ITPub博客

首页 > Linux操作系统 > Linux操作系统 > Detecting Potential I/O Bottlenecks in AIX

Detecting Potential I/O Bottlenecks in AIX

原创 Linux操作系统 作者:v_fantasy 时间:2009-04-13 23:31:44 0 删除 编辑

Using IBM's AIX dialect of UNIX, an I/O bound database server is usually evidenced by a high value in the wa (wait) column of vmstat. In the next listing, we see that 22 percent of the CPU time is being used waiting for database I/O.

L 5-19

root> vmstat
kthr     memory             page              faults        cpu    
----- ----------- ------------------------ ------------ -----------
r  b   avm   fre  re  pi  po  fr   sr  cy  in   sy  cs us sy id wa
0  0 217485   386  0   0   0   4   14   0 202  300 210 14 19 45 22

It is important to note that a high wait value does not always signal an I/O bottleneck. The CPU wait could be due to other factors, and this metric simply means that the CPU is waiting for external OS services.

Now that we understand the basics of the vmstat utility, let's explore an easy method for capturing vmstat data, storing it inside an Oracle table, and producing server performance reports.

来自 “ ITPUB博客 ” ,链接:http://blog.itpub.net/16158219/viewspace-588900/,如需转载,请注明出处,否则将追究法律责任。

请登录后发表评论 登录
全部评论

注册时间:2008-10-07

  • 博文量
    98
  • 访问量
    179843