Click here to monitor SSC
SQLServerCentral is supported by Red Gate Software Ltd.
 
Log in  ::  Register  ::  Not logged in
 
 
 
        
Home       Members    Calendar    Who's On


Add to briefcase

Understanding messages about long checkpoint Expand / Collapse
Author
Message
Posted Tuesday, September 3, 2013 3:48 AM
SSCrazy

SSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazySSCrazy

Group: General Forum Members
Last Login: Today @ 4:02 AM
Points: 2,122, Visits: 5,468
Hi all

Lately I've seen some messages about long checkpoints in our server. The messages look like this:

FlushCache: cleaned up 311 bufs with 275 writes in 100497 ms (avoided 7 new dirty bufs) for db 4:0
average throughput: 0.02 MB/sec, I/O saturation: 3139, context switches 6607
last target outstanding: 2, avgWriteLatency 627

I understand parts of this message. I do understand that the checkpoint had to write 311 pages and it did it by issuing 275 reads operation. This took 10004.97 seconds. The throughput was 0.02 mb/sec (kind of shocking, I think that hand writing can be quicker). What I don't understand is the other parts of this message (e.g. I/O saturation: 3139, context switches 6607 last target outstanding: 2, avgWritLatancy 627). Any explanation about this part would be greatly appreciated.

Adi


--------------------------------------------------------------
To know how to ask questions and increase the chances of getting asnwers:
http://www.sqlservercentral.com/articles/Best+Practices/61537/

For better answers on performance questions, click on the following...
http://www.sqlservercentral.com/articles/SQLServerCentral/66909/
Post #1490850
« Prev Topic | Next Topic »

Add to briefcase

Permissions Expand / Collapse