Tom , Thank you for your reply! I am encountering a context-switch storm problem . We got the pg_locks data when context-switch value over 200K/sec We fount that the value of CS relate to the count of Exclutivelocks . And I donnt know how to make the problem appear again by testing to collect evidence to update postgreSQL . So I want to redo your testing for that. Thank you！ Best regards， Ray Huang -----邮件原件----- 发件人: Tom Lane [mailto:tgl@xxxxxxxxxxxxx] 发送时间: 2010年5月27日 22:10 收件人: 黄永卫 抄送: pgsql-performance@xxxxxxxxxxxxxx 主题: Re: About Tom Lane's Xeon CS test case =?gb2312?B?u8bTwM7A?= <yongwei.huang@xxxxxxxxx> writes: > My postgres version: 8.1.3; You do realize that version was obsoleted four years ago last week? If you're encountering multiprocessor performance problems you really need to get onto 8.3.x or later. regards, tom lane -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
