PostgreSQLã®è¡Œãƒ¬ãƒ™ãƒ«ã‚»ã‚ュリティ㨠SpringAOPã§ãƒžãƒ«ãƒãƒ†ãƒŠãƒ³ãƒˆã® ãƒ¦ãƒ¼ã‚¶ãƒ¼é–“æƒ…å ±æ¼æ´©ã‚’防æ¢ã™ã‚‹ (JJUG CCC 2021 Spring)
雑ãªMySQLパフォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°
PostgreSQLã®è¡Œãƒ¬ãƒ™ãƒ«ã‚»ã‚ュリティ㨠SpringAOPã§ãƒžãƒ«ãƒãƒ†ãƒŠãƒ³ãƒˆã® ãƒ¦ãƒ¼ã‚¶ãƒ¼é–“æƒ…å ±æ¼æ´©ã‚’防æ¢ã™ã‚‹ (JJUG CCC 2021 Spring)
Kazuho's Weblog: ウェブページã®æç”» (first-paint) ã¾ã§ã®æ™‚間を測定ã™ã‚‹ãƒ„ールを作ã£ãŸä»¶ã€ã‚‚ã—ãã¯HTTP2時代ã®ãƒ‘フォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã®è©±
ウェブページã®æç”» (first-paint) ã¾ã§ã®æ™‚間を測定ã™ã‚‹ãƒ„ールを作ã£ãŸä»¶ã€ã‚‚ã—ãã¯HTTP2時代ã®ãƒ‘フォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã®è©± ウェブページã®è¡¨ç¤ºã¾ã§ã«ã‹ã‹ã‚‹æ™‚é–“ã‚’ã„ã‹ã«çŸãã™ã‚‹ã‹ã£ã¦ã®ã¯ã€å„²ã‹ã‚‹ã‚¦ã‚§ãƒ–サイトを構築ã™ã‚‹ä¸Šã§é¿ã‘ã¦é€šã‚Œãªã„ã€ã¨ã¦ã‚‚é‡è¦ãªè¦ç´ ã§ã™ã€‚ å°‘ã—å¤ã„データã¨ã—ã¦ã¯ã€ãŸã¨ãˆã°ã€ã‚¦ã‚§ãƒ–ページã®è¡¨ç¤ºãŒ500ミリ秒é…ããªã‚‹ã¨åºƒå‘Šå£²ä¸ŠãŒ1.2%低下ã™ã‚‹ã¨ã„ã†Bingã®ä¾‹ãªã‚“ã‹ã‚‚知られã¦ã„ã‚‹ã‚ã‘ã§ã™ã€‚ 「ウェブページã®è¡¨ç¤ºã¾ã§ã«ã‹ã‹ã‚‹æ™‚é–“ã€ã¨è¨€ã£ãŸå ´åˆã€å®Ÿéš›ã«ã¯ä»¥ä¸‹ã®ã‚ˆã†ãªã„ãã¤ã‹ã®ãƒ¡ãƒˆãƒªãƒƒã‚¯ãŒã‚ã‚Šã¾ã™ã€‚ イベント æ„味
Redirecting…
Redirecting… Click here if you are not redirected.
Skylight
New: Source Locations Do the answer dance Don’t struggle to learn why your app is slow. Get answers with Skylight. Skylight is a smart profiler for Ruby and Rails applications. It turns performance data into actionable insights, so you spend less time diagnosing and more time improving. Performance data can be confusing Where's my app spending its time? How bad are my slowdowns? How can I improve
innodbã®ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã¨EC2ã®IO - å·¥å ´é•·ã®ãƒ–ãƒã‚°
ã“ã“ã§ã¯ã€EC2ã®ä¸Šã§innodbã‚’ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã—ã¦ä½¿ã†ã¨ã„ã†è¦³ç‚¹ã§TIPSã‚’ã¾ã¨ã‚ã¦ã¿ãŸã€‚ RDS便利ã ã‹ã‚‰ä½¿ãŠã†ãœã£ã¦ã®ã¯ä»Šå›žã®è©±ã®ã‚¹ã‚³ãƒ¼ãƒ—ã«ã¯å«ã¿ã¾ã›ã‚“。 ã‚ã¨ã€innodbã«ã¤ã„ã¦ã€å‰²ã‚Šã¨ã¡ã‚ƒã‚“ã¨èª¿ã¹ã¦ã¿ãŸã®ã¯åˆã‚ã¦ã ã£ãŸã‚Šã™ã‚‹ã®ã§ã€é–“é•ã£ã¦ã‚‹ç‚¹ãªã©è¦‹ã¤ã‘ãŸã‚‰ãœã²ã”指摘ãã ã•ã„ã¾ã›ã€‚ innodb関連 ãƒãƒƒãƒ•ã‚¡ãƒ—ール ワーã‚ングセットを乗ã›ã¦ãŠãオンメモリã®ãƒãƒƒãƒ•ã‚¡é ˜åŸŸã€‚èªã¿æ›¸ãå…±ã«ã“ã®é ˜åŸŸã‚’経由ã—ã¦å®Ÿæ–½ã•ã‚Œã‚‹ã€‚ å‚照時ã¯ãƒãƒƒãƒ•ã‚¡ãƒ—ール上ã§ãƒ‡ãƒ¼ã‚¿ã‚’探ã—ã¦ã€ãªã‘ã‚Œã°ãƒ†ãƒ¼ãƒ–ルスペースã‹ã‚‰å–å¾—ã™ã‚‹ã€‚（ãã®ãƒ‡ãƒ¼ã‚¿ã¯ãƒãƒƒãƒ•ã‚¡ãƒ—ール上ã«æ ¼ç´ã•ã‚Œã‚‹ï¼‰ 書ãè¾¼ã¿æ™‚ã¯ãƒªã‚¯ã‚¨ã‚¹ãƒˆã‚’å—ã‘付ã‘ã¦ãƒ¯ãƒ¼ã‚ングセットを更新ã—ã€ãƒã‚°ã®æ›¸ãè¾¼ã¿ã¸ç§»è¡Œã™ã‚‹ã€‚ è¨å®šã¯innodb_buffer_pool_size 監視ã¯SHOW ENGINE INNODB STATUSã‹ã€mysqladmin extended-sta
Etsukata blog: FreakOut DSP å…¥æœã‚µãƒ¼ãƒã® CPU 使用率を 30% 削減ã™ã‚‹ Performance Tuning
ã¯ã˜ã‚㫠勤務先㮠FreakOut 社ã§ã¯ RTB ã§åºƒå‘Šæž ã‚’è²·ã„付ã‘ã‚‹ DSP ã®é–‹ç™ºãƒ»é‹ç”¨ã‚’è¡Œã£ã¦ã„ã¾ã™ã€‚RTB ã¨ã¯ã€ã‚¤ãƒ³ã‚¿ãƒ¼ãƒãƒƒãƒˆåºƒå‘Šã®ã‚¤ãƒ³ãƒ—レッションãŒç”Ÿã˜ã‚‹æ¯Žã«ã€åºƒå‘Šæž ã®ç«¶äº‰å…¥æœã‚’è¡Œã†ä»•çµ„ã¿ã§ã™ã€‚ DSP ã¨ã¯ã€ RTB ã«ãŠã„ã¦ã€ç«¶äº‰å…¥æœã‚’ã™ã‚‹å´ã®ã‚·ã‚¹ãƒ†ãƒ ã«ãªã‚Šã¾ã™ã€‚åºƒå‘Šæž /広告を見ã¦ã„る人 ã«å¯¾ã—ã€æœ€é©ãªåºƒå‘Šã‚’ã€æœ€é©ãªã‚¿ã‚¤ãƒŸãƒ³ã‚°ã§å±Šã‘る機能を広告主ã«æä¾›ã™ã‚‹ä»•çµ„ã¿ã§ã™ã€‚ FreakOut DSP ã¯æœ€é©ãªåºƒå‘ŠæŽ¢ç´¢ãƒ»å…¥æœä¾¡æ ¼èª¿æ•´ã®ãŸã‚ã€éžå¸¸ã«å¤šãã®ãƒ‡ãƒ¼ã‚¿ã‚’å‚ç…§ã—ã€æ²¢å±±ã®æ¼”算処ç†ã‚’è¡Œã„ã¾ã™ã€‚広告を見ã¦ã„る人ãŒéŽåŽ»ã«ã‚¢ã‚¯ã‚»ã‚¹ã—㟠Web ページã®æƒ…å ±ã‚„æ¤œç´¢ãƒ¯ãƒ¼ãƒ‰ã€ã•ã‚‰ã« 広告ãŒã‚¯ãƒªãƒƒã‚¯ã•ã‚Œã‚‹äºˆæ¸¬ç¢ºçŽ‡(éŽåŽ»ã®ãƒã‚°ã‹ã‚‰æ©Ÿæ¢°å¦ç¿’ã§ç®—出) ãªã©ã‚’å‚ç…§ã—ã€å…¥æœä¾¡æ ¼ã‚’決定ã™ã‚‹ã®ã§ã™ã€‚ãã®ãŸã‚ã€DSP ã§å…¥æœã‚’担当ã™ã‚‹ã‚µãƒ¼ãƒã¯ CPU ãŒãƒœãƒˆãƒ«ãƒãƒƒã‚¯ã«ãªã£ã¦ãŠã‚Šã€å°æ•°ã‚‚数百å°ã«åµ©ã‚“ã§ã„ã¾ã™ã€‚
How to receive a million packets per second
Last week during a casual conversation I overheard a colleague saying: "The Linux network stack is slow! You can't expect it to do more than 50 thousand packets per second per core!" That got me thinking. While I agree that 50kpps per core is probably the limit for any practical application, what is the Linux networking stack capable of? Let's rephrase that to make it more fun: On Linux, how hard
Socket Sharding in NGINX Release 1.9.1
NGINX 1.9.1 introduces a new feature that enables use of the SO_REUSEPORT socket option, which is available in newer versions of many operating systems, including DragonFly BSD and Linux (kernel version 3.9 and later). This socket option allows multiple sockets to listen on the same IP address and port combination. The kernel then load balances incoming connections across the sockets. Editor – For
HPCユーザãŒçŸ¥ã£ã¦ãŠããŸã„TCP/IPã®è©± ~クラスタ・グリッド環境ã®è½ã¨ã—ç©´~
HPCユーザãŒçŸ¥ã£ã¦ãŠããŸã„TCP/IPã®è©± ~クラスタ・グリッド環境ã®è½ã¨ã—ç©´~AI-enhanced description The document discusses using TCP/IP for high-performance computing and describes how TCP performance is impacted by factors like round-trip time, bandwidth limitations, and window size. It provides measurements of bandwidth over TCP for different round-trip times and explains TCP congestion control algorithms and how they influence t
JAWS目黒 EC2ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°Tips #jawsmeguro #jawsug
2015/5/22 JAWS-UG 目黒ã§ç™ºè¡¨ã—ãŸè³‡æ–™ã§ã™ã€‚ http://connpass.com/event/14016/
Mackerelを支ãˆã‚‹æ™‚系列データベース技術 - ゆã†ã†ãブãƒã‚°
ã€è¿½è¨˜ 2018/01/06】ç¾åœ¨Mackerelã¯ã€æ™‚系列データベースã¨ã„ã†æ¦‚念をクラウドã®æŠ€ã§å†æ§‹ç¯‰ã™ã‚‹ - ゆã†ã†ãブãƒã‚°ã®æ™‚系列データベース実装ã¸ç§»è¡Œã—ã¦ã„ã¾ã™ã€‚ サーãƒãƒ¢ãƒ‹ã‚¿ãƒªãƒ³ã‚°ã‚µãƒ¼ãƒ“ス Mackerel ã§æŽ¡ç”¨ã—ã¦ã„る時系列データベース Graphite を用ã„ãŸã‚·ã‚¹ãƒ†ãƒ ã®æ§‹ç¯‰ã¨é‹ç”¨äº‹æƒ…を紹介ã—ã¾ã™ã€‚Graphiteã«ã¤ã„ã¦ã¯ã€ãƒ—ãƒãƒ“ジョニングやアプリケーションã‹ã‚‰ã®ä½¿ã„æ–¹ã€Graphite自体ã®ãƒ¢ãƒ‹ã‚¿ãƒªãƒ³ã‚°ãªã©æ§˜ã€…ãªãƒˆãƒ”ックãŒã‚ã‚Šã¾ã™ãŒã€ç‰¹ã«å¤§è¦æ¨¡ãªã‚‰ã§ã¯ã®ãƒˆãƒ”ックã¨ã—ã¦ã€Graphiteã®å†…部アーã‚テクãƒãƒ£ã€ãƒ‘フォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ãŠã‚ˆã³ã‚¯ãƒ©ã‚¹ã‚¿æ§‹æˆã«ã¤ã„ã¦ã®çŸ¥è¦‹ã‚’書ãã¾ã™ã€‚ 背景 Graphiteシステム概観 ãƒ‡ãƒ¼ã‚¿æ§‹é€ ã¨ã‚¢ãƒ¼ã‚テクãƒãƒ£ whisperã®ãƒ‡ãƒ¼ã‚¿æ§‹é€ carbon-cacheã®ã‚¢ãƒ¼ã‚テクãƒãƒ£ パフォーマンス特性 パフォーマンスãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚° ミドルウェアレ
GlusterFSã®ãƒ‘フォーマンス関連トピック
GlusterFSã®ãƒ‘フォーマンス関連トピック Presentation Transcript GlusterFS㮠パフォーマンス関連トピック ã‚‚ã‚Šã‚ã‹ã‹ã‹ãšãŠ 構æˆä¾‹ 10GbEスイッムx 2 10GbE接続 10GbE / 1GbE接続 ・・・ ストレージノード x 4 クライアントノード 群 s213s211 s212 s214 GlusterFS サームGlusterFS クライアント Native Clientイメージ gluster srv. gluster cli. gluster srv. gluster srv. gluster srv. glusterfs独自プãƒãƒˆã‚³ãƒ« FUSE Native Clientã®ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯è»¢é€é‡ ◠レプリケーション個数分ã€ã‚¯ãƒ©ã‚¤ã‚¢ãƒ³ãƒˆã¯å„サーãƒã« writeã‚’ãŠã“ãªã† ◠サーãƒé–“ã®é€šä¿¡ã¯åŸºæœ¬çš„ã«ç™ºç”Ÿã—ãªã„ ◠例: レプリケーシ
Deep Dive: Maximizing EC2 and EBS Performance
Amazon Elastic Block Store (Amazon EBS) provides persistent block level storage volumes for use with Amazon EC2 instances. In this technical session, we conduct a detailed analysis of the differences among the three types of Amazon EBS block storage: General Purpose (SSD), Provisioned IOPS (SSD), and Magnetic. We discuss how to maximize Amazon EBS performance, with a special eye towards low-latenc
Timing Details With cURL
You have likely heard of “The Golden Ruleâ€, which is usually described as: treat other people the way that you want to be treated. The New Testament version comes from Matthew 7:12 ( emphasis mine ): 12 Therefore all things whatsoever ye would that men should do to you, do ye even so to them: for this is the law and the prophets. While talking with others about the current political situation I th
faml 㨠slimã€hamlit ã®ãƒ‘フォーマンスã®å·® - eagletmt's blog
http://k0kubun.hatenablog.com/entry/2015/03/31/004021 を見ã¦ã€ã€Œfaml 㯠slim ã¨åŒç‰ã¨ã‹è¨€ã„ãªãŒã‚‰ slim よりãšã£ã¨é…ã„ã˜ã‚ƒã‚“ã€ã¨æ€ã‚れるã¨æ‚”ã—ã„ã®ã§ä¸€å¿œè§£èª¬ã—ã¦ãŠã。 ãªãŠã€ã“ã®ã‚¨ãƒ³ãƒˆãƒªã¯ slim 3.0.3 㨠hamlit 0.4.2 ã«åŸºã„ã¦ã„る。 ãªãœã“ã®ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ã§ faml ãŒé…ã„ã®ã‹ã€çµè«–ã‹ã‚‰è¨€ã†ã¨ã€ã“ã®ä¸ã§ faml ã ã‘自動 html エスケープãŒæœ‰åŠ¹ã«ãªã£ã¦ã„ã‚‹ã‹ã‚‰ã 。 haml ã¯ãƒ‡ãƒ•ã‚©ãƒ«ãƒˆã§ã¯è‡ªå‹• html エスケープã¯ç„¡åŠ¹ã§ã‚ã‚Šã€hamlit ã‚‚ãã‚Œã«å€£ã£ã¦ã„る。 slim ã®ãƒ™ãƒ³ãƒãƒžãƒ¼ã‚¯ã§ç”¨ã„ã¦ã„ã‚‹ slim ã®ãƒ“ューã§ã¯ == を使ã£ã¦æ˜Žç¤ºçš„ã« html エスケープを無効化ã—ã¦ã„ã‚‹ https://github.com/slim-template/slim/blob/v3.0.3/ben
Linuxã§ãƒãƒ¼ãƒ‰ãƒãƒ©ãƒ³ã‚µã‚„ã‚ャッシュサーãƒã‚’マルãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã•ã›ã‚‹ãŸã‚ã®ã‚«ãƒ¼ãƒãƒ«ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚° - ゆã†ã†ãブãƒã‚°
本記事ã®å…¬é–‹å¾Œã®2016å¹´7月ã«ã¯ã¦ãªã«ãŠã‘ã‚‹ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°äº‹ä¾‹ã‚’紹介ã—ãŸã€‚ ã¯ã¦ãªã«ãŠã‘ã‚‹Linuxãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ã‚¿ãƒƒã‚¯ãƒ‘フォーマンス改善 / Linux network performance improvement at hatena - Speaker Deck HAProxy ã‚„ nginx ãªã©ã®ã‚½ãƒ•ãƒˆã‚¦ã‚§ã‚¢ãƒãƒ¼ãƒ‰ãƒãƒ©ãƒ³ã‚µã‚„リãƒãƒ¼ã‚¹ãƒ—ãƒã‚ã‚·ã€memcached ãªã©ã® KVS ã®ã‚ˆã†ãªé«˜ãƒ‘ケットレートã«ãªã‚Šã‚„ã™ã„ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¢ãƒ—リケーションã«ãŠã„ã¦ã€å˜ä¸€ã® CPU コアã«è² è·ãŒåã‚Šã€ãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã—ãªã„ã“ã¨ãŒã‚ã‚Šã¾ã™ã€‚ 今回ã¯ã€ã“ã®ã‚ˆã†ãªãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¢ãƒ—リケーションã«ãŠã„㦠CPU è² è·ãŒãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã—ãªã„ç†ç”±ã¨ã€ãƒžãƒ«ãƒã‚³ã‚¢ã‚¹ã‚±ãƒ¼ãƒ«ã•ã›ã‚‹ãŸã‚ã® Linux カーãƒãƒ«ã®ãƒãƒƒãƒˆãƒ¯ãƒ¼ã‚¯ã‚¹ã‚¿ãƒƒã‚¯ã®ãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°æ‰‹æ³•ã¨ã—㦠RFS (Receive Flow Steering) ã‚’
DNS Prefetching
Problem DNS resolution time can lead to a significant amount of user perceived latency. The time that DNS resolution takes is highly variable. Latency delays range from around 1ms (locally cached results) to commonly reported times of several seconds. Solution DNS prefetching is an attempt to resolve domain names before a user tries to follow a link. This is done using the computer's normal DNS re
ランã‚ング
ランã‚ング
メンテナンス
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
Crafting performance alerting tools
Akka Actorã¨AMQPã§LINEã®ãƒ¡ãƒƒã‚»ãƒ¼ã‚¸ãƒ³ã‚°ãƒ‘イプラインをリプレースã—ãŸè©±
nginxã®ãƒ‘ラメータãƒãƒ¥ãƒ¼ãƒ‹ãƒ³ã‚°ã¨h2o - Qiita
{{#tags}}- {{label}}
{{/tags}}