Go 言語ã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«æ©Ÿèƒ½ã¨ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚³ãƒã‚¯ã‚·ãƒ§ãƒ³ã«ã¾ã¤ã‚るトラブルシューティング
ãƒ¤ãƒ•ãƒ¼æ ªå¼ä¼šç¤¾ã¯ã€2023å¹´10月1æ—¥ã«LINEãƒ¤ãƒ•ãƒ¼æ ªå¼ä¼šç¤¾ã«ãªã‚Šã¾ã—ãŸã€‚LINEãƒ¤ãƒ•ãƒ¼æ ªå¼ä¼šç¤¾ã®æ–°ã—ã„ブãƒã‚°ã¯ã“ã¡ã‚‰ã§ã™ã€‚LINEヤフー Tech Blog インフラエンジニア見習ã„ã®æ£®æœ¬ã§ã™ã€‚ã“ã‚Œã¾ã§æ•°å¹´ã»ã©ã‚µãƒ¼ãƒãƒ¼ã‚µã‚¤ãƒ‰ã‚¨ãƒ³ã‚¸ãƒ‹ã‚¢ã¨ã—ã¦é–‹ç™ºã°ã‹ã‚Šã‚’ã—ã¦ããŸäººãŒæœ€è¿‘インフラエンジニアã«ãªã‚Šã¾ã—ãŸã€‚ 3月末ã«é–‹å‚¬ã•ã‚ŒãŸ Go Conference 2017 Spring ã§é–‹ç™ºãƒãƒ¼ãƒ ã®å¾Œè—¤ã‹ã‚‰å¼Šç¤¾ã§é–‹ç™ºãƒ»é‹ç”¨ã—ã¦ã„ã‚‹ AWS S3 互æ›ã®åˆ†æ•£ã‚ªãƒ–ジェクトストレージ Dragon ã«ã¤ã„ã¦ã®ç™ºè¡¨ãŒã‚ã‚Šã¾ã—ãŸã€‚ Goã§ãƒ¤ãƒ•ãƒ¼ã®åˆ†æ•£ã‚ªãƒ–ジェクトストレージを作ã£ãŸè©± ç§ã¯ Dragon ã®é‹ç”¨ã‚’æ‹…ã†ãƒãƒ¼ãƒ ã«æ‰€å±žã—ã¦ãŠã‚Šã€æœ¬ç¨¿ã§ã¯ãã®æ¥å‹™ã®ä¸ã§ç™ºç”Ÿã—ãŸãƒˆãƒ©ãƒ–ルシューティングã«ã¤ã„ã¦ç´¹ä»‹ã—ãŸã„ã¨æ€ã„ã¾ã™ã€‚ 分散オブジェクトストレージ Dragon ã§ç™ºç”Ÿã—ã¦ã„ãŸèª²é¡Œ Dragon 㧠gorout
Profiling and optimizing Go web applications - Artem Krylysov
Go has a powerful built-in profiler that supports CPU, memory, goroutine and block (contention) profiling. Enabling the profiler # Go provides a low-level profiling API runtime/pprof, but if you are developing a long-running service, it's more convenient to work with a high-level net/http/pprof package. All you need to enable the profiler is to import net/http/pprof and it will automatically regis
Daily code optimization using benchmarks and profiling in Golang - Gophercon India 2016 talk
Need some help with analyzing whether algorithm A performs faster than B? Or can I know whether method A performs better than method B? Or is approach A to send JSON response to the client performs better than approach B? How to find answers for everyday performance related questions? These questions can be answered easily if you are happened to program in Golang. Golang comes bundled with facilit
Profiling Python in Production - Nylas
We recently reduced CPU usage across our fleet by 80%. One key technique that made this possible was a lightweight profiling strategy that we could run in production. This post is about the ways we approached instrumentation, the tradeoffs involved, and some tools you can use to optimize your own apps (including code!). Background Nylas is a developer platform that provides APIs to integrate with
Golangã§Flame Graphã‚’æã
アプリケーションã®ãƒ‘フォーマンスå•é¡Œã®è§£æ±ºã‚„ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã§å¤§åˆ‡ãªã®ã¯å•é¡Œã®ã‚³ã‚¢ã‚„ボトルãƒãƒƒã‚¯ã«æœ€çŸãƒ‘スã§åˆ°é”ã™ã‚‹ã“ã¨ã§ã‚る． 基本的ãªãƒ‘フォーマンス分æžã®å…¥ã‚Šå£ã¯ã‚¢ãƒ—リケーションã®ã‚¹ãƒ¬ãƒƒãƒ‰ãŒon-CPUã§æ™‚間を消費ã—ã¦ã„ã‚‹ã‹off-CPUã§æ™‚間を消費ã—ã¦ã„ã‚‹ã‹ã‚’ç†è§£ã™ã‚‹ã¨ã“ã‚ã‹ã‚‰å§‹ã¾ã‚‹ï¼Žon-CPUã®å ´åˆã¯ãã‚ŒãŒuserモードã‹kernelモードã‹ã‚’特定ã—，ã•ã‚‰ã«CPUプãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ã«ã‚ˆã£ã¦ã©ã®code pathãŒCPUを消費ã—ã¦ã„ã‚‹ã®ã‹ã®åˆ†æžã«å‘ã‹ã†ï¼Žoff-CPUã®å ´åˆã¯I/Oã‚„Lock，pagingã¨ã„ã£ãŸå•é¡Œã®åˆ†æžã«å‘ã‹ã†ï¼Ž Flame Graphã¯on-CPUã§ã®ãƒ‘フォーマンスã®å•é¡ŒãŒç™ºè¦šã—ãŸæ™‚ã«è¡Œã†CPUプãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ã‚’助ã‘る．ã©ã®code pathãŒãƒœãƒˆãƒ«ãƒãƒƒã‚¯ã«ãªã£ã¦ã„ã‚‹ã®ã‹ã‚’1ã¤ã®ã‚°ãƒ©ãƒ•ä¸Šã§ç†è§£ã§ãる．本記事ã§ã¯Flame Graphã¨ã¯ä½•ã‹? ãªãœå¿…è¦ãªã®ã‹? を解
YAPC: Go Debugging, Profiling, and Optimization
ã“ã®ãƒ–ラウザ ãƒãƒ¼ã‚¸ãƒ§ãƒ³ã®ã‚µãƒãƒ¼ãƒˆã¯çµ‚了ã—ã¾ã—ãŸã€‚サãƒãƒ¼ãƒˆã•ã‚Œã¦ã„るブラウザã«ã‚¢ãƒƒãƒ—グレードã—ã¦ãã ã•ã„。
Profiling Ansible Tasks
Introduction The source on GitHub. After spending far too much time wondering which of the tasks in my long list of Ansible tasks were slow, I decided to do something about it. In the pursuit of automating everything, I've constructed a large collection of playbooks, complete with their own dependencies, to do just about anything that needs to be done for my web service stacks. Where I used to gro
デコレータã§æ¥½ã«cProfileã™ã‚‹ - Qiita
ãã“ãã“ã®è¦æ¨¡ã®ãƒ©ã‚¤ãƒ–ラリ書ã„ã¦ã‚‹ã¨ã€ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ã™ã‚‹ã¨ã毎回コマンドラインã‹ã‚‰ã‚¹ã‚¯ãƒªãƒ—トを実行ã—ãŸã‚Šunittestã®setUpã¨tearDown書ãã®ãŒé¢å€’ã ã£ãŸã‚Šã™ã‚‹ã®ã§ã€ãƒ‡ã‚³ãƒ¬ãƒ¼ã‚¿ã‚’作ã£ã¦ãŠãã¾ã™ã€‚ import cProfile import pstats def profile(func): def _f(*args, **kwargs): pr = cProfile.Profile() pr.enable() print("\n<<<---") res = func(*args, **kwargs) p = pstats.Stats(pr) p.strip_dirs().sort_stats('cumtime').print_stats(20) print("\n--->>>") return res return _f
Enhanced Intel SpeedStep® Technology and Demand-Based Switching on Linux* - Intel® Software Network
Using Intel.com Search You can easily search the entire Intel.com site in several ways. Brand Name: Core i9 Document Number: 123456 Code Name: Emerald Rapids Special Operators: “Ice Lakeâ€, Ice AND Lake, Ice OR Lake, Ice* Quick Links You can also try the quick links below to see results for most popular searches. Product Information Support Drivers & Software
Profiling Go Programs - The Go Programming Language
Russ Cox, July 2011; updated by Shenghou Ma, May 2013 24 June 2011 At Scala Days 2011, Robert Hundt presented a paper titled Loop Recognition in C++/Java/Go/Scala. The paper implemented a specific loop finding algorithm, such as you might use in a flow analysis pass of a compiler, in C++, Go, Java, Scala, and then used those programs to draw conclusions about typical performance concerns in these
Go Conference 2014 Autumnã€ãã®ä»–ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã®ã¾ã¨ã‚ | gihyo.jp
å‰å›žã®è¨˜äº‹ã§ã¯ã€3ã¤ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã«ã¤ã„ã¦ãƒ¬ãƒãƒ¼ãƒˆã‚’記ã—ã¾ã—ãŸã€‚ã“ã®è¨˜äº‹ã§ã¯ã€æ®‹ã‚Šã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã«ã¤ã„ã¦ã¾ã¨ã‚ãŸã„ã¨æ€ã„ã¾ã™ã€‚ Why my Go program is slow?（@methaneæ°ï¼‰ ã“ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã§ã¯ã€@methaneæ°ã‹ã‚‰pprofã¨ã„ã†CPUプãƒãƒ•ã‚¡ã‚¤ãƒ©ã¨Go言語ã«ãŠã‘るパフォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã«ã¤ã„ã¦ã®è©±ãŒã•ã‚Œã¾ã—ãŸã€‚ãªãŠã€ã“ã®ã‚»ãƒƒã‚·ãƒ§ãƒ³ã®ç™ºè¡¨è³‡æ–™ã¯ã€ä»¥ä¸‹ã®URLã‹ã‚‰é–²è¦§ã™ã‚‹ã“ã¨ãŒã§ãã¾ã™ã€‚ http://www.slideshare.net/InadaNaoki/gocon2014-pprof 写真1　@methaneæ°ã®ç™ºè¡¨ã®æ§˜å サンプリングプãƒãƒ•ã‚¡ã‚¤ãƒ© pprofã¯ã‚µãƒ³ãƒ—リングプãƒãƒ•ã‚¡ã‚¤ãƒ©ã§ã€ä¸€å®šæ™‚é–“ã”ã¨ã«CPUãŒã©ã“を処ç†ã—ã¦ã„ã‚‹ã‹ã¨ã„ã†ã‚µãƒ³ãƒ—ルをå–ã‚‹ã“ã¨ã§ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ã‚’è¡Œã†ãã†ã§ã™ã€‚ãã®ãŸã‚ã€å¤šãã®ã‚µãƒ³ãƒ—ルãŒå–ã‚ŒãŸé–¢æ•°ã¯æ™‚é–“ãŒã‹ã‹ã£ã¦ã„ã‚‹ã¨ã„ã†ã“ã¨
Go ã§åˆ©ç”¨ã§ãるプãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ãƒ„ール pprof ã®èªã¿æ–¹ - Qiita
æ¦‚è¦ Go ã«ã¯ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ãƒ„ールãŒã¤ã„ã¦ã„ã‚‹ã®ã ã‘ã©ï¼Œä½•ã‚’出力ã—ã¦ãã‚Œã¦ã„ã‚‹ã®ã‹ã‚ˆã分ã‹ã‚‰ãªã‹ã£ãŸã®ã§ãƒ¡ãƒ¢ï¼Žé–“é•ã„ã«æ°—ã¥ã‹ã‚ŒãŸæ–¹ã„らã£ã—ゃã£ãŸã‚‰ã‚³ãƒ¡ãƒ³ãƒˆã„ãŸã ã‘ã‚Œã°å¹¸ã„ã§ã™ ． 出典 Russ Cox ã•ã‚“ãŒä¸‹è¨˜ã«æ›¸ã„ã¦ãã‚Œã¦ã¾ã—ãŸï¼Žã“れをèªã‚€ã®ãŒé–“é•ã„ãªã„ã§ã™ï¼Ž Russ Cox, Profiling Go Programs CPU Profiling プãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ç”¨ã®ã‚³ãƒ¼ãƒ‰ã®åŸ‹ã‚込㿠ã¾ãšï¼Œ runtime/pprof ã‚’ import ã—ã¦ãŠãã¾ã™ï¼Ž(net/http/pprofã¨ã„ã†ã®ã‚‚ã‚ã‚Šã¾ã™)． CPU Profiling ã¯è¨ˆæ¸¬ã—ãŸã„実行コードã®å‰ã§StartCPUProfile()を呼んã§ï¼Œè¨ˆæ¸¬çµ‚了ã®ã‚¿ã‚¤ãƒŸãƒ³ã‚°ã§pprof.StopCPUProfile()を呼ã³ã¾ã™ï¼Žçµ‚了ã®æ–¹ã¯ defer ã«ç™»éŒ²ã—ã¦ãŠã‘ã°é–¢æ•°æŠœã‘ã‚‹ã¨ãã«è‡ªå‹•ã§å‘¼ã°ã‚Œã‚‹ã®ã§ä¾¿åˆ©ã§ã™ãŒï¼ŒCtr-Cã¨
MySQLをプãƒãƒ•ã‚¡ã‚¤ã‚‹(ä»®)
5. How to profile your MySQL? â— dstat -apr â— ä»–ã®ã‚µãƒ¼ãƒãƒ¼ã¨æ¯”ã¹ã‚‹æ™‚ã«ä¾¿åˆ© â— SET GLOBAL ã™ã‚‹æ™‚ã¯ä¸€ç·’ã«æµã—ã¦ãŠã ◠定常値的ãªã‚‚ã®ã‚’把æ¡ã™ã‚‹ã®ã« ◠カーãƒãƒ«ã«ã‚ˆã£ã¦ã¯ -r 使ãˆãªã„ â— iostat -mx â— await, svctm ã‚ãŸã‚Š â— I/O ãªã®ã‹ãれ以外ãªã®ã‹â€¦ã¨ã„ã†ã‚ãŸã‚Š â— top -iH -p `pidof mysqld` ◠見るã®ã¯ State ãらㄠ◠"f"=> "J"(CPU ç•ªå· ), "P"(SWAP) 6. How to profile your MySQL? â— perf top -p `pidof mysqld` â— å½¹ã«ç«‹ã£ãŸã“ã¨ã¯ãªã„ã‘ã©ç´³å£«ã®ãŸã—ãªã¿ â— pt-ioprofile --cell sizes --run-time 10 â— ã©ã®ãƒ•ã‚¡ã‚¤ãƒ«ã«å¯¾ã—㦠I/O ã—ãŸã‹è¦‹
Javaã®CPUプãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚° - nekop's blog
JBoss / WildFly (全部俺) Advent Calendar 2013ã®19日目ã§ã™ã€‚Java一般ãªãƒˆãƒ”ック寄りã§ã€‚ 自分ã§ä¸»ã«åˆ©ç”¨ã™ã‚‹Javaã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ©ã¯YourKitãªã®ã§ã™ãŒã€YourKitを購入ã—ã¦ã„ãªã„å ´åˆã§æ€§èƒ½å•é¡Œã‚’解æžã™ã‚‹ã¨ãã¨ã‹ã«ä½•ãŒä½¿ãˆã‚‹ã‹ãªãã¨æ€ã£ã¦èª¿ã¹ã¦ã¿ã¾ã—ãŸã€‚ $ uname -a Linux localhost.localdomain 3.11.10-301.fc20.x86_64 #1 SMP Thu Dec 5 14:01:17 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux $ java -version java version "1.7.0_45" OpenJDK Runtime Environment (fedora-2.4.3.0.fc20-x86_64 u45-b15) OpenJDK 64-Bi
WSGI Application Profiler — Werkzeug Documentation (0.16.x)
Warning werkzeug.contrib.profiler has moved to werkzeug.middleware.profiler. The old import is deprecated as of version 0.15 and will be removed in version 1.0. class werkzeug.contrib.profiler.MergeStream(*streams)¶ An object that redirects write calls to multiple streams. Use this to log to both sys.stdout and a file: f = open('profiler.log', 'w') stream = MergeStream(sys.stdout, f) profiler = Pr
Devel::NYTProf 㧠Starlet/Starman (Plack) ã§ã†ã”ãウェブアプリケーションã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ã‚’ãŠã“ãªã†æ–¹æ³•ã¾ã¨ã‚ - tokuhirom's blog
Devel::NYTProf 㧠Starlet/Starman (Plack) ã§ã†ã”ãウェブアプリケーションã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒªãƒ³ã‚°ã‚’ãŠã“ãªã†æ–¹æ³•ã¾ã¨ã‚ Devel::NYTProf 㯠Perl5 ã®ä¸–ç•Œã§ã‚‚ã£ã¨ã‚‚人気ãŒã‚るプãƒãƒ•ã‚¡ã‚¤ãƒ©ã§ã‚る。表示ãŒç¾Žéº—ã§ã‚ã‚‹ã—ã€ã‚¹ãƒ†ãƒƒãƒ—ã”ã¨ã®å‡¦ç†é€Ÿåº¦ãŒç°¡å˜ã«ã‚ã‹ã‚‹ã®ã§ã¨ã¦ã‚‚便利ã 。 ãん㪠Devel::NYTProf ã§ã‚ã‚‹ãŒã€Starlet/Starman ã®ã‚ˆã†ãªãƒ—リフォークå¼ã®ã‚µãƒ¼ãƒãƒ¼ã§ã†ã”ãウェブアプリケーションã¨ãã¿ã‚ã‚ã›ã‚‹å ´åˆã®æ–¹æ³•è«–ã¨ã—ã¦ã€ã‚ã‹ã‚Šã‚„ã™ã„資料ãŒã¿ã‚ãŸã‚‰ãªã‹ã£ãŸã®ã§ã“ã“ã«è¨˜ã™ã‚‚ã®ã§ã‚る。 環境変数 NYTPROF ã‚’è¨å®šã™ã‚‹ Devel::NYTProf ã¯ç’°å¢ƒå¤‰æ•°ã§æŒ™å‹•ã‚’変ãˆã‚‰ã‚Œã‚‹ã€‚ plack ã¨ãã¿ã‚ã‚ã›ã‚‹å ´åˆã«ã¯ã€ä»¥ä¸‹ã®ã‚ˆã†ã«ã™ã‚‹ã¨ã‚ˆã„。 NYTPROF=sigexit=int:savesrc=0:start=no s
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
Profiling and Optimizing Go
GitHub - bradfitz/talk-yapc-asia-2015: talk-yapc-asia-2015
GitHub - rakyll/gom: A visual interface to work with runtime profiling data for Go
Client Challenge
{{#tags}}- {{label}}
{{/tags}}