 9.9±0.18ms ? ?/sec 1.03 10.1±0.16ms ? ?/sec
wide_struct/select_two_string_fields 1.00 4.1±0.11ms ? ?/sec 2.57 10.6±0.29ms ? ?/sec
wide_struct/sum_small_field 1.00 624.4±9.24µs ? ?/sec 17.29 10.8±0.19ms ? ?/sec
```
</p>
</details>
<details><summary>Resource Usage</summary>
**base (merge-base)**
| Metric | Value |
|--------|-------|
| Wall time | 69.5s |
| Peak memory | 4.3 GiB |
| Avg memory | 4.2 GiB |
| CPU user | 68.2s |
| CPU sys | 41.9s |
| Disk read | 0 B |
| Disk write | 612.9 MiB |
**branch**
| Metric | Value |
|--------|-------|
| Wall time | 66.0s |
| Peak memory | 4.3 GiB |
| Avg memory | 4.3 GiB |
| CPU user | 87.8s |
| CPU sys | 19.5s |
| Disk read | 0 B |
| Disk write | 4.6 MiB |
</details>
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4137725898
🤖 Criterion benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4137698911)
`Linux bench-c4137698911-567-67wr6 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
Comparing friendlymatthew/update-projections-for-structs (63724faa17280cfe4ab2ccf8220406d8c4084cd2) to 1416ed4 (merge-base) [diff](https://github.com/apache/datafusion/compare/1416ed4d5007180136dae0aaeb921f0681ec001e..63724faa17280cfe4ab2ccf8220406d8c4084cd2)
BENCH_NAME=parquet_struct_projection
BENCH_COMMAND=cargo bench --features=parquet --bench parquet_struct_projection
BENCH_FILTER=
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb commented on code in PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#discussion_r2990369821
##########
datafusion/datasource-parquet/src/row_filter.rs:
##########
@@ -579,6 +595,59 @@ pub(crate) fn build_parquet_read_plan(
)))
}
+/// Builds a unified [`ParquetReadPlan`] for a set of projection expressions
+///
+/// Unlike [`build_parquet_read_plan`] (which is used for filter pushdown and
+/// returns `None` when an expression references unsupported nested types or
+/// missing columns), this function always succeeds. It collects every column
+/// that *can* be resolved in the file and produces a leaf-level projection
+/// mask. Columns missing from the file are silently skipped since the projection
+/// layer handles those by inserting nulls.
Review Comment:
If we never expect to be called with columns missing from the file, shouldn't we error if we are?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangbot (via GitHub)" <gi...@apache.org>.
adriangbot commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4143013455
🤖 Criterion benchmark running (GKE) | [trigger](https://github.com/apache/datafusion/pull/20925#issuecomment-4143001255)
**Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux bench-c4143001255-581-ln9lr 6.12.55+ #1 SMP Sun Feb 1 08:59:41 UTC 2026 aarch64 GNU/Linux`
<details><summary>CPU Details (lscpu)</summary>
```
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 friendlymatthew/update-projections-for-structs (4e3eca988796d4747570111a4365b187c1ad406b) to 37cd3de (merge-base) [diff](https://github.com/apache/datafusion/compare/37cd3de82fcfa7619b04cfb9f19607ff55d44bc4..4e3eca988796d4747570111a4365b187c1ad406b)
BENCH_NAME=parquet_struct_projection
BENCH_COMMAND=cargo bench --features=parquet --bench parquet_struct_projection
BENCH_FILTER=
Results will be posted here when complete
---
[File an issue](https://github.com/adriangb/datafusion-benchmarking/issues) against this benchmark runner
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "friendlymatthew (via GitHub)" <gi...@apache.org>.
friendlymatthew commented on PR #20925:
URL: https://github.com/apache/datafusion/pull/20925#issuecomment-4142992979
run benchmark parquet_struct_projection
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org
Re: [PR] Use leaf level `ProjectionMask` for parquet projections [datafusion]
Posted by "adriangb (via GitHub)" <gi...@apache.org>.
adriangb merged PR #20925:
URL: https://github.com/apache/datafusion/pull/20925
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org