Repository navigation
CCFQueue: append that iterates over the shorter operand (+ bench) - #495
Merged
Merged
Conversation
Throughputs for "CCList.append", "List.append", "CCFun_vec.append", "Pvec.append", "Sek.concat", "CCFQueue.append", "CCFQueue.append (r-assoc)" each running 3 times for at least 2 CPU seconds:
CCList.append: 2.02 WALL ( 2.01 usr + 0.00 sys = 2.02 CPU, minor = 9993.60 MB, major = 0 B) @ 6881.88/s (n=13880)
2.01 WALL ( 2.01 usr + 0.00 sys = 2.01 CPU, minor = 9993.60 MB, major = 0 B) @ 6897.54/s (n=13880)
2.09 WALL ( 2.08 usr + 0.00 sys = 2.09 CPU, minor = 9993.60 MB, major = 0 B) @ 6650.14/s (n=13880)
List.append: 2.20 WALL ( 2.20 usr + 0.00 sys = 2.20 CPU, minor = 11758.32 MB, major = 0 B) @ 7418.49/s (n=16331)
2.23 WALL ( 2.22 usr + 0.00 sys = 2.23 CPU, minor = 11758.32 MB, major = 0 B) @ 7336.55/s (n=16331)
2.20 WALL ( 2.20 usr + 0.00 sys = 2.20 CPU, minor = 11758.32 MB, major = 0 B) @ 7430.92/s (n=16331)
CCFun_vec.append: 2.12 WALL ( 2.12 usr + 0.00 sys = 2.12 CPU, minor = 28591.61 MB, major = 0 B) @ 871.78/s (n=1851)
2.13 WALL ( 2.13 usr + 0.00 sys = 2.13 CPU, minor = 28591.61 MB, major = 0 B) @ 869.39/s (n=1851)
2.12 WALL ( 2.11 usr + 0.00 sys = 2.12 CPU, minor = 28591.61 MB, major = 0 B) @ 874.99/s (n=1851)
Pvec.append: 2.20 WALL ( 2.20 usr + 0.00 sys = 2.20 CPU, minor = 12912.84 MB, major = 0 B) @ 2209.16/s (n=4865)
2.24 WALL ( 2.24 usr + 0.00 sys = 2.24 CPU, minor = 12912.84 MB, major = 0 B) @ 2171.91/s (n=4865)
2.25 WALL ( 2.25 usr + 0.00 sys = 2.25 CPU, minor = 12912.84 MB, major = 0 B) @ 2161.75/s (n=4865)
Sek.concat: 2.18 WALL ( 2.17 usr + 0.00 sys = 2.18 CPU, minor = 17026.66 MB, major = 0 B) @ 1444267.83/s (n=3143771)
2.17 WALL ( 2.17 usr + 0.00 sys = 2.17 CPU, minor = 17026.66 MB, major = 0 B) @ 1448128.44/s (n=3143771)
2.16 WALL ( 2.16 usr + 0.00 sys = 2.16 CPU, minor = 17026.66 MB, major = 0 B) @ 1452894.19/s (n=3143771)
CCFQueue.append: 2.25 WALL ( 2.24 usr + 0.00 sys = 2.25 CPU, minor = 14796.36 MB, major = 0 B) @ 1420.12/s (n=3191)
2.25 WALL ( 2.24 usr + 0.00 sys = 2.25 CPU, minor = 14796.36 MB, major = 0 B) @ 1421.24/s (n=3191)
2.22 WALL ( 2.22 usr + 0.00 sys = 2.22 CPU, minor = 14796.36 MB, major = 0 B) @ 1435.99/s (n=3191)
CCFQueue.append (r-assoc): 2.21 WALL ( 2.21 usr + 0.00 sys = 2.21 CPU, minor = 14615.03 MB, major = 0 B) @ 949.97/s (n=2101)
2.22 WALL ( 2.22 usr + 0.00 sys = 2.22 CPU, minor = 14615.03 MB, major = 0 B) @ 944.86/s (n=2101)
2.30 WALL ( 2.30 usr + 0.00 sys = 2.30 CPU, minor = 14615.03 MB, major = 0 B) @ 913.02/s (n=2101)
to achieve `min(m, n)` runtime.
*** Run benchmarks for path "list.append.10000"
Throughputs for "CCList.append", "List.append", "CCFun_vec.append", "Pvec.append", "Sek.concat", "CCFQueue.append", "CCFQueue.append (r-assoc)" each running 3 times for at least 2 CPU seconds:
CCList.append: 2.17 WALL ( 2.16 usr + 0.00 sys = 2.17 CPU, minor = 10821.60 MB, major = 0 B) @ 6939.78/s (n=15030)
2.20 WALL ( 2.19 usr + 0.00 sys = 2.20 CPU, minor = 10821.60 MB, major = 0 B) @ 6840.74/s (n=15030)
2.18 WALL ( 2.18 usr + 0.00 sys = 2.18 CPU, minor = 10821.60 MB, major = 0 B) @ 6899.68/s (n=15030)
List.append: 2.24 WALL ( 2.24 usr + 0.00 sys = 2.24 CPU, minor = 11154.96 MB, major = 0 B) @ 6921.46/s (n=15493)
2.24 WALL ( 2.24 usr + 0.00 sys = 2.24 CPU, minor = 11154.96 MB, major = 0 B) @ 6911.63/s (n=15493)
2.22 WALL ( 2.21 usr + 0.00 sys = 2.22 CPU, minor = 11154.96 MB, major = 0 B) @ 6990.08/s (n=15493)
CCFun_vec.append: 2.07 WALL ( 2.06 usr + 0.00 sys = 2.07 CPU, minor = 26367.31 MB, major = 0 B) @ 826.18/s (n=1707)
2.05 WALL ( 2.05 usr + 0.00 sys = 2.05 CPU, minor = 26367.31 MB, major = 0 B) @ 831.38/s (n=1707)
2.06 WALL ( 2.05 usr + 0.00 sys = 2.06 CPU, minor = 26367.31 MB, major = 0 B) @ 830.34/s (n=1707)
Pvec.append: 2.06 WALL ( 2.06 usr + 0.00 sys = 2.06 CPU, minor = 11609.61 MB, major = 0 B) @ 2123.31/s (n=4374)
2.01 WALL ( 2.01 usr + 0.00 sys = 2.01 CPU, minor = 11609.61 MB, major = 0 B) @ 2176.21/s (n=4374)
2.05 WALL ( 2.05 usr + 0.00 sys = 2.05 CPU, minor = 11609.61 MB, major = 0 B) @ 2134.59/s (n=4374)
Sek.concat: 2.02 WALL ( 2.01 usr + 0.00 sys = 2.02 CPU, minor = 15610.98 MB, major = 0 B) @ 1429794.82/s (n=2882382)
2.17 WALL ( 2.17 usr + 0.00 sys = 2.17 CPU, minor = 17030.75 MB, major = 0 B) @ 1447474.09/s (n=3144526)
2.00 WALL ( 2.00 usr + 0.00 sys = 2.00 CPU, minor = 15610.98 MB, major = 0 B) @ 1440124.59/s (n=2882382)
CCFQueue.append: 2.07 WALL ( 2.07 usr + 0.00 sys = 2.07 CPU, minor = 13725.24 MB, major = 0 B) @ 1427.06/s (n=2960)
2.09 WALL ( 2.08 usr + 0.00 sys = 2.08 CPU, minor = 13725.24 MB, major = 0 B) @ 1419.77/s (n=2960)
2.07 WALL ( 2.06 usr + 0.00 sys = 2.06 CPU, minor = 13725.24 MB, major = 0 B) @ 1433.62/s (n=2960)
CCFQueue.append (r-assoc): 2.03 WALL ( 2.03 usr + 0.00 sys = 2.03 CPU, minor = 13524.57 MB, major = 0 B) @ 1434.53/s (n=2917)
2.06 WALL ( 2.06 usr + 0.00 sys = 2.06 CPU, minor = 13524.57 MB, major = 0 B) @ 1416.82/s (n=2917)
2.06 WALL ( 2.06 usr + 0.00 sys = 2.06 CPU, minor = 13524.57 MB, major = 0 B) @ 1417.23/s (n=2917)
*** Run benchmarks for path "list.append.100000"
Throughputs for "CCList.append", "List.append", "CCFun_vec.append", "Pvec.append", "Sek.concat", "CCFQueue.append", "CCFQueue.append (r-assoc)" each running 3 times for at least 4 CPU seconds:
CCList.append: 4.28 WALL ( 4.27 usr + 0.01 sys = 4.28 CPU, minor = 5961.60 MB, major = 0 B) @ 193.64/s (n=828)
4.38 WALL ( 4.37 usr + 0.01 sys = 4.38 CPU, minor = 5961.60 MB, major = 0 B) @ 189.06/s (n=828)
4.39 WALL ( 4.37 usr + 0.02 sys = 4.39 CPU, minor = 5961.60 MB, major = 0 B) @ 188.82/s (n=828)
List.append: 4.27 WALL ( 4.26 usr + 0.01 sys = 4.27 CPU, minor = 5846.40 MB, major = 0 B) @ 190.31/s (n=812)
4.25 WALL ( 4.23 usr + 0.01 sys = 4.24 CPU, minor = 5846.40 MB, major = 0 B) @ 191.29/s (n=812)
4.25 WALL ( 4.24 usr + 0.01 sys = 4.25 CPU, minor = 5846.40 MB, major = 0 B) @ 191.05/s (n=812)
CCFun_vec.append: 4.38 WALL ( 4.37 usr + 0.01 sys = 4.38 CPU, minor = 61842.82 MB, major = 0 B) @ 64.60/s (n=283)
4.43 WALL ( 4.42 usr + 0.01 sys = 4.43 CPU, minor = 61842.82 MB, major = 0 B) @ 63.86/s (n=283)
4.41 WALL ( 4.40 usr + 0.01 sys = 4.41 CPU, minor = 61842.82 MB, major = 0 B) @ 64.11/s (n=283)
Pvec.append: 4.22 WALL ( 4.21 usr + 0.01 sys = 4.22 CPU, minor = 22446.20 MB, major = 0 B) @ 193.31/s (n=816)
4.26 WALL ( 4.25 usr + 0.01 sys = 4.26 CPU, minor = 22446.20 MB, major = 0 B) @ 191.71/s (n=816)
4.27 WALL ( 4.27 usr + 0.01 sys = 4.27 CPU, minor = 22446.20 MB, major = 0 B) @ 190.89/s (n=816)
Sek.concat: 4.18 WALL ( 4.17 usr + 0.00 sys = 4.18 CPU, minor = 32803.53 MB, major = 0 B) @ 973097.90/s (n=4063867)
4.23 WALL ( 4.22 usr + 0.00 sys = 4.23 CPU, minor = 32803.53 MB, major = 0 B) @ 961667.63/s (n=4063867)
4.20 WALL ( 4.20 usr + 0.01 sys = 4.20 CPU, minor = 32803.53 MB, major = 0 B) @ 967258.97/s (n=4063867)
CCFQueue.append: 4.50 WALL ( 4.50 usr + 0.01 sys = 4.50 CPU, minor = 27234.49 MB, major = 0 B) @ 130.43/s (n=587)
4.48 WALL ( 4.47 usr + 0.01 sys = 4.48 CPU, minor = 27234.49 MB, major = 0 B) @ 130.98/s (n=587)
4.42 WALL ( 4.40 usr + 0.01 sys = 4.42 CPU, minor = 27234.49 MB, major = 0 B) @ 132.92/s (n=587)
CCFQueue.append (r-assoc): 4.42 WALL ( 4.42 usr + 0.01 sys = 4.42 CPU, minor = 27559.08 MB, major = 0 B) @ 134.31/s (n=594)
4.44 WALL ( 4.43 usr + 0.01 sys = 4.44 CPU, minor = 27559.08 MB, major = 0 B) @ 133.83/s (n=594)
4.43 WALL ( 4.42 usr + 0.01 sys = 4.43 CPU, minor = 27559.08 MB, major = 0 B) @ 134.14/s (n=594)
Owner
|
thank you, that looks good! |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Fixes #493
I have also added two benchs for CCFQueue.append, to show the difference in left vs right associative running times. The before/after results are in the commit messages, it's about a 50% speed up in the 10000 length.
I also added two functions
to_iter_revandadd_iter_front_rev. These aren't exposed through the mli at the moment. Let me know if you think they should be, or whether they should be inlined or_prefixed.