[B! profiling] toritori0318ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯

toritori0318 id:toritori0318

profilingã«é–¢ã™ã‚‹toritori0318ã®ãƒ–ãƒƒã‚¯ãƒžãƒ¼ã‚¯ (5)

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

${{author_name}}$
{{author_name}}{{created}}
{{ #comment }}{{ comment }}{{ /comment }}
- {{ label }}

${{author_name}}$

{{{comment_expanded}}}

{{label}}

{{#is_bookmark}}ãƒªã‚¹ãƒˆ{{/is_bookmark}}{{^is_bookmark}}ãƒªãƒ³ã‚¯{{/is_bookmark}}

CPython ã® GC ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚° - methaneã®ãƒ–ãƒã‚°
ISUCON ã¯ Go ã§å‚æˆ¦ã—ã¦ã„ã‚‹ã‚“ã ã‘ã©ã€è¤‡æ•°ã®ãƒãƒ¼ãƒ ãŒ Python ã§äºˆé¸é€šéŽã—ãŸã‚‰ã—ã„ã®ã§ã€å¿œæ´ã®ãŸã‚ã« Tips ã‚’å…¬é–‹ã—ã¦ã„ã“ã†ã¨æ€ã†ã€‚ ç›®æ¬¡ CPython ã® GC ã«ã¤ã„ã¦ çµ±è¨ˆæƒ…å ±ã‚’å‡ºåŠ›ã™ã‚‹ ç¬¬ä¸€ä¸–ä»£GCã®é–“éš”ã‚’èª¿æ•´ã™ã‚‹ Out of Band GC å¾ªç’°å‚ç…§ã‚’è¦‹ã¤ã‘ã¦å¯¾å‡¦ã™ã‚‹ CPython ã® GC ã«ã¤ã„ã¦ CPython ã®GCã¯å‚ç…§ã‚«ã‚¦ãƒ³ãƒˆï¼‹å¾ªç’°å‚ç…§ã‚³ãƒ¬ã‚¯ã‚¿ã ã€‚ãã—ã¦å‚ç…§ã‚«ã‚¦ãƒ³ãƒˆæ–¹å¼ã¯(å¹¾ã¤ã‹ã®æ¬ ç‚¹ã¯ã‚ã‚‹ã‚‚ã®ã®)Webã‚¢ãƒ—ãƒªã®ãƒœãƒˆãƒ«ãƒãƒƒã‚¯ã«ãªã£ãŸã‚Šã¯ã—ã«ãã„ã€‚ ãªã®ã§GCãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã®åŸºæœ¬ã¯æ¬¡ã®ã‚ˆã†ã«ãªã‚‹ã€‚ å¾ªç’°å‚ç…§ã‚’é¿ã‘ã‚‹ å¾ªç’°å‚ç…§ã‚³ãƒ¬ã‚¯ã‚¿ã®å‘¼ã³å‡ºã—ã‚¿ã‚¤ãƒŸãƒ³ã‚°ã‚’åˆ¶å¾¡ã™ã‚‹ å¾ªç’°å‚ç…§ã‚³ãƒ¬ã‚¯ã‚¿ã¯ã€ç”Ÿãã¦ã„ã‚‹ã‚ªãƒ–ã‚¸ã‚§ã‚¯ãƒˆã®æ•°ãŒã‚ã‚‹ç¨‹åº¦å¢—ãˆã‚‹ã¨ç¬¬ä¸€ä¸–ä»£ãŒå®Ÿè¡Œã•ã‚Œã€ç¬¬ä¸€ä¸–ä»£ãŒä¸€å®šå›žæ•°å®Ÿè¡Œã•ã‚Œã‚‹ã¨ç¬¬äºŒä¸–ä»£ãŒã€ç¬¬äºŒä¸–ä»£ãŒä¸€å®šå›žæ•°å®Ÿè¡Œã•ã‚Œã‚‹ã¨ç¬¬ä¸‰ä¸–ä»£ãŒå®Ÿè¡Œã•ã‚Œã‚‹ã€‚ å„ä¸–ä»£
toritori0318 2015/10/14
python

tuning

profiling
ãƒªãƒ³ã‚¯
Linux Profiling at Netflix
This talk discusses Linux profiling using perf_events (also called "perf") based on Netflix's use of it. It covers how to use perf to get CPU profiling working and overcome common issues. The speaker will give a tour of perf_events features and show how Netflix uses it to analyze performance across their massive Amazon EC2 Linux cloud. They rely on tools like perf for customer satisfaction, cost o
toritori0318 2015/02/24
profiling

netflix

linux

performance
ãƒªãƒ³ã‚¯
MySQLã‚’ãƒ—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 â— å®šå¸¸å€¤çš„ãªã‚‚ã®ã‚’æŠŠæ¡ã™ã‚‹ã®ã« â— ã‚«ãƒ¼ãƒãƒ«ã«ã‚ˆã£ã¦ã¯ -r ä½¿ãˆãªã„ â— iostat -mx â— await, svctm ã‚ãŸã‚Š â— I/O ãªã®ã‹ãã‚Œä»¥å¤–ãªã®ã‹â€¦ã¨ã„ã†ã‚ãŸã‚Š â— top -iH -p `pidof mysqld` â— è¦‹ã‚‹ã®ã¯ State ãã‚‰ã„ â— "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—ã¦ I/O ã—ãŸã‹è¦‹
toritori0318 2014/04/15
mysql

profiling
ãƒªãƒ³ã‚¯
node-tick-processorã§nodejsã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ã‚’æ•´å½¢ã™ã‚‹ - Islands in the byte stream (legacy)
ã‚‚ã¨ã‚‚ã¨V8ã«ã¯ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«æ©Ÿèƒ½ãŒçµ„ã¿è¾¼ã¾ã‚Œã¦ã„ã¦ã€V8ã®ãƒªãƒã‚¸ãƒˆãƒªã«ã¯ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ãƒ‡ãƒ¼ã‚¿ã‚’æ•´å½¢ã—ã¦è¡¨ç¤ºã™ã‚‹ãŸã‚ã® *-tick-processor ã¨ã„ã†ã‚¹ã‚¯ãƒªãƒ—ãƒˆãŒå…¥ã£ã¦ã„ã‚‹ã€‚ https://github.com/v8/v8/tree/master/tools linux-tick-processor mac-tick-processor freebsd-tick-processor windows-tick-processor.bat ãã—ã¦ nodejs ã¯ --prof ã‚ªãƒ—ã‚·ãƒ§ãƒ³ã§V8ã®ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ã‚’æœ‰åŠ¹ã«ã—ã¦ã€ãƒ—ãƒãƒ•ã‚¡ã‚¤ãƒ«ãƒ‡ãƒ¼ã‚¿ã‚’å‡ºåŠ›ã§ãã‚‹ã€‚ãŸã ã€æ•´å½¢ã«ã¯V8ã® *-tick-processor ãŒå¿…è¦ã§ã€ã“ã® *-tick-processor ã‚’ä½¿ã†ã«ã¯V8ã‚’ãƒ“ãƒ«ãƒ‰ã—ãªã‘ã‚Œã°ãªã‚‰ãªã„ã€‚ *-tick-processor ã¯V8ã‚’ãƒ“ãƒ«ãƒ‰ã—ãŸã¨ãã«ã§ãã‚‹d8ã¨ã„ã†JSã®ã‚¤ãƒ³ã‚¿
toritori0318 2013/01/10
Node.js

v8

profiling
ãƒªãƒ³ã‚¯
Node.js â€” Profiling Node.js
It's incredibly easy to visualize where your Node program spends its time using DTrace and node-stackvis (a Node port of Brendan Gregg's FlameGraph tool): Run your Node.js program as usual. In another terminal, run: $ dtrace -n 'profile-97/execname == "node" && arg1/{ @[jstack(150, 8000)] = count(); } tick-60s { exit(0); }' > stacks.out This will sample about 100 times per second for 60 seconds an
toritori0318 2012/04/26
node.js

profiling
ãƒªãƒ³ã‚¯
1