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应用性能调优实践
Search
Li Zhe
October 12, 2015
Programming
630
10
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
Ruby应用性能调优实践
了解Ruby的性能调优关键,以及都有哪些工具可以利用
Li Zhe
October 12, 2015
Other Decks in Programming
See All in Programming
Hello, Hiroshima Geospatial Data! — Exploring DoboX with Python
ra0kley
0
120
Go 1.27 における memory allocation の高速化
andpad
0
350
[PyCon KR 2026] More Variants, More Diversity for AI Accelerators
achimnol
0
120
T3DD26: From RAGs to Riches
martinhelmich
0
120
Hono + Inertia + React で LP を構築した話
oukayuka
2
190
Discordを用いたラボオートメーション関連情報収集の自動化
noguhiro2002
0
470
異なる設計思想のフレームワークを経験して得た学び
amekuhideki
2
860
AI駆動開発にグラフDBを重ねてみた
satoshi256kbyte
1
400
AI時代に学ぶ 好きなルール 嫌いなルール Linter編
shorty5121
0
230
30年振りにコンパイラの定数整数除算を改善した
herumi
9
4.3k
PyConJP2026_wat_Python × Signal Processing: How to Draw Pictures with Sound Using Spectrogram Art
wat
0
590
Press start. Python's next generation.
willingc
PRO
3
270
Featured
See All Featured
XXLCSS - How to scale CSS and keep your sanity
sugarenia
249
1.3M
"I'm Feeling Lucky" - Building Great Search Experiences for Today's Users (#IAC19)
danielanewman
230
23k
Agile Actions for Facilitating Distributed Teams - ADO2019
mkilby
0
270
SERP Conf. Vienna - Web Accessibility: Optimizing for Inclusivity and SEO
sarafernandez
2
1.6k
Building the Perfect Custom Keyboard
takai
2
850
Designing Experiences People Love
moore
143
24k
Neural Spatial Audio Processing for Sound Field Analysis and Control
skoyamalab
0
420
コードの90%をAIが書く世界で何が待っているのか / What awaits us in a world where 90% of the code is written by AI
rkaga
63
45k
JAMstack: Web Apps at Ludicrous Speed - All Things Open 2022
reverentgeek
1
580
Leading Effective Engineering Teams in the AI Era
addyosmani
9
2.5k
Marketing Yourself as an Engineer | Alaka | Gurzu
gurzu
0
280
The SEO Collaboration Effect
kristinabergwall1
1
540
Transcript
Rubyଫአᚆ᧣սਫ᪢ ߽ - OneAPM 2015.10.10
ञฎՋԍҘ
ञ: ݰԧ੪ : ݰԧ੪ݺ
֦మݰᗤᶟҘ ඓԀஞހҘ ౯݄ԧᬟѺ
ڹݶԪ { ڹݶԪ }
ӳ۹Ոࣁଠ
ৼ֯Ո
۹Ղཛྷୗސ …
ํᅩᔲୟ …
౯ժฎଗՋԍጱ? • OneAPM • ᚆፊഴଘݣ • 2015ଙڡਠ౮Cᣟᩒ • 350Ո •
ᬪ200ՈฎᎸݎ • قࢵԆᥝITउ૱᮷ํړلݪېԪ॒
౯ฎ᧡Ҙ • ߽ • Java 2007 ~ 2010 • Ruby
2011 ~ Now • ፓڹ੪ᘳOneAPM - RubyAgent, Cloud Insight • github.com/markgeek • douban.com/people/markgeek/
౯ժᘱᘱRails ౌ ᬯկԪ
ᘏฎRack ౌ ᬯկԪ
ٌਫฎRuby ౌ ᬯկԪ
֦ᦊԅRubyౌހҘ ౌጱᤒሿฎՋԍҘ ፥ጱฎ᭜౮ጱހҘ ᦶ፳አૡٍ༄ၥᬦހҘ
30%զӤጱౌฎຝӤ᧣սጱ
– ӧᎣ᭲ฎ᧡᧔ጱ “ᑕଧާ᮷ฎࡅෛܵ෯ጱਹվ̶”
ӞӻRubyଫአ • 2000ଚݎ • क़᮱ߥଫᳵ300ms • ଫአߥଫᳵ < 150ms •
4ݣԆ / 8core,8G • nginx, goliath, grape, redis
None
None
፡፡RubyጱचᏐܲݥ
RubyVMጱݎܲݥ • 1.9 Koichi Sasada : YARV • 1.9 Narihiro
Nakamura ғLazySweep • 2.0 Narihiro Nakamura ғBitmap Marking GC • 2.1 Koichi Sasada ғGenerational Collector • 2.2 Koichi Sasada ғIncremental Generational Collector
http://tmm1.net/ruby21-rgengc
Rubyጱᬩᤈ᭛ଶӞፗࣁ܋
ӞԶच༷ஷ • Benchmark • Profiling • Tracing • Debugging
Benchmarkғ᪒ړ
benchmark, benchmark-ips
Profilingғᦕ୯ᬩᤈמ௳
None
RubyProf,StackProf
Tracingғᦕ୯ොဩ᧣አמ௳
rbtrace,allocation_tracer
Ӟӻੜጱbenchmark
Ruby Performance
1.98 / 1.14 = 1.7368…
None
RubyProf • flat - Prints a flat profile as text
(default). • dot - Prints a graph profile as a dot file • call_stack - prints a HTML visualization of the call tree • call_tree - format for KCacheGrind
None
None
None
None
StackProf • CPUཛྷୗ • ړᯈཛྷୗ • middlewareཛྷୗ
None
None
None
rbtrace • rbtrace -p <PID> —firehose • rbtrace -p <PID>
—slow=<N> • rbtrace -p <PID> --methods "Kernel#sleep" “Proc#call" • rbtrace -p <PID> --gc
Ꮯਧኒ᷀ԏݸ • fast-ruby • fasterer
Ӟӻਫᴬጱֺৼ
None
None
ݚक़ጱӞԶૡٍ
http://rbkit.codemancers.com/
https://github.com/peek/peek-rblineprof
https://github.com/MiniProfiler/rack-mini-profiler
ইຎ֦ବ੶உఽي᪁ Dtrace / SystemTap
ݳᭇጱૡٍ؉ݳᭇጱૡ֢
Rubyਖ਼տᬩᤈጱๅள
– Yukihiro 'Matz' Matsumoto “Super Dry Ruby”
https://www.youtube.com/watch?v=bqWBB8-iEac&list=PLE7tQUdRKcyZdMYKatu_nics8Pkjxl3-F
ই֜೮ᖅጱፊഴଫአ ጱᚆᳯ᷌
Application Performance Monitor or Application Performance Management
None
None
ӧՐՐฎᬯԶ
None
None
რᩒრ • https://github.com/JuanitoFatas/fast-ruby • https://github.com/DamirSvrtan/fasterer • https://github.com/schneems/derailed_benchmarks • https://github.com/change/method_profiler •
https://github.com/schneems/heapy • https://github.com/peek/peek-rblineprof • https://github.com/tmm1/rblineprof
݇ᘍᩒා1 • https://engineering.heroku.com/blogs/2015-02-04-incremental-gc/ • http://tmm1.net/ruby21-rgengc/ • https://www.youtube.com/watch?v=nRehkd22j78&spfreload=10 • https://avsej.net/2012/systemtap-and-ruby-20/ •
https://www.shopify.com/technology/16777156-tuning-rubys-global-method-cache • https://www.youtube.com/watch?v=pZ_BcEcFGj0 • http://www.atdot.net/~ko1/activities/2015_railsconf.pdf • http://samsaffron.com/archive/2013/11/22/demystifying-the-ruby-gc
݇ᘍᩒා2 • http://www.theirishpenguin.com/2009/10/29/understanding-how-ruby-stores- objects-in-memory-the-ruby-heap.html • http://samsaffron.com/archive/2013/11/22/demystifying-the-ruby-gc • http://thorstenball.com/blog/2014/03/12/watching-understanding-ruby-2.1- garbage-collector/
https://github.com/mattbaker/ruby-heap-viz
ᨀᨀ Q&A