Mon, 17 Oct 2011 16:00:00 GMTI learn to find my way, I learn to be myself.http://www.feedsky.com/feed/totaoy/sc/gifhttp://www.totaoy.com/bloghttp://www.totaoy.com/bloghourly1enMon, 29 Aug 2011 02:16:28 GMThttp://www.google.com/reader/shared/03930582029292863514#2011-10-18http://www.google.com/reader/shared/03930582029292863514#2011-10-18Tue, 18 Oct 2011 00:00:00 +0800<ul><li><a href='http://item.feedsky.com/~feedsky/kenwu/~7967402/565755899/5691693/1/item.html' title='link to ..'>HBase二级索引与Join</a><br/><div>二级索引与索引Join是Online业务系统要求存储引擎提供的基本特性。RDBMS支持得比较好，NOSQL阵营也在摸索着符合自身特点的最佳解决方案。 这篇文章会以HBase做为对象来探讨如何基于Hbase构建二级索引与实现索引join。文末同时会列出目前已知的包括0.19.3版secondary i..</div></li> </ul>mark dayhttp://item.feedsky.com/~feedsky/kenwu/~7967402/565755899/5691693/1/item.htmlhttp://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164415/6162548http://www.google.com/reader/shared/03930582029292863514#2011-09-30http://www.google.com/reader/shared/03930582029292863514#2011-09-30Fri, 30 Sep 2011 00:00:00 +0800<ul><li><a href='http://rdc.taobao.com/blog/cs/?p=1220' title='link to ..'>在JVM中加入CRC32 Intrinsic加速CRC校验</a><br/><div>问题： 在Java应用中例如在Hadoop中，经常用到CRC校验计算。对于Java开发者来说，目前可用的实现是java.util.zip类，他通过JNI调用zlib中crc32函数。这里有两个性能瓶颈。其一，JNI的调用性能很低，如果每次计算的数据量很小，比如1、2个byte，那么JNI的调用开销远远..</div></li> </ul>mark dayhttp://rdc.taobao.com/blog/cs/?p=1220http://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164416/6162548http://www.google.com/reader/shared/03930582029292863514#2011-09-16http://www.google.com/reader/shared/03930582029292863514#2011-09-16Fri, 16 Sep 2011 00:00:00 +0800<ul><li><a href='http://highscalability.com/blog/2011/9/15/paper-its-time-for-low-latency-inventing-the-1-microsecond-d.html' title='link to ..'>Paper: It's Time for Low Latency - Inventing the 1 Microsecond Datacenter</a><br/><div>In It's Time for Low Latency  Stephen Rumble et al. explore the idea that it&#39;s time to rearchitect our stack to live in the modern era of low-latency datacenter instead of high-latency WANs. The implications for program architectures will be ..</div></li> </ul>mark dayhttp://highscalability.com/blog/2011/9/15/paper-its-time-for-low-latency-inventing-the-1-microsecond-d.htmlhttp://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164417/6162548http://www.google.com/reader/shared/03930582029292863514#2011-09-10http://www.google.com/reader/shared/03930582029292863514#2011-09-10Sat, 10 Sep 2011 00:00:00 +0800<ul><li><a href='http://rdc.taobao.com/team/jm/archives/1124' title='link to ..'>Nginx走HttpProxy连JBoss在上传文件时的一个bug,及其解决</a><br/><div> 这是帮其他团队的同学解决的一个问题.这里做个记录. 问题描述: Nginx作为前端HttpServer 使用HttpProxy与JBoss进行连接 在上传比较大的文件时,JBoss已经返回结果,但是Nginx不立即将结果返回给浏览器,直到Nginx的HttpProxy与JBoss因为超时而断开后才返回结果给浏览器..</div></li> </ul>mark dayhttp://rdc.taobao.com/team/jm/archives/1124http://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164418/6162548http://www.google.com/reader/shared/03930582029292863514#2011-09-09http://www.google.com/reader/shared/03930582029292863514#2011-09-09Fri, 09 Sep 2011 00:00:00 +0800<ul><li><a href='http://www.iteye.com/news/22761' title='link to ..'>Java 8决定采用跟C#一样的Lambda句法</a><br/><div>Java 7 正式版已经发布，甲骨文和Java社区进程组织（JCP）的成员们开始考虑为这种编程语言的下一个版本Java SE 8添加新特性了。 经过对备选方案的筛选，JCP团队决定Java 8 将采用跟C#一样的Lambda句法，一些细节还在仔细考虑，涉及到方法引用的语法问题也还在讨论中..</div></li> </ul>mark dayhttp://www.iteye.com/news/22761http://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164419/6162548http://item.feedsky.com/~feedsky/totaoy/~8627228/551192744/6162548/1/item.html<p>有些东西，还是博客合适。<br /> 此后博客内容只分两类：生活、技术。</p><img src="http://www1.feedsky.com/t1/551192744/totaoy/feedsky/s.gif?r=http://item.feedsky.com/~feedsky/totaoy/~8627228/551192744/6162548/1/item.html" border="0" height="0" width="0" style="position:absolute" />http://www.totaoy.com/blog/life_22/microblogging-blog.html/feed0有些东西，还是博客合适。 此后博客内容只分两类：生活、技术。<img src="http://www1.feedsky.com/t1/551192744/totaoy/feedsky/s.gif?r=http://item.feedsky.com/~feedsky/totaoy/~8627228/551192744/6162548/1/item.html" border="0" height="0" width="0" style="position:absolute" />LifeMon, 29 Aug 2011 10:08:39 +0800BThttp://www.totaoy.com/blog/life_22/microblogging-blog.html#commentshttp://www.totaoy.com/blog/?p=22BThttp://www.totaoy.com/blog/life_22/microblogging-blog.htmlhttp://www.totaoy.com/blog/?feed=rss2feedsky/totaoy/~8627228/551192744/6162548http://www.google.com/reader/shared/03930582029292863514#2011-08-14http://www.google.com/reader/shared/03930582029292863514#2011-08-14Sun, 14 Aug 2011 00:00:00 +0800<ul><li><a href='http://blog.bluedavy.com/?p=309' title='link to ..'>HBase随机写以及随机读性能测试</a><br/><div>根据最近生产环境使用的经验，更多的项目的采用，以及采用了更加自动的测试平台，对HBase做了更多的场景的测试，在这篇blog中来分享下纯粹的随机写和随机读的性能数据，同时也分享下我们调整过后的参数。 测试环境说明： 1、Region Server： 5台，12块1T SATA盘(7200 ..</div></li> </ul>mark dayhttp://blog.bluedavy.com/?p=309http://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164420/6162548http://www.google.com/reader/shared/03930582029292863514#2011-07-19http://www.google.com/reader/shared/03930582029292863514#2011-07-19Tue, 19 Jul 2011 00:00:00 +0800<ul><li><a href='http://highscalability.com/blog/2011/7/18/new-relic-architecture-collecting-20-billion-metrics-a-day.html' title='link to ..'>New Relic Architecture - Collecting 20+ Billion Metrics a Day</a><br/><div> This is a guest post by Brian Doll, Application Performance Engineer at New Relic. New Relic’s multitenant, SaaS web application monitoring service collects and persists over 100,000 metrics every second on a sustained basis, while still delive..</div></li> <li><a href='http://highscalability.com/blog/2011/7/18/building-your-own-facebook-realtime-analytics-system.html' title='link to ..'>Building your own Facebook Realtime Analytics System</a><br/><div>Recently, I was reading Todd Hoff&#39;s write-up on FaceBook real time analytics system. As usual, Todd did an excellent job in summarizing this video from Engineering Manager at Facebook Alex Himel. In the first post, I’d like to summarize th..</div></li> </ul>mark dayhttp://highscalability.com/blog/2011/7/18/new-relic-architecture-collecting-20-billion-metrics-a-day.htmlhttp://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/607164421/6162548http://www.google.com/reader/shared/03930582029292863514#2011-06-23http://www.google.com/reader/shared/03930582029292863514#2011-06-23Thu, 23 Jun 2011 00:00:00 +0800<ul><li><a href='http://t.cn/a9mL2R' title='link to ..'>喷血……网络美女聚合网站，通过点击来进行质量推荐，不错的idea，肯定火爆！</a><br/><div></div></li> <li><a href='http://t.cn/a9mL2R' title='link to ..'>喷血……网络美女聚合网站，通过点击来进行质量推荐，不错的idea，肯定火爆！</a><br/><div></div></li> </ul>mark dayhttp://t.cn/a9mL2Rhttp://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/531447011/6162548http://www.google.com/reader/shared/03930582029292863514#2011-06-21http://www.google.com/reader/shared/03930582029292863514#2011-06-21Tue, 21 Jun 2011 00:00:00 +0800<ul><li><a href='http://item.feedsky.com/~feedsky/kenwu/~7967402/607904724/5691693/1/item.html' title='link to ..'>利用Arena Allocation避免HBase触发Full GC</a><br/><div>Arena Allocation，是一种GC优化技术，它可以有效地减少因内存碎片导致的Full GC，从而提高系统的整体性能。本文介绍Arena Allocation的原理及其在Hbase中的应用-MSLAB。 背景 假设有1G内存，我顺序创建了1百万个对象，每个对象大小1K，Heap会被渐渐充满且每个对象以..</div></li> </ul>mark dayhttp://item.feedsky.com/~feedsky/kenwu/~7967402/607904724/5691693/1/item.htmlhttp://www.google.com/reader/public/atom/user/03930582029292863514/state/com.google/broadcastfeedsky/totaoy/~8269169/608369276/6162548