You were brought to this page based on an internet search
and as a free service to Oracle DBAs.
The text below is an except from the book,
Oracle Performance Firefighting, written by
Craig Shallahamer of
OraPub, Inc.
Figures and tables are not included on this page, only their reference.
To order the book in either print or PDF form, click
here.
©2009, 2010 by Craig Shallahamer. This is copyrighted material.
PleaseOut of respect for those involved in the creation of the book and also for
their familes, we ask you to respect the copyright both in intent and deed. Thank you.
-------------------------------
The database server bottleneck is the IO subsystem. Simply put, Oracle's IO read requirements have exceeded the IO subsystem's read capacity. We will expect our response-time graph to reflect the general queue time increase of an IO bottleneck, which has a continual and steady increase in queue time until the elbow of the curve is reached, and then response time skyrockets.
While the wait event situation, shown in Figure 9-14, looks like an IO bottleneck, especially with the 20 ms average db file scattered read time, there could be also be a CPU bottleneck. To double-check, calculate both the operating system and the Oracle CPU utilization. Using Figure 9-15 to calculate Oracle CPU requirements and considering both server process (DB CPU of 2,065.87 seconds) and background process (background cpu time of 25.95 seconds) CPU consumption, the total instance CPU consumption is 2,091.82 seconds.8 The database server CPU capacity is based on the four CPU cores and the 60-minute reporting interval.
Figure 9-14. Shown is a snippet from both v$system_event (wait events) and v$sys_time_model (v$sys_time_model, DB CPU). Oracle does not include background CPU when calculating "CPU time."
©2009, 2010 by Craig Shallahamer. This is copyrighted material.
PleaseOut of respect for those involved in the creation of the book and also for
their familes, we ask you to respect the copyright both in intent and deed. Thank you.
|