大规模数据处理 云计算 Lecture 1 Introduction to MapReduce.ppt

* CNNIC发布《第26次中国互联网络发展状况统计报告》 /html/Dir/2010/07/15/5921.htm 中国互联网络信息中心（CNNIC） 2009年上半年中国互联网用户，从年初算起的半年间增长了13.4%，约4000万人，达到了3亿3千800万人（CNNIC2009年6月数据），通 过手机连接到互联网也迅速增加了32.1%。但从另一方面看，普及率仅为25.5%比发达国家低（日本：73.8%），因此目前预计仍然能保持这个速度增长。 * 2009年全国新闻出版业基本情况 /cms/html/21/493/201009/702538.html * * /programs/view/3lO90FaU9XY/ * * /archives/972/mind-boggling_exponential_growth.aspx Tell the story of 漫游者， 一个人可能获取的信息不断增加。当然，有可能知道，并不意味着你需要知道。 * I have a philosophy that if you throw everything, or nearly everything, away, the amount of time that you are likely to spend reacquiring something that you threw away and then found that you needed is much, much lower than the amount of time you might spend doing things with those things that you didn’t throw away. * 我们该怎样选择? Quality .vs. Quantity “马关条约” “经典作品” .vs. 30万出版物 * * LAMP is the industry standard But management is a hassle: Configuration, tuning Backup and recovery, disk space management Hardware failures, system crashes Software updates, security patches Log rotation, cron jobs, and much more Redesign needed once your database exceeds one box * .vs. 传统的EULA(End User License Agreement ) Level4, Scalable, Configurable, Multi-Tenant-Efficient：增加multitier architecutre，支持load-balanced farm of identical application instances，在数量变化的多个服务器上运行。Provider按用户需求demand来增减服务器，而不需要修改任何软件的体系结构。 /198/1988278.html * * /news.ars/postgoogle-and-ibm-team-on-cloud-computing-initiative-for-universitiesgoogle-and-ibm-team-on-cloud-computing-initiative-for-universities.html The idea for the program from Google senior software engineer Christophe Bisciglia despite the extreme talent of these potential job candidates, they sort of stalled when asked to think about algorithms the way that Google does. They just didnt have the background to think about computation in a fundamentally distributed manner, * /wired/archive/14.10/cloudware_pr.html * * /blog/archives/2007/07/yahoo-hadoop.html There are some big names involved with Cloudera.