Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Ruby 2.4 のハッシュテーブル高速化を理解する

2b4f9f20b554d4c77c46fe6a9930d6fe?s=47 Nao Minami
April 20, 2017
Programming
3
6.1k

Ruby 2.4 のハッシュテーブル高速化を理解する

第2回 meguro.rb LT で Ruby 2.4 のハッシュテーブル実装について話しました
https://megurorb.connpass.com/event/55107/

2b4f9f20b554d4c77c46fe6a9930d6fe?s=128

Nao Minami

April 20, 2017
Tweet

More Decks by Nao Minami

See All by Nao Minami
Real World Migration from HTTP to gRPC #CNDT2020
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
3
4.3k
Real World Migration from HTTP to gRPC in Ruby #grpcconf
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
2
3.5k
Getting Things Done をベースにした仕事の進め方 / How to Work with Getting Things Done
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
5
4.9k
Web API に秩序を与える Protocol Buffers / Protocol Buffers for Web API #builderscon
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
18
13k
puma v4 では SIGTERM での worker process ゾンビ化に気をつけよう / Be aware of zombie processes in puma v4
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
0
2.7k
理想的なマイクロサービスアーキテクチャを目指す継続的改善 / Re-architecturing of Microservices #CNDT2019
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
10
12k
gcpc: Google Cloud Pub/Sub Client for Ruby #tqrk13
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
1
490
実行計画から学ぶ PostgreSQL の内部動作とクエリ最適化 / Learn PostgreSQL from Explain
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
7
14k
学びを得るための新卒 ISUCON / New Grad ISUCON for Learning
2b4f9f20b554d4c77c46fe6a9930d6fe?s=48 south37
3
17k

Other Decks in Programming

See All in Programming
#JJUG_CCC 「サポート」は製品開発？　- JDBCライブラリ屋さんが実践する攻めのテクニカルサポートとJavaエンジニアのキャリアについて -
244963b5eab1dd775692490daea81a7f?s=48 cdataj
0
420
設計ナイト2022 トランザクションスクリプト
6521324f6ab5266cdcde1dd05945a27b?s=48 shinpeim
11
2k
engineer
E6706306bf41c5a5675e7296ecdb4103?s=48 spacemarket
0
810
Java初心者が知っておくべきプログラミングのこと - JJUG CCC 2022 Spring
9e840766611f942c7c0a9ad6987a5d78?s=48 kishida
5
540
インターン生・新卒向け、学校でもっと教えてほしいITエンジニア基本スキル
E72f241475dfe632c6b1666debc16e8a?s=48 nearme_tech
0
120
The strategies behind ddd – AdeoDevSummit 2022
Beb422437c1dfb5366f197919e41ac50?s=48 lilobase
PRO
4
240
RFC 9111: HTTP Caching
1ff811939fd0923df8321ec6d8bf9d4b?s=48 jxck
0
160
trocco® の品質を守る、とても普通な取り組み
E7d321bef653bcb8fce8032609321398?s=48 kekekenta
0
350
実践エクストリームプログラミング / Extreme Programming in Practice
00cbada6669966805d91061d5a234d3b?s=48 enk
1
500
"What's new in Swift"の要約 / swift_5_7_summary
42a6a049ac8f5265f31858a9509217fb?s=48 uhooi
1
300
パターンマッチングを学んで新しいJavaの世界へ！Java 18までの目玉機能をおさらいしよう / Java 18 pattern matching
Fb025419ed38f98df595eb2eafc859f4?s=48 ihcomega56
3
390
Oracle REST Data Service: APEX Office Hours
0ed10d1154c696886ca483fe827cb299?s=48 thatjeffsmith
0
720

Featured

See All Featured
VelocityConf: Rendering Performance Case Studies
96270e4c3e5e9806cf7245475c00b275?s=48 addyosmani
316
22k
Code Review Best Practice
3d6ace9554821d552146413bcdf874f6?s=48 trishagee
43
9.2k
Dealing with People You Can't Stand - Big Design 2015
44b54d2ffcb7857004071cc6795dde11?s=48 cassininazir
351
21k
Code Reviewing Like a Champion
C7393b7ba7ec9c8890dd77d209fbb3c9?s=48 maltzj
506
37k
The Pragmatic Product Professional
1177e050db6bafe62885362edf6e3537?s=48 lauravandoore
19
3k
Rails Girls Zürich Keynote
24fc194843a71f10949be18d5a692682?s=48 gr2m
86
12k
Web development in the modern age
465724d73fe3a92c0879fdfb43a3a6f3?s=48 philhawksworth
197
9.3k
Helping Users Find Their Own Way: Creating Modern Search Experiences
3d4519fd34b76fb265fc0237f3792bd4?s=48 danielanewman
7
1.1k
Music & Morning Musume
A60068bce2e73de3a37ca9d2dbe36092?s=48 bryan
35
4.2k
Agile that works and the tools we love
462dad0dd24c568a32dcdb0ae895ae1b?s=48 rasmusluckow
319
19k
Why Our Code Smells
20bfe76b3d6105641f879fe45cfc9272?s=48 bkeepers
PRO
324
55k
The Cult of Friendly URLs
Ded01cdab114abe4ec13511e4c25b9bb?s=48 andyhume
68
4.8k

Transcript

  1. 3VCZͷϋογϡςʔϒϧ ߴ଎ԽΛཧղ͢Δ  /BP.JOBNJ !TPVUI

  2. ࣗݾ঺հ

  3. /BP.JOBNJ!TPVUI !NJOBNJP 4PGUXBSFFOHJOFFS !8BOUFEMZ *OD

  4. w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ೔࿩͢಺༰

  5. w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ೔࿩͢಺༰

  6. 3VCZY

  7. w ೥·Ͱʹ3VCZ͸ഒ଎͘ͳΔCZ!NBU[ 3VCZͱύϑΥʔϚϯε w .VMUJ5ISFBE +*5 FUD w 3VCZY w

    3VCZͰͷϋογϡςʔϒϧߴ଎Խ /&8

  8. w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ೔࿩͢಺༰

  9. ϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ

  10. ϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ

  11. ϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w 36#:#&/$) w ߴ଎Խ w ϝϞϦ࢖༻ྔ࡟ݮ 1000000.times.map{|i| a={}; 8.times{|j|

    a[j]=j}; a} IUUQTSVCZCFODIPSHSVCZSVCZSFMFBTFT SFTVMU@UZQFIBTI@TNBMM

  12. ϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w QJDP@IUUQ@QBSTFS w ߴ଎Խ IUUQLB[FCVSPIBUFOBCMPHDPNFOUSZ ྘ʜ3VCZ ੨ʜ3VCZQSF

  13. ܶతʹ଎͘ͳͬͯΔ

  14. w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ೔࿩͢಺༰

  15. ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ϋογϡςʔϒϧͷিಥ࣌ͷڍಈΛมߋ $IBJOJOH 0QFO"ESFTTJOH w ϙΠϯτ͸ʮσʔλͷہॴੑͷ޲্ʯ

  16. σʔλͷہॴੑ͸ͳͥॏཁ͔ w 1SPDFTTPSʹ͸ଟஈΩϟογϡ͕ଘࡏ w ΩϟογϡʹIJU͢ΔͱύϑΥʔϚϯε޲্ $16 L# .BJO.FNPSZ -$BDIF -$BDIF

    -$BDIF .# (# L# # dOT IUUQTUBDLPWFSqPXDPNRVFTUJPOTBQQSPYJNBUFDPTUUPBDDFTTWBSJPVTDBDIFTBOENBJONFNPSZ dOT

  17. σʔλͷہॴੑ͸ͳͥॏཁ͔ w 1SPDFTTPSʹ͸ଟஈΩϟογϡ͕ଘࡏ w σʔλͷہॴੑ͕ॏཁ $16 L# .BJO.FNPSZ -$BDIF -$BDIF

    -$BDIF .# (# L# # dOT IUUQTUBDLPWFSqPXDPNRVFTUJPOTBQQSPYJNBUFDPTUUPBDDFTTWBSJPVTDBDIFTBOENBJONFNPSZ dOT

  18. 0QFO"ESFTTJOHͰσʔλͷہॴੑ্͕͕Δͷ͸ͳͥʁ w $IBJOJOH 0QFO"ESFTTJOHͭͷΞϧΰϦζϜΛൺֱ w ૬ҧ఺ϋογϡςʔϒϧͷিಥ࣌ͷڍಈ w $IBJOJOH࿈݁ϦετΛḷͬͯ୳ࡧ w 0QFO"ESFTTJOH"SSBZ্ͰJOEFYΛม͑ͯ୳ࡧ

    w ڞ௨఺ w ϋογϡςʔϒϧ͸LFZͷIBTI஋ΛJOEFYʹར༻

  19. $IBJOJOHͷ࢓૊Έ w UBCMF͸࿈݁Ϧετ΁ͷϙΠϯλΛอ࣋ w ಉ͡IBTIΛ࣋ͭ৔߹͸ɺ࿈݁ϦετΛḷΔ IUUQXXXBMHPMJTUOFU%BUB@TUSVDUVSFT)BTI@UBCMF$IBJOJOH FOUSZ UBCMF LFZ

  20. $IBJOJOHͷ໰୊఺ w ࿈݁ϦετͷFOUSZ͸ɺNFNPSZ্Ͱ཭Εͯ഑ஔ w σʔλͷہॴੑ͕௿͍ FOUSZ UBCMF LFZ

  21. 3VCZҎલͷ$IBJOJOH w ૒ํ޲Ϧετʹͳ͓ͬͯΓɺσʔλͷہॴੑ͕௿͍ ͚ͩͰͳ͘QSFWϙΠϯλͳͲͷ෼͚ͩFOUSZͷσʔ λ͕Ͱ͔͍ͷ΋໰୊ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT UBCMF FOUSZ FOUSZ FOUSZ

  22. 3VCZҎલͷ$IBJOJOH w ૒ํ޲Ϧετʹͳ͓ͬͯΓɺσʔλͷہॴੑ͕௿͍ ͚ͩͰͳ͘QSFWϙΠϯλͳͲͷ෼͚ͩFOUSZͷσʔ λ͕Ͱ͔͍ͷ΋໰୊ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT UBCMF FOUSZ FOUSZ FOUSZ

    ஗͍

  23. 0QFO"ESFTTJOHͷ࢓૊Έ w UBCMF͸FOUSZ΁ͷQPJOUFSΛ௚઀อ࣋ w ಉ͡IBTIΛ࣋ͭ৔߹͸ɺJOEFYΛͣΒ͢ FOUSZ UBCMF LFZ IUUQXXXBMHPMJTUOFU%BUB@TUSVDUVSFT)BTI@UBCMF0QFO@BEESFTTJOH

  24. 3VCZͷ0QFO"ESFTTJOH UBCMF FOUSJFT TUBSU CPVOE w UBCMF FOUSJFT͕྆ํ"SSBZʹͳ͓ͬͯΓσʔλͷہॴ ੑ͕ߴ͘ɺͭͭͷFOUSZͷσʔλαΠζ΋খ͍͞ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT

  25. 3VCZͷ0QFO"ESFTTJOH UBCMF FOUSJFT TUBSU CPVOE w UBCMF FOUSJFT͕྆ํ"SSBZʹͳ͓ͬͯΓσʔλͷہॴ ੑ͕ߴ͘ɺͭͭͷFOUSZͷσʔλαΠζ΋খ͍͞ ߴ଎Խ

  26. 3VCZͰ͸ߋʹࡉ͔͘࠷దԽ w ݅਺͕গͳ͍࣌ͷ࠷దԽʢলϝϞϦɺߴ଎Խʣ w ݅਺͕গͳ͍࣌͸BSSBZͭͰMJOFBSTFBSDI w ݅਺͕গͳ͍࣌͸CJU CJUͳͲͰJOEFYΛදݱ w 'VMMDZDMFMJOFBSDPOHSVFOUJBMHFOFSBUPSΛ

    TFDPOEBSZIBTIͱͯ͠ར༻ w ߴ଎͔ͭIBTIͱͯ͠ͷੑೳ΋ྑ͍

  27. w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴ଎ԽͱͦͷԸܙ w ߴ଎Խ͸Ͳ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ೔࿩͢಺༰

  28. ·ͱΊ w 3VCZ͸ϋογϡςʔϒϧ͕ܶతʹߴ଎Խɻ࣮ࡍ ʹΞϓϦέʔγϣϯ΋ߴ଎Խɻ w 3VCZʹͯ͠շదͳ3VCZϥΠϑΛૹΖ͏ʂ

  29. ࢀߟϦϯΫ w 3VCZϦϦʔε w IUUQTXXXSVCZMBOHPSHKBOFXT SVCZSFMFBTFE w 'FBUVSF)BTIUBCMFTXJUIPQFOBESFTTJOH w IUUQTCVHTSVCZMBOHPSHJTTVFT

    w 5PXBSET'BTUFS3VCZ)BTI5BCMFT w IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSET GBTUFSSVCZIBTIUBCMFT