Craig Shallahamer's Blog
|
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.
-------------------------------
<p>* db file parallel write: By the far the most common database writer wait event, a parallel write is simply a multiblock write, like most of the calls shown in Figure 6-24. This is the result of the database writer feeding off the write list and writing a batch of dirty blocks to disk. I like to see the wait time below 5 ms, but every organization has its own budget and service level requirements. A write less than 5 ms indicates write caching is working well. The write calls in Figure 6-24 range from around 1.5 ms to nearly 40 ms!
</p><p>* db file single write: This should never be the top wait event. This can possibly occur at the end of a checkpoint when all the database file header blocks must be written. These are done one at a time by multiple single-block writes.
</p><p>I look at IO issues in term of requirements and capacity. When there is an IO issue, the requirements have exceeded the capacity. The only exception is when a locking or blocking type of issue occurs, such as with a free buffer waits event (discussed in the next section). When you see a database file write issue, except for locking/blocking reasons, you know the IO requirements have exceeded the IO subsystem's capacity. With modern IO subsystems, I never assume the requirements are due to the database I'm working on, or even another Oracle database. With complex IO management comes the increased chance a different system's files are residing on the same physical disk as my database's files. So I'm very careful in my IO subsystem assumptions.
</p>
©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.
|
|
Know what's important before it's too late!
OraPub's Performance Training is like no other...
More Class Pics...
Get student testimonials!
|