Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-15 Thread via GitHub


alamb commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4460587549

   Thanks for the diligence everyone. This is great


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-15 Thread via GitHub


alamb merged PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-14 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4455067334

   Running manually on an ARM instance
   ```
   group   branch   
  branch2main2
   -   --   
  ----
   decode metadata (wide) with schema  1.02 26.4±0.51ms 
   ? ?/sec1.00 26.0±0.43ms? ?/sec1.03 26.8±0.54ms   
 ? ?/sec
   decode metadata (wide) with skip PES1.02 26.6±0.78ms 
   ? ?/sec1.09 28.3±0.39ms? ?/sec1.00 26.0±0.56ms   
 ? ?/sec
   decode metadata (wide) with skip all stats  1.12 32.4±0.26ms 
   ? ?/sec1.02 29.6±0.44ms? ?/sec1.00 28.9±0.46ms   
 ? ?/sec
   decode metadata (wide) with skip column stats   1.03 28.9±0.32ms 
   ? ?/sec1.00 28.1±0.49ms? ?/sec1.01 28.3±0.45ms   
 ? ?/sec
   decode metadata (wide) with skip size stats 1.07 31.7±0.43ms 
   ? ?/sec1.00 29.7±0.49ms? ?/sec1.04 30.7±0.68ms   
 ? ?/sec
   decode metadata (wide) with stats mask  1.02 25.8±0.56ms 
   ? ?/sec1.00 25.2±0.42ms? ?/sec1.03 26.0±0.50ms   
 ? ?/sec
   decode metadata with schema 1.02  4.1±0.01µs 
   ? ?/sec1.02  4.1±0.01µs? ?/sec1.00  4.0±0.02µs   
 ? ?/sec
   decode metadata with skip PES   1.00  6.9±0.04µs 
   ? ?/sec1.00  7.0±0.01µs? ?/sec1.00  6.9±0.04µs   
 ? ?/sec
   decode metadata with skip column stats  1.00  6.8±0.02µs 
   ? ?/sec1.01  6.9±0.09µs? ?/sec1.01  6.9±0.04µs   
 ? ?/sec
   decode metadata with stats mask 1.00  6.9±0.05µs 
   ? ?/sec1.00  6.9±0.02µs? ?/sec1.01  6.9±0.04µs   
 ? ?/sec
   decode parquet metadata 1.00  7.1±0.02µs 
   ? ?/sec1.00  7.2±0.02µs? ?/sec1.01  7.2±0.04µs   
 ? ?/sec
   decode parquet metadata (wide)  1.03 28.7±0.47ms 
   ? ?/sec1.00 27.8±0.35ms? ?/sec1.05 29.2±0.53ms   
 ? ?/sec
   decode parquet metadata no path_in_schema (wide)1.04 28.4±0.54ms 
   ? ?/sec1.00 27.3±0.54ms? ?/sec1.03 28.2±0.49ms   
 ? ?/sec
   decode parquet metadata w/ size stats (wide)1.04 35.6±0.42ms 
   ? ?/sec1.00 34.3±0.44ms? ?/sec1.04 35.6±0.49ms   
 ? ?/sec
   open(default)   1.00  7.8±0.05µs 
   ? ?/sec1.00  7.8±0.02µs? ?/sec1.00  7.8±0.04µs   
 ? ?/sec
   open(page index)1.00132.0±0.22µs 
   ? ?/sec1.00132.0±0.18µs? ?/sec1.00132.0±0.23µs   
 ? ?/sec
   
   ```
   picked the best run of the current main branch vs 2 runs of this branch. I 
think we can conclude it's just noise, and if there is a difference it's not 
really showing up strongly (although on x86 there's a definite advantage to 
this branch).
   
   Thanks @alamb!


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-14 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4454089542

   > > Thanks @alamb. I'll see if I can hunt up some AWS time and see if this 
is really a thing on ARM.
   > 
   > I can provide access to an AWS machine -- DM or email me a public SSH key 
and i'll setup one for you to test on
   
   Ooh, awesome! Thanks! Just sent you one via email. 


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-14 Thread via GitHub


alamb commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4453707530

   > Thanks @alamb. I'll see if I can hunt up some AWS time and see if this is 
really a thing on ARM.
   
   I can provide access to an AWS machine -- DM or email me a public SSH key 
and i'll setup one for you to test on 


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-14 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4452463877

   Thanks @alamb. I'll see if I can hunt up some AWS time and see if this is 
really a thing on ARM.


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-13 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4445241078

   🤖 Arrow criterion benchmark completed (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4445185505)
   
   **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
   
   
   ```
   group   main 
  thrift_bool
   -    
  ---
   decode metadata (wide) with schema  1.00 25.9±0.42ms 
   ? ?/sec1.03 26.7±0.30ms? ?/sec
   decode metadata (wide) with skip PES1.00 25.8±0.28ms 
   ? ?/sec1.05 27.0±0.17ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 28.2±0.41ms 
   ? ?/sec1.07 30.2±0.33ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 26.8±0.38ms 
   ? ?/sec1.06 28.3±0.34ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 30.0±0.36ms 
   ? ?/sec1.01 30.4±0.18ms? ?/sec
   decode metadata (wide) with stats mask  1.00 25.7±0.34ms 
   ? ?/sec1.02 26.3±0.24ms? ?/sec
   decode metadata with schema 1.00  4.1±0.02µs 
   ? ?/sec1.04  4.2±0.01µs? ?/sec
   decode metadata with skip PES   1.00  6.8±0.05µs 
   ? ?/sec1.03  7.1±0.05µs? ?/sec
   decode metadata with skip column stats  1.00  6.7±0.08µs 
   ? ?/sec1.03  6.9±0.04µs? ?/sec
   decode metadata with stats mask 1.00  6.8±0.07µs 
   ? ?/sec1.02  6.9±0.04µs? ?/sec
   decode parquet metadata 1.00  7.1±0.03µs 
   ? ?/sec1.02  7.2±0.01µs? ?/sec
   decode parquet metadata (wide)  1.00 28.0±0.36ms 
   ? ?/sec1.03 28.9±0.37ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 27.2±0.33ms 
   ? ?/sec1.03 27.9±0.26ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 35.1±0.40ms 
   ? ?/sec1.00 35.1±0.18ms? ?/sec
   open(default)   1.00  7.7±0.09µs 
   ? ?/sec1.02  7.9±0.01µs? ?/sec
   open(p

Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-13 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4445204263

   🤖 Arrow criterion benchmark running (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4445185505)
   **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux 
bench-c4445185505-49-9dv7c 6.12.68+ #1 SMP Wed Apr  1 02:23:28 UTC 2026 aarch64 
GNU/Linux`
   CPU Details (lscpu)
   
   ```
   Architecture:aarch64
   CPU op-mode(s):  64-bit
   Byte Order:  Little Endian
   CPU(s):  16
   On-line CPU(s) list: 0-15
   Vendor ID:   ARM
   Model name:  Neoverse-V2
   Model:   1
   Thread(s) per core:  1
   Core(s) per cluster: 16
   Socket(s):   -
   Cluster(s):  1
   Stepping:r0p1
   BogoMIPS:2000.00
   Flags:   fp asimd evtstrm aes pmull sha1 
sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 
sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 
sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm 
bf16 dgh rng bti
   L1d cache:   1 MiB (16 instances)
   L1i cache:   1 MiB (16 instances)
   L2 cache:32 MiB (16 instances)
   L3 cache:80 MiB (1 instance)
   NUMA node(s):1
   NUMA node0 CPU(s):   0-15
   Vulnerability Gather data sampling:  Not affected
   Vulnerability Indirect target selection: Not affected
   Vulnerability Itlb multihit: Not affected
   Vulnerability L1tf:  Not affected
   Vulnerability Mds:   Not affected
   Vulnerability Meltdown:  Not affected
   Vulnerability Mmio stale data:   Not affected
   Vulnerability Reg file data sampling:Not affected
   Vulnerability Retbleed:  Not affected
   Vulnerability Spec rstack overflow:  Not affected
   Vulnerability Spec store bypass: Mitigation; Speculative Store 
Bypass disabled via prctl
   Vulnerability Spectre v1:Mitigation; __user pointer 
sanitization
   Vulnerability Spectre v2:Mitigation; CSV2, BHB
   Vulnerability Srbds: Not affected
   Vulnerability Tsa:   Not affected
   Vulnerability Tsx async abort:   Not affected
   Vulnerability Vmscape:   Not affected
   ```
   
   
   
   Comparing thrift_bool (876fc37be252ad144e020e94dfa031f031e08887) to 1ffd202 
(merge-base) 
[diff](https://github.com/apache/arrow-rs/compare/1ffd202aeef1af2a3b0ab2b4ae604a9edbac6e09..876fc37be252ad144e020e94dfa031f031e08887)
   BENCH_NAME=metadata
   BENCH_COMMAND=cargo bench 
--features=arrow,async,test_common,experimental,object_store --bench metadata
   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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-13 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4445185505

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-12 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4432401208

   > I ran the benchmarks locally on my machine (Intel Tigerlake), and this 
branch seems to be around 5% faster on all metadata benchmarks, compared to 
`main`.
   
   Same, but all on intel as well. I've been playing around on godbolt and 
honestly do not understand why there's any difference at all. However I write 
`read_field_begin` it always boils down to:
   ```rust
   let field_type = self.read_byte()?;
   if field_type == 0 {
  // exit early
   } else if field_type < 16 { // unlikely
  // read full i16 field_id and return
   } else {
  // compute field_id from delta and return
   }
   ```
   Lacking an aarch64 machine to profile on, it's hard to see why the 
benchmarks here are so different. It's possible that the difference comes from 
either the new `bool_val()` function or some quirk in the error handling. 


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-12 Thread via GitHub


jhorstmann commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4428709246

   I ran the benchmarks locally on my machine (Intel Tigerlake), and this 
branch seems to be around 5% faster on all metadata benchmarks, compared to 
`main`.


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4426773523

   🤖 Arrow criterion benchmark completed (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4426738697)
   
   **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
   
   
   ```
   group   main 
  thrift_bool
   -    
  ---
   decode metadata (wide) with schema  1.00 26.4±0.47ms 
   ? ?/sec1.02 26.8±0.26ms? ?/sec
   decode metadata (wide) with skip PES1.00 26.1±0.43ms 
   ? ?/sec1.05 27.4±0.24ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 28.0±0.36ms 
   ? ?/sec1.08 30.3±0.25ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 27.2±0.40ms 
   ? ?/sec1.04 28.2±0.27ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 30.3±0.48ms 
   ? ?/sec1.02 30.8±0.26ms? ?/sec
   decode metadata (wide) with stats mask  1.00 26.0±0.41ms 
   ? ?/sec1.02 26.4±0.22ms? ?/sec
   decode metadata with schema 1.00  4.0±0.01µs 
   ? ?/sec1.04  4.2±0.01µs? ?/sec
   decode metadata with skip PES   1.00  6.9±0.01µs 
   ? ?/sec1.03  7.1±0.01µs? ?/sec
   decode metadata with skip column stats  1.00  6.7±0.07µs 
   ? ?/sec1.01  6.8±0.01µs? ?/sec
   decode metadata with stats mask 1.00  6.8±0.01µs 
   ? ?/sec1.03  7.0±0.04µs? ?/sec
   decode parquet metadata 1.00  7.1±0.02µs 
   ? ?/sec1.02  7.3±0.01µs? ?/sec
   decode parquet metadata (wide)  1.00 28.3±0.48ms 
   ? ?/sec1.02 28.9±0.31ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 27.6±0.48ms 
   ? ?/sec1.02 28.3±0.30ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 35.5±0.43ms 
   ? ?/sec1.00 35.3±0.21ms? ?/sec
   open(default)   1.00  7.7±0.03µs 
   ? ?/sec1.03  7.9±0.02µs? ?/sec
   open(p

Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4426746300

   🤖 Arrow criterion benchmark running (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4426738697)
   **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux 
bench-c4426738697-17-xj782 6.12.68+ #1 SMP Wed Apr  1 02:23:28 UTC 2026 aarch64 
GNU/Linux`
   CPU Details (lscpu)
   
   ```
   Architecture:aarch64
   CPU op-mode(s):  64-bit
   Byte Order:  Little Endian
   CPU(s):  16
   On-line CPU(s) list: 0-15
   Vendor ID:   ARM
   Model name:  Neoverse-V2
   Model:   1
   Thread(s) per core:  1
   Core(s) per cluster: 16
   Socket(s):   -
   Cluster(s):  1
   Stepping:r0p1
   BogoMIPS:2000.00
   Flags:   fp asimd evtstrm aes pmull sha1 
sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 
sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 
sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm 
bf16 dgh rng bti
   L1d cache:   1 MiB (16 instances)
   L1i cache:   1 MiB (16 instances)
   L2 cache:32 MiB (16 instances)
   L3 cache:80 MiB (1 instance)
   NUMA node(s):1
   NUMA node0 CPU(s):   0-15
   Vulnerability Gather data sampling:  Not affected
   Vulnerability Indirect target selection: Not affected
   Vulnerability Itlb multihit: Not affected
   Vulnerability L1tf:  Not affected
   Vulnerability Mds:   Not affected
   Vulnerability Meltdown:  Not affected
   Vulnerability Mmio stale data:   Not affected
   Vulnerability Reg file data sampling:Not affected
   Vulnerability Retbleed:  Not affected
   Vulnerability Spec rstack overflow:  Not affected
   Vulnerability Spec store bypass: Mitigation; Speculative Store 
Bypass disabled via prctl
   Vulnerability Spectre v1:Mitigation; __user pointer 
sanitization
   Vulnerability Spectre v2:Mitigation; CSV2, BHB
   Vulnerability Srbds: Not affected
   Vulnerability Tsa:   Not affected
   Vulnerability Tsx async abort:   Not affected
   Vulnerability Vmscape:   Not affected
   ```
   
   
   
   Comparing thrift_bool (99de9817d9dcaa63c6f4842326fe6ab4b8f6f972) to 7abb225 
(merge-base) 
[diff](https://github.com/apache/arrow-rs/compare/7abb2255f2de045ca8dcd12e1e46377e11de5b9f..99de9817d9dcaa63c6f4842326fe6ab4b8f6f972)
   BENCH_NAME=metadata
   BENCH_COMMAND=cargo bench 
--features=arrow,async,test_common,experimental,object_store --bench metadata
   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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4426738697

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425743233

   🤖 Arrow criterion benchmark completed (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425693420)
   
   **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
   
   
   ```
   group   main 
  thrift_bool
   -    
  ---
   decode metadata (wide) with schema  1.00 27.2±0.48ms 
   ? ?/sec1.02 27.7±0.47ms? ?/sec
   decode metadata (wide) with skip PES1.00 26.7±0.38ms 
   ? ?/sec1.02 27.4±0.33ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 28.5±0.50ms 
   ? ?/sec1.02 29.1±0.40ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 27.6±0.41ms 
   ? ?/sec1.03 28.4±0.36ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 30.8±0.46ms 
   ? ?/sec1.02 31.3±0.40ms? ?/sec
   decode metadata (wide) with stats mask  1.00 26.9±0.49ms 
   ? ?/sec1.02 27.4±0.40ms? ?/sec
   decode metadata with schema 1.00  4.0±0.02µs 
   ? ?/sec1.02  4.1±0.02µs? ?/sec
   decode metadata with skip PES   1.00  6.8±0.07µs 
   ? ?/sec1.00  6.8±0.01µs? ?/sec
   decode metadata with skip column stats  1.01  6.7±0.02µs 
   ? ?/sec1.00  6.7±0.02µs? ?/sec
   decode metadata with stats mask 1.00  6.8±0.07µs 
   ? ?/sec1.02  6.9±0.03µs? ?/sec
   decode parquet metadata 1.00  7.0±0.02µs 
   ? ?/sec1.03  7.2±0.02µs? ?/sec
   decode parquet metadata (wide)  1.00 29.2±0.49ms 
   ? ?/sec1.02 29.7±0.46ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 28.3±0.43ms 
   ? ?/sec1.02 28.8±0.44ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 35.8±0.41ms 
   ? ?/sec1.01 36.3±0.37ms? ?/sec
   open(default)   1.00  7.6±0.02µs 
   ? ?/sec1.02  7.8±0.03µs? ?/sec
   open(p

Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425715683

   🤖 Arrow criterion benchmark running (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425693420)
   **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux 
bench-c4425693420-16-dslmw 6.12.68+ #1 SMP Wed Apr  1 02:23:28 UTC 2026 aarch64 
GNU/Linux`
   CPU Details (lscpu)
   
   ```
   Architecture:aarch64
   CPU op-mode(s):  64-bit
   Byte Order:  Little Endian
   CPU(s):  16
   On-line CPU(s) list: 0-15
   Vendor ID:   ARM
   Model name:  Neoverse-V2
   Model:   1
   Thread(s) per core:  1
   Core(s) per cluster: 16
   Socket(s):   -
   Cluster(s):  1
   Stepping:r0p1
   BogoMIPS:2000.00
   Flags:   fp asimd evtstrm aes pmull sha1 
sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 
sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 
sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm 
bf16 dgh rng bti
   L1d cache:   1 MiB (16 instances)
   L1i cache:   1 MiB (16 instances)
   L2 cache:32 MiB (16 instances)
   L3 cache:80 MiB (1 instance)
   NUMA node(s):1
   NUMA node0 CPU(s):   0-15
   Vulnerability Gather data sampling:  Not affected
   Vulnerability Indirect target selection: Not affected
   Vulnerability Itlb multihit: Not affected
   Vulnerability L1tf:  Not affected
   Vulnerability Mds:   Not affected
   Vulnerability Meltdown:  Not affected
   Vulnerability Mmio stale data:   Not affected
   Vulnerability Reg file data sampling:Not affected
   Vulnerability Retbleed:  Not affected
   Vulnerability Spec rstack overflow:  Not affected
   Vulnerability Spec store bypass: Mitigation; Speculative Store 
Bypass disabled via prctl
   Vulnerability Spectre v1:Mitigation; __user pointer 
sanitization
   Vulnerability Spectre v2:Mitigation; CSV2, BHB
   Vulnerability Srbds: Not affected
   Vulnerability Tsa:   Not affected
   Vulnerability Tsx async abort:   Not affected
   Vulnerability Vmscape:   Not affected
   ```
   
   
   
   Comparing thrift_bool (00326ea4dc02309428ac2cdfb82a9ac7d0c29303) to 7abb225 
(merge-base) 
[diff](https://github.com/apache/arrow-rs/compare/7abb2255f2de045ca8dcd12e1e46377e11de5b9f..00326ea4dc02309428ac2cdfb82a9ac7d0c29303)
   BENCH_NAME=metadata
   BENCH_COMMAND=cargo bench 
--features=arrow,async,test_common,experimental,object_store --bench metadata
   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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425693420

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425489201

   🤖 Arrow criterion benchmark completed (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425427881)
   
   **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
   
   
   ```
   group   main 
  thrift_bool
   -    
  ---
   decode metadata (wide) with schema  1.00 26.8±0.43ms 
   ? ?/sec1.03 27.6±0.34ms? ?/sec
   decode metadata (wide) with skip PES1.00 26.4±0.44ms 
   ? ?/sec1.04 27.4±0.33ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 28.3±0.45ms 
   ? ?/sec1.02 29.0±0.36ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 27.6±0.45ms 
   ? ?/sec1.03 28.3±0.34ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 30.7±0.47ms 
   ? ?/sec1.02 31.3±0.34ms? ?/sec
   decode metadata (wide) with stats mask  1.00 26.5±0.44ms 
   ? ?/sec1.03 27.2±0.32ms? ?/sec
   decode metadata with schema 1.00  4.0±0.01µs 
   ? ?/sec1.01  4.1±0.02µs? ?/sec
   decode metadata with skip PES   1.00  6.9±0.03µs 
   ? ?/sec1.02  7.0±0.07µs? ?/sec
   decode metadata with skip column stats  1.00  6.7±0.03µs 
   ? ?/sec1.01  6.7±0.02µs? ?/sec
   decode metadata with stats mask 1.00  6.8±0.03µs 
   ? ?/sec1.02  6.9±0.01µs? ?/sec
   decode parquet metadata 1.00  7.0±0.02µs 
   ? ?/sec1.03  7.3±0.02µs? ?/sec
   decode parquet metadata (wide)  1.00 29.1±0.48ms 
   ? ?/sec1.01 29.4±0.46ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 28.2±0.49ms 
   ? ?/sec1.02 28.8±0.49ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 35.8±0.38ms 
   ? ?/sec1.01 36.0±0.32ms? ?/sec
   open(default)   1.00  7.7±0.03µs 
   ? ?/sec1.02  7.9±0.05µs? ?/sec
   open(p

Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425453386

   https://github.com/adriangb/datafusion-benchmarking/pull/7 seems to have 
fixed benchmarks


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425448787

   🤖 Arrow criterion benchmark running (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425427881)
   **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux 
bench-c4425427881-15-czwcp 6.12.68+ #1 SMP Wed Apr  1 02:23:28 UTC 2026 aarch64 
GNU/Linux`
   CPU Details (lscpu)
   
   ```
   Architecture:aarch64
   CPU op-mode(s):  64-bit
   Byte Order:  Little Endian
   CPU(s):  16
   On-line CPU(s) list: 0-15
   Vendor ID:   ARM
   Model name:  Neoverse-V2
   Model:   1
   Thread(s) per core:  1
   Core(s) per cluster: 16
   Socket(s):   -
   Cluster(s):  1
   Stepping:r0p1
   BogoMIPS:2000.00
   Flags:   fp asimd evtstrm aes pmull sha1 
sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 
sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 
sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm 
bf16 dgh rng bti
   L1d cache:   1 MiB (16 instances)
   L1i cache:   1 MiB (16 instances)
   L2 cache:32 MiB (16 instances)
   L3 cache:80 MiB (1 instance)
   NUMA node(s):1
   NUMA node0 CPU(s):   0-15
   Vulnerability Gather data sampling:  Not affected
   Vulnerability Indirect target selection: Not affected
   Vulnerability Itlb multihit: Not affected
   Vulnerability L1tf:  Not affected
   Vulnerability Mds:   Not affected
   Vulnerability Meltdown:  Not affected
   Vulnerability Mmio stale data:   Not affected
   Vulnerability Reg file data sampling:Not affected
   Vulnerability Retbleed:  Not affected
   Vulnerability Spec rstack overflow:  Not affected
   Vulnerability Spec store bypass: Mitigation; Speculative Store 
Bypass disabled via prctl
   Vulnerability Spectre v1:Mitigation; __user pointer 
sanitization
   Vulnerability Spectre v2:Mitigation; CSV2, BHB
   Vulnerability Srbds: Not affected
   Vulnerability Tsa:   Not affected
   Vulnerability Tsx async abort:   Not affected
   Vulnerability Vmscape:   Not affected
   ```
   
   
   
   Comparing thrift_bool (f58ff9be29b053c07c83ac35fa77b48a4c1effaa) to 7abb225 
(merge-base) 
[diff](https://github.com/apache/arrow-rs/compare/7abb2255f2de045ca8dcd12e1e46377e11de5b9f..f58ff9be29b053c07c83ac35fa77b48a4c1effaa)
   BENCH_NAME=metadata
   BENCH_COMMAND=cargo bench 
--features=arrow,async,test_common,experimental,object_store --bench metadata
   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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-11 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4425427881

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-08 Thread via GitHub


alamb commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4409558451

   Benchmarks seem not to be working. Maybe related to this
   - https://github.com/adriangb/datafusion-benchmarking/pull/5


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-08 Thread via GitHub


alamb commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4407300963

   run benchmarks metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-07 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4404030421

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-07 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4400226858

   bench results look weird. On x86 WS got
   ```
   group   bool 
  main
   -    
  
   decode metadata (wide) with schema  1.00 15.5±0.12ms 
   ? ?/sec1.09 16.9±0.13ms? ?/sec
   decode metadata (wide) with skip PES1.00 15.6±0.23ms 
   ? ?/sec1.09 17.1±0.23ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 18.0±0.10ms 
   ? ?/sec1.08 19.4±0.10ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 17.0±0.11ms 
   ? ?/sec1.08 18.3±0.15ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 18.2±0.12ms 
   ? ?/sec1.11 20.1±0.11ms? ?/sec
   decode metadata (wide) with stats mask  1.00 15.6±0.10ms 
   ? ?/sec1.09 17.1±0.11ms? ?/sec
   decode metadata with schema 1.00  2.6±0.01µs 
   ? ?/sec1.05  2.8±0.02µs? ?/sec
   decode metadata with skip PES   1.00  4.5±0.06µs 
   ? ?/sec1.04  4.7±0.05µs? ?/sec
   decode metadata with skip column stats  1.00  4.7±0.02µs 
   ? ?/sec1.03  4.8±0.05µs? ?/sec
   decode metadata with stats mask 1.00  4.5±0.06µs 
   ? ?/sec1.05  4.7±0.06µs? ?/sec
   decode parquet metadata 1.00  4.8±0.05µs 
   ? ?/sec1.06  5.1±0.02µs? ?/sec
   decode parquet metadata (wide)  1.00 17.3±0.31ms 
   ? ?/sec1.08 18.7±0.24ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 16.7±0.08ms 
   ? ?/sec1.09 18.2±0.11ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 21.0±0.19ms 
   ? ?/sec1.07 22.4±0.28ms? ?/sec
   open(default)   1.00  5.3±0.02µs 
   ? ?/sec1.04  5.5±0.03µs? ?/sec
   open(page index)1.00 94.4±1.19µs 
   ? ?/sec1.02 96.4±2.76µs? ?/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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-07 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4400059665

   🤖 Arrow criterion benchmark completed (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-431702)
   
   **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
   
   
   ```
   group   main 
  thrift_bool
   -    
  ---
   decode metadata (wide) with schema  1.00 26.9±0.49ms 
   ? ?/sec1.02 27.4±0.32ms? ?/sec
   decode metadata (wide) with skip PES1.00 26.5±0.47ms 
   ? ?/sec1.02 27.1±0.33ms? ?/sec
   decode metadata (wide) with skip all stats  1.00 28.1±0.42ms 
   ? ?/sec1.02 28.7±0.32ms? ?/sec
   decode metadata (wide) with skip column stats   1.00 27.8±0.51ms 
   ? ?/sec1.01 28.1±0.28ms? ?/sec
   decode metadata (wide) with skip size stats 1.00 30.8±0.50ms 
   ? ?/sec1.02 31.4±0.28ms? ?/sec
   decode metadata (wide) with stats mask  1.00 26.6±0.53ms 
   ? ?/sec1.01 26.8±0.31ms? ?/sec
   decode metadata with schema 1.00  4.0±0.02µs 
   ? ?/sec1.02  4.1±0.01µs? ?/sec
   decode metadata with skip PES   1.00  6.8±0.07µs 
   ? ?/sec1.01  6.9±0.05µs? ?/sec
   decode metadata with skip column stats  1.01  6.7±0.02µs 
   ? ?/sec1.00  6.7±0.02µs? ?/sec
   decode metadata with stats mask 1.00  6.8±0.07µs 
   ? ?/sec1.02  6.9±0.05µs? ?/sec
   decode parquet metadata 1.00  7.1±0.02µs 
   ? ?/sec1.02  7.2±0.02µs? ?/sec
   decode parquet metadata (wide)  1.00 29.0±0.59ms 
   ? ?/sec1.02 29.6±0.35ms? ?/sec
   decode parquet metadata no path_in_schema (wide)1.00 28.4±0.47ms 
   ? ?/sec1.01 28.7±0.34ms? ?/sec
   decode parquet metadata w/ size stats (wide)1.00 35.9±0.53ms 
   ? ?/sec1.01 36.2±0.32ms? ?/sec
   open(default)   1.00  7.7±0.02µs 
   ? ?/sec1.02  7.8±0.02µs? ?/sec
   open(p

Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-07 Thread via GitHub


adriangbot commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4400015440

   🤖 Arrow criterion benchmark running (GKE) | 
[trigger](https://github.com/apache/arrow-rs/pull/9945#issuecomment-431702)
   **Instance:** `c4a-highmem-16` (12 vCPU / 65 GiB) | `Linux 
bench-c431702-2062-vv7n4 6.12.68+ #1 SMP Wed Apr  1 02:23:28 UTC 2026 
aarch64 GNU/Linux`
   CPU Details (lscpu)
   
   ```
   Architecture:aarch64
   CPU op-mode(s):  64-bit
   Byte Order:  Little Endian
   CPU(s):  16
   On-line CPU(s) list: 0-15
   Vendor ID:   ARM
   Model name:  Neoverse-V2
   Model:   1
   Thread(s) per core:  1
   Core(s) per cluster: 16
   Socket(s):   -
   Cluster(s):  1
   Stepping:r0p1
   BogoMIPS:2000.00
   Flags:   fp asimd evtstrm aes pmull sha1 
sha2 crc32 atomics fphp asimdhp cpuid asimdrdm jscvt fcma lrcpc dcpop sha3 sm3 
sm4 asimddp sha512 sve asimdfhm dit uscat ilrcpc flagm sb paca pacg dcpodp sve2 
sveaes svepmull svebitperm svesha3 svesm4 flagm2 frint svei8mm svebf16 i8mm 
bf16 dgh rng bti
   L1d cache:   1 MiB (16 instances)
   L1i cache:   1 MiB (16 instances)
   L2 cache:32 MiB (16 instances)
   L3 cache:80 MiB (1 instance)
   NUMA node(s):1
   NUMA node0 CPU(s):   0-15
   Vulnerability Gather data sampling:  Not affected
   Vulnerability Indirect target selection: Not affected
   Vulnerability Itlb multihit: Not affected
   Vulnerability L1tf:  Not affected
   Vulnerability Mds:   Not affected
   Vulnerability Meltdown:  Not affected
   Vulnerability Mmio stale data:   Not affected
   Vulnerability Reg file data sampling:Not affected
   Vulnerability Retbleed:  Not affected
   Vulnerability Spec rstack overflow:  Not affected
   Vulnerability Spec store bypass: Mitigation; Speculative Store 
Bypass disabled via prctl
   Vulnerability Spectre v1:Mitigation; __user pointer 
sanitization
   Vulnerability Spectre v2:Mitigation; CSV2, BHB
   Vulnerability Srbds: Not affected
   Vulnerability Tsa:   Not affected
   Vulnerability Tsx async abort:   Not affected
   Vulnerability Vmscape:   Not affected
   ```
   
   
   
   Comparing thrift_bool (1081e7afe510794fc9eb9db35ca25c9bb0dece65) to 6ce4bc8 
(merge-base) 
[diff](https://github.com/apache/arrow-rs/compare/6ce4bc899644c1b0816072f3260d835a0c14d148..1081e7afe510794fc9eb9db35ca25c9bb0dece65)
   BENCH_NAME=metadata
   BENCH_COMMAND=cargo bench 
--features=arrow,async,test_common,experimental,object_store --bench metadata
   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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



Re: [PR] perf: Remove `bool_val` from Parquet Thrift `FieldIdentifier` [arrow-rs]

2026-05-07 Thread via GitHub


etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-431702

   run benchmark metadata


-- 
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: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]