ARROW-10540: [Rust] Extended filter kernel to all types and improved performance by jorgecarleitao · Pull Request #8954 · apache/arrow

jorgecarleitao · 2020-12-17T13:18:36Z

This PR improves the filter kernel:

made the filter benchmarks more realistic
performance improved by 1.2-4x for all multi-filter operations
performance decreased by 30% for a single-filter operation
filter now supports all types supported by MutableArrayData
removed 500 LOC

There are two novel ideas here:

it minimizes the number of memcopies when building the filtered array, both for single filter and multi-filter operations.
for single filter operations, it leverages an iterator to create the new array on the fly. For multi filter operations, it persists the iterator's result in a vector and iterates over it per array.

This PR also improves the performance of MutableArrayData by avoiding some bound checks via unsafe (properly documented).

Summary of the benchmarks:

benchmark	variation (%)
filter u8	29.5
filter u8 low selectivity	7.3
filter context u8 w NULLs	-17.5
filter context u8 w NULLs high selectivity	-21.9
filter context f32 high selectivity	-22.0
filter context f32	-26.8
filter context string high selectivity	-27.5
filter context string	-31.4
filter context u8	-40.3
filter u8 high selectivity	-47.3
filter context string low selectivity	-55.3
filter context u8 w NULLs low selectivity	-57.7
filter context f32 low selectivity	-64.8
filter context u8 low selectivity	-66.0
filter context u8 high selectivity	-77.2

Code used to benchmark:

git checkout 54da4378d138df12bd4e09a68b0f4c80218834c7
cargo bench --bench filter_kernels
git checkout mutable_filter2
cargo bench --bench filter_kernels

Benchmark result:

   Compiling arrow v3.0.0-SNAPSHOT (/Users/jorgecarleitao/projects/arrow/rust/arrow)
    Finished bench [optimized] target(s) in 1m 01s
     Running /Users/jorgecarleitao/projects/arrow/rust/target/release/deps/filter_kernels-5208f9a404de52c9
Gnuplot not found, using plotters backend
filter u8               time:   [512.54 us 513.43 us 514.37 us]                      
                        change: [+29.070% +29.548% +30.003%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 5 outliers among 100 measurements (5.00%)
  3 (3.00%) high mild
  2 (2.00%) high severe

filter u8 high selectivity                                                                             
                        time:   [11.494 us 11.513 us 11.532 us]
                        change: [-47.846% -47.337% -46.755%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) high mild
  5 (5.00%) high severe

filter u8 low selectivity                                                                             
                        time:   [7.0342 us 7.0520 us 7.0693 us]
                        change: [+6.5543% +7.3409% +8.1080%] (p = 0.00 < 0.05)
                        Performance has regressed.
Found 5 outliers among 100 measurements (5.00%)
  1 (1.00%) high mild
  4 (4.00%) high severe

filter context u8       time:   [233.81 us 234.31 us 234.93 us]                              
                        change: [-40.715% -40.329% -39.886%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  3 (3.00%) high mild
  5 (5.00%) high severe

filter context u8 high selectivity                                                                             
                        time:   [4.5943 us 4.6100 us 4.6276 us]
                        change: [-77.449% -77.231% -77.022%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 18 outliers among 100 measurements (18.00%)
  10 (10.00%) high mild
  8 (8.00%) high severe

filter context u8 low selectivity                                                                             
                        time:   [1.7582 us 1.7664 us 1.7742 us]
                        change: [-66.250% -65.989% -65.669%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high severe

filter context u8 w NULLs                                                                            
                        time:   [476.99 us 477.71 us 478.44 us]
                        change: [-17.852% -17.457% -17.000%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  3 (3.00%) high mild
  3 (3.00%) high severe

filter context u8 w NULLs high selectivity                                                                            
                        time:   [296.46 us 297.03 us 297.67 us]
                        change: [-22.297% -21.871% -21.393%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  3 (3.00%) high mild
  4 (4.00%) high severe

filter context u8 w NULLs low selectivity                                                                             
                        time:   [2.5988 us 2.6124 us 2.6268 us]
                        change: [-58.065% -57.668% -57.237%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 5 outliers among 100 measurements (5.00%)
  4 (4.00%) high mild
  1 (1.00%) high severe

filter context f32      time:   [470.69 us 472.39 us 474.73 us]                               
                        change: [-29.574% -26.769% -24.242%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 14 outliers among 100 measurements (14.00%)
  9 (9.00%) high mild
  5 (5.00%) high severe

filter context f32 high selectivity                                                                            
                        time:   [307.16 us 307.58 us 308.03 us]
                        change: [-22.472% -22.039% -21.532%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  2 (2.00%) high mild
  4 (4.00%) high severe

filter context f32 low selectivity                                                                             
                        time:   [2.4266 us 2.4323 us 2.4384 us]
                        change: [-65.024% -64.764% -64.517%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  5 (5.00%) high mild
  2 (2.00%) high severe

filter context string   time:   [645.82 us 647.32 us 649.04 us]                                  
                        change: [-31.810% -31.427% -31.046%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  6 (6.00%) high mild
  5 (5.00%) high severe

Benchmarking filter context string high selectivity: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 5.2s, enable flat sampling, or reduce sample count to 60.
filter context string high selectivity                                                                             
                        time:   [999.11 us 1.0008 ms 1.0027 ms]
                        change: [-28.133% -27.524% -26.930%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  7 (7.00%) high mild
  4 (4.00%) high severe

filter context string low selectivity                                                                             
                        time:   [3.6441 us 3.6623 us 3.6799 us]
                        change: [-55.650% -55.329% -55.013%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  6 (6.00%) low mild
  2 (2.00%) high severe

github-actions · 2020-12-17T13:32:28Z

https://issues.apache.org/jira/browse/ARROW-10540

mqy · 2020-12-17T14:36:01Z

Amazing, great work! Here is my bench result on mac book pro 2018:

$ git checkout mutable_filter2
Previous HEAD position was 54da4378d Improved filter bench.
Switched to branch 'mutable_filter2'
Your branch is up to date with 'jorgecarleitao/mutable_filter2'.
$ cargo bench --bench filter_kernels
   Compiling arrow v3.0.0-SNAPSHOT (/Users/mqy/tools/data/arrow/rust/arrow)
    Finished bench [optimized] target(s) in 30.81s
     Running target/release/deps/filter_kernels-9a777356dd1c1b39
Gnuplot not found, using plotters backend
filter u8               time:   [3.3223 us 3.3282 us 3.3340 us]                       
                        change: [-99.062% -99.059% -99.056%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 6 outliers among 100 measurements (6.00%)
  3 (3.00%) low mild
  3 (3.00%) high severe

filter u8 high selectivity                                                                             
                        time:   [6.6873 us 6.8811 us 7.0893 us]
                        change: [-65.458% -64.142% -62.812%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  7 (7.00%) high mild

filter u8 low selectivity                                                                             
                        time:   [2.6626 us 2.6704 us 2.6783 us]
                        change: [-53.949% -53.755% -53.558%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild

filter context u8       time:   [1.6725 us 1.6814 us 1.6902 us]                               
                        change: [-99.525% -99.522% -99.519%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) low mild

filter context u8 high selectivity                                                                             
                        time:   [4.3977 us 4.5770 us 4.7594 us]
                        change: [-75.323% -73.815% -72.286%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  11 (11.00%) high mild

filter context u8 low selectivity                                                                             
                        time:   [1.1797 us 1.1843 us 1.1890 us]
                        change: [-74.379% -74.076% -73.824%] (p = 0.00 < 0.05)
                        Performance has improved.

filter context u8 w NULLs                                                                             
                        time:   [2.0055 us 2.0120 us 2.0187 us]
                        change: [-99.621% -99.619% -99.618%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) low mild
  1 (1.00%) high mild

filter context u8 w NULLs high selectivity                                                                            
                        time:   [243.76 us 244.08 us 244.41 us]
                        change: [-32.511% -32.318% -32.144%] (p = 0.00 < 0.05)
                        Performance has improved.

filter context u8 w NULLs low selectivity                                                                             
                        time:   [1.4073 us 1.4151 us 1.4230 us]
                        change: [-73.174% -72.860% -72.574%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild

filter context f32      time:   [3.7693 us 3.7804 us 3.7918 us]                                
                        change: [-99.341% -99.337% -99.332%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 8 outliers among 100 measurements (8.00%)
  3 (3.00%) low mild
  4 (4.00%) high mild
  1 (1.00%) high severe

filter context f32 high selectivity                                                                            
                        time:   [257.52 us 258.42 us 259.27 us]
                        change: [-31.098% -30.752% -30.429%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high mild

filter context f32 low selectivity                                                                             
                        time:   [1.3838 us 1.3901 us 1.3964 us]
                        change: [-77.474% -77.227% -77.021%] (p = 0.00 < 0.05)
                        Performance has improved.

filter context string   time:   [4.7862 us 4.8040 us 4.8206 us]                                   
                        change: [-99.393% -99.389% -99.386%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 11 outliers among 100 measurements (11.00%)
  1 (1.00%) low mild
  7 (7.00%) high mild
  3 (3.00%) high severe

filter context string high selectivity                                                                            
                        time:   [833.56 us 836.78 us 843.25 us]
                        change: [-20.015% -19.639% -19.202%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 9 outliers among 100 measurements (9.00%)
  2 (2.00%) low mild
  3 (3.00%) high mild
  4 (4.00%) high severe

filter context string low selectivity                                                                             
                        time:   [1.6389 us 1.6486 us 1.6584 us]
                        change: [-76.728% -76.552% -76.376%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high mild

jorgecarleitao · 2020-12-17T15:23:04Z

I am moving this to draft as I do not believe the 200x. It looks too good to be true.

jorgecarleitao added 7 commits December 17, 2020 06:44

Added bench for mutable array.

a174359

Trying unsafe.

0ccaa66

Removed clone.

cc317db

Simplified bit chunk iterator.

1271414

Improved filter bench.

54da437

Migrated filter to mutableArrayData.

bf92235

Improved.

c8586f9

github-actions bot added the Component: Rust label Dec 17, 2020

jorgecarleitao marked this pull request as draft December 17, 2020 15:22

jorgecarleitao changed the title ~~ARROW-10540: [Rust] Extended filter kernel to all types and improved performance (200x)~~ ARROW-10540: [Rust] Extended filter kernel to all types and improved performance Dec 17, 2020

jorgecarleitao closed this Dec 17, 2020

jorgecarleitao mentioned this pull request Dec 19, 2020

[Rust] [Experiment] Vec<u8> vs current allocations #8796

Closed

asfimport mentioned this pull request Dec 24, 2020

[Rust] Allow unary kernels of arbitrary array types #26507

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

ARROW-10540: [Rust] Extended filter kernel to all types and improved performance#8954

ARROW-10540: [Rust] Extended filter kernel to all types and improved performance#8954
jorgecarleitao wants to merge 7 commits intoapache:masterfrom
jorgecarleitao:mutable_filter2

jorgecarleitao commented Dec 17, 2020 •

edited

Loading

Uh oh!

github-actions bot commented Dec 17, 2020

Uh oh!

mqy commented Dec 17, 2020 •

edited

Loading

Uh oh!

jorgecarleitao commented Dec 17, 2020 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

jorgecarleitao commented Dec 17, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Dec 17, 2020

Uh oh!

mqy commented Dec 17, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jorgecarleitao commented Dec 17, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

jorgecarleitao commented Dec 17, 2020 •

edited

Loading

mqy commented Dec 17, 2020 •

edited

Loading

jorgecarleitao commented Dec 17, 2020 •

edited

Loading