BitVector::build_index: 100x speedup #28
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Process the BitVector unit-by-unit instead of bit-by-bit. Use PopCount::count() to update num_1s and use select_bit to find the bit positions for the select0s_ and select1s_ index. According to my benchmarks, the old bit-by-bit version processed a 256kbit vector at about 20MB/s independent of enables_select0/enables_select1. The new version is 50x-150x faster, depending on the compiler and build_index options. enables_select_0=enables_select1=false: popcnt, no bmi2: 1600MB/s popcnt, no bmi2: 2500MB/s popcnt and bmi2: 2900MB/s enables_select_0=enables_select1=true: no popcnt, no bmi2: 1100MB/s popcnt, no bmi2: 1600MB/s popcnt and bmi2: 1800MB/s
glebm
approved these changes
Jun 2, 2020
This is safe and there is no truncation.
jmr
commented
Jun 2, 2020
The 32-bit select_bit will be used to make the new build_index implementation work for MARISA_WORD_SIZE == 32.
This is already used by build_index and fixes the 32-bit build.
jmr
commented
Jun 3, 2020
Owner
Benchmark
The following table shows build speed [1,000 keys/second].
jmr:build-index is 2-3% faster than s-yata:master. |
Contributor
Author
Did you configure with My benchmark was just on I will have time to run/profile the benchmarks myself later in the week. |
Owner
|
The table shows the speed of dictionary construction and |
Owner
|
It looks good tome. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Process the
BitVectorunit-by-unit instead of bit-by-bit.Use
PopCount::count()to updatenum_1sand useselect_bitto findthe bit positions for the
select0s_andselect1s_indexes.According to my benchmarks, the old bit-by-bit version processed a
256kbit vector at about 20MB/s independent of
enables_select0andenables_select1.The new version is 50x-150x faster, depending on the compiler and build_index
options.
enables_select_0=enables_select1=false:popcnt, no bmi2: 1600MB/s
popcnt, no bmi2: 2500MB/s
popcnt and bmi2: 2900MB/s
enables_select_0=enables_select1=true:no popcnt, no bmi2: 1100MB/s
popcnt, no bmi2: 1600MB/s
popcnt and bmi2: 1800MB/s