Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Ruby 2.4 のハッシュテーブル高速化を理解する
Search
Nao Minami
April 20, 2017
Programming
3
6.8k
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/
Nao Minami
April 20, 2017
Tweet
Share
More Decks by Nao Minami
See All by Nao Minami
Real World Migration from HTTP to gRPC #CNDT2020
south37
3
5.8k
Real World Migration from HTTP to gRPC in Ruby #grpcconf
south37
2
4.4k
Getting Things Done をベースにした仕事の進め方 / How to Work with Getting Things Done
south37
8
7.9k
Web API に秩序を与える Protocol Buffers / Protocol Buffers for Web API #builderscon
south37
18
16k
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
south37
1
3.9k
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
south37
10
15k
gcpc: Google Cloud Pub/Sub Client for Ruby #tqrk13
south37
1
800
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
south37
8
40k
学びを得るための新卒 ISUCON / New Grad ISUCON for Learning
south37
4
43k
Other Decks in Programming
See All in Programming
リアルタイムレイトレーシング + ニューラルレンダリング簡単紹介 / Real-Time Ray Tracing & Neural Rendering: A Quick Introduction (2025)
shocker_0x15
1
290
AWSで雰囲気でつくる! VRChatの写真変換ピタゴラスイッチ
anatofuz
0
140
Code smarter, not harder - How AI Coding Tools Boost Your Productivity | Webinar 2025
danielsogl
0
120
タイムゾーンの奥地は思ったよりも闇深いかもしれない
suguruooki
1
570
custom_lintで始めるチームルール管理
akaboshinit
0
200
The Weight of Data: Rethinking Cloud-Native Systems for the Age of AI
hollycummins
0
270
5年間継続して開発した自作OSSの記録
bebeji_nappa
0
170
Building a macOS screen saver with Kotlin (Android Makers 2025)
zsmb
1
140
Bedrock×MCPで社内ブログ執筆文化を育てたい!
har1101
6
900
Building Scalable Mobile Projects: Fast Builds, High Reusability and Clear Ownership
cyrilmottier
2
260
Devinのメモリ活用の学びを自社サービスにどう組み込むか?
itarutomy
0
2.1k
リアクティブシステムの変遷から理解するalien-signals / Learning alien-signals from the evolution of reactive systems
yamanoku
3
1.2k
Featured
See All Featured
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
45
9.5k
Faster Mobile Websites
deanohume
306
31k
Optimizing for Happiness
mojombo
377
70k
Rebuilding a faster, lazier Slack
samanthasiow
80
8.9k
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
31
1.1k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
29
1.6k
GraphQLとの向き合い方2022年版
quramy
46
14k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
49k
How STYLIGHT went responsive
nonsquared
99
5.5k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
45
7.2k
Art, The Web, and Tiny UX
lynnandtonic
298
20k
Why You Should Never Use an ORM
jnunemaker
PRO
55
9.3k
Transcript
3VCZͷϋογϡςʔϒϧ ߴԽΛཧղ͢Δ /BP.JOBNJ !TPVUI
ࣗݾհ
/BP.JOBNJ!TPVUI !NJOBNJP 4PGUXBSFFOHJOFFS !8BOUFEMZ *OD
w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴԽͱͦͷԸܙ w ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ͢༰
w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴԽͱͦͷԸܙ w ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ͢༰
3VCZY
w ·Ͱʹ3VCZഒ͘ͳΔCZ!NBU[ 3VCZͱύϑΥʔϚϯε w .VMUJ5ISFBE +*5 FUD w 3VCZY w
3VCZͰͷϋογϡςʔϒϧߴԽ /&8
w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴԽͱͦͷԸܙ w ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ͢༰
ϋογϡςʔϒϧߴԽͱͦͷԸܙ
ϋογϡςʔϒϧߴԽͱͦͷԸܙ
ϋογϡςʔϒϧߴԽͱͦͷԸܙ w 36#:#&/$) w ߴԽ w ϝϞϦ༻ྔݮ 1000000.times.map{|i| a={}; 8.times{|j|
a[j]=j}; a} IUUQTSVCZCFODIPSHSVCZSVCZSFMFBTFT SFTVMU@UZQFIBTI@TNBMM
ϋογϡςʔϒϧߴԽͱͦͷԸܙ w QJDP@IUUQ@QBSTFS w ߴԽ IUUQLB[FCVSPIBUFOBCMPHDPNFOUSZ ʜ3VCZ ੨ʜ3VCZQSF
ܶతʹ͘ͳͬͯΔ
w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴԽͱͦͷԸܙ w ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ͢༰
ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ϋογϡςʔϒϧͷিಥ࣌ͷڍಈΛมߋ $IBJOJOH 0QFO"ESFTTJOH w ϙΠϯτʮσʔλͷہॴੑͷ্ʯ
σʔλͷہॴੑͳͥॏཁ͔ w 1SPDFTTPSʹଟஈΩϟογϡ͕ଘࡏ w ΩϟογϡʹIJU͢ΔͱύϑΥʔϚϯε্ $16 L# .BJO.FNPSZ -$BDIF -$BDIF
-$BDIF .# (# L# # dOT IUUQTUBDLPWFSqPXDPNRVFTUJPOTBQQSPYJNBUFDPTUUPBDDFTTWBSJPVTDBDIFTBOENBJONFNPSZ dOT
σʔλͷہॴੑͳͥॏཁ͔ w 1SPDFTTPSʹଟஈΩϟογϡ͕ଘࡏ w σʔλͷہॴੑ͕ॏཁ $16 L# .BJO.FNPSZ -$BDIF -$BDIF
-$BDIF .# (# L# # dOT IUUQTUBDLPWFSqPXDPNRVFTUJPOTBQQSPYJNBUFDPTUUPBDDFTTWBSJPVTDBDIFTBOENBJONFNPSZ dOT
0QFO"ESFTTJOHͰσʔλͷہॴੑ্͕͕Δͷͳͥʁ w $IBJOJOH 0QFO"ESFTTJOHͭͷΞϧΰϦζϜΛൺֱ w ૬ҧϋογϡςʔϒϧͷিಥ࣌ͷڍಈ w $IBJOJOH࿈݁ϦετΛḷͬͯ୳ࡧ w 0QFO"ESFTTJOH"SSBZ্ͰJOEFYΛม͑ͯ୳ࡧ
w ڞ௨ w ϋογϡςʔϒϧLFZͷIBTIΛJOEFYʹར༻
$IBJOJOHͷΈ w UBCMF࿈݁ϦετͷϙΠϯλΛอ࣋ w ಉ͡IBTIΛ࣋ͭ߹ɺ࿈݁ϦετΛḷΔ IUUQXXXBMHPMJTUOFU%BUB@TUSVDUVSFT)BTI@UBCMF$IBJOJOH FOUSZ UBCMF LFZ
$IBJOJOHͷ w ࿈݁ϦετͷFOUSZɺNFNPSZ্ͰΕͯஔ w σʔλͷہॴੑ͕͍ FOUSZ UBCMF LFZ
3VCZҎલͷ$IBJOJOH w ํϦετʹͳ͓ͬͯΓɺσʔλͷہॴੑ͕͍ ͚ͩͰͳ͘QSFWϙΠϯλͳͲͷ͚ͩFOUSZͷσʔ λ͕Ͱ͔͍ͷ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT UBCMF FOUSZ FOUSZ FOUSZ
3VCZҎલͷ$IBJOJOH w ํϦετʹͳ͓ͬͯΓɺσʔλͷہॴੑ͕͍ ͚ͩͰͳ͘QSFWϙΠϯλͳͲͷ͚ͩFOUSZͷσʔ λ͕Ͱ͔͍ͷ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT UBCMF FOUSZ FOUSZ FOUSZ
͍
0QFO"ESFTTJOHͷΈ w UBCMFFOUSZͷQPJOUFSΛอ࣋ w ಉ͡IBTIΛ࣋ͭ߹ɺJOEFYΛͣΒ͢ FOUSZ UBCMF LFZ IUUQXXXBMHPMJTUOFU%BUB@TUSVDUVSFT)BTI@UBCMF0QFO@BEESFTTJOH
3VCZͷ0QFO"ESFTTJOH UBCMF FOUSJFT TUBSU CPVOE w UBCMF FOUSJFT͕྆ํ"SSBZʹͳ͓ͬͯΓσʔλͷہॴ ੑ͕ߴ͘ɺͭͭͷFOUSZͷσʔλαΠζখ͍͞ IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSETGBTUFSSVCZIBTIUBCMFT
3VCZͷ0QFO"ESFTTJOH UBCMF FOUSJFT TUBSU CPVOE w UBCMF FOUSJFT͕྆ํ"SSBZʹͳ͓ͬͯΓσʔλͷہॴ ੑ͕ߴ͘ɺͭͭͷFOUSZͷσʔλαΠζখ͍͞ ߴԽ
3VCZͰߋʹࡉ͔͘࠷దԽ w ͕݅গͳ͍࣌ͷ࠷దԽʢলϝϞϦɺߴԽʣ w ͕݅গͳ͍࣌BSSBZͭͰMJOFBSTFBSDI w ͕݅গͳ͍࣌CJU CJUͳͲͰJOEFYΛදݱ w 'VMMDZDMFMJOFBSDPOHSVFOUJBMHFOFSBUPSΛ
TFDPOEBSZIBTIͱͯ͠ར༻ w ߴ͔ͭIBTIͱͯ͠ͷੑೳྑ͍
w 3VCZͱύϑΥʔϚϯε w 3VCZͰͷϋογϡςʔϒϧߴԽͱͦͷԸܙ w ߴԽͲ͏࣮ݱ͞Εͨͷ͔ʁ w ·ͱΊ ࠓ͢༰
·ͱΊ w 3VCZϋογϡςʔϒϧ͕ܶతʹߴԽɻ࣮ࡍ ʹΞϓϦέʔγϣϯߴԽɻ w 3VCZʹͯ͠շదͳ3VCZϥΠϑΛૹΖ͏ʂ
ࢀߟϦϯΫ w 3VCZϦϦʔε w IUUQTXXXSVCZMBOHPSHKBOFXT SVCZSFMFBTFE w 'FBUVSF)BTIUBCMFTXJUIPQFOBESFTTJOH w IUUQTCVHTSVCZMBOHPSHJTTVFT
w 5PXBSET'BTUFS3VCZ)BTI5BCMFT w IUUQTEFWFMPQFSTSFEIBUDPNCMPHUPXBSET GBTUFSSVCZIBTIUBCMFT