03-29-2011 08:43 AM
The 7.1 release notes state:
Sybase ASA performance is poor with an UltraSparc-T series processor.
The Sybase ASA database does not perform well when Solaris is used with the
UltraSparc-T series processor. Thus, Symantec recommends that you do not
run your EMM server on this type of hardware.
Is there anyone out there in netbackup land that has gone to 7.1 and runs EMM on ultrasparc T?
Yeah I do, dont want to find the thing grinds to a halt, so interested in the experience of others
Curious to know the impact and why this is now frowned upon.Surely tis not beyond the realms of possiblity to determine why...and sort it?!
Thanks in advance,Jim
Solved! Go to Solution.
03-29-2011 02:22 PM
Seems the performance issue has been around for a while...
I belong to a 'veritas-bu' mailing list (http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu)
and the following email was sent out in June 2009:
"Has anyone here had performance problems with 6.5 masters running on Solaris with T5220 or T2 Sparc class processors? I have noticed that during periods of heavy job load the nbrb evaluation cycle, which appears to handled via a single threaded dbsrv9 process, can take forever and leave jobs stuck between requesting & granted resources states. Sometime for 15 minutes a time, depending on the job load.
I am familiar with nbrb.conf tweaks, however this doesn't make much difference in my environment where we have many disk-based destinations. Seems like in large environments, the master needs to be on a processor better suited for single threaded performance."
There were a couple of responses - all complaining about the same problem....
03-29-2011 01:27 PM
Well, this is an issue with Sybase and Oracle/Sun, not with NetBackup...but because Sybase is IN NetBackup, that's how it becomes a NetBackup problem.
What version are you running now? If you're at 7.0 already and not noticing any problems, you may continue to not notice any problems at 7.1. (How's that for covering my butt?)
03-29-2011 02:22 PM
Seems the performance issue has been around for a while...
I belong to a 'veritas-bu' mailing list (http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu)
and the following email was sent out in June 2009:
"Has anyone here had performance problems with 6.5 masters running on Solaris with T5220 or T2 Sparc class processors? I have noticed that during periods of heavy job load the nbrb evaluation cycle, which appears to handled via a single threaded dbsrv9 process, can take forever and leave jobs stuck between requesting & granted resources states. Sometime for 15 minutes a time, depending on the job load.
I am familiar with nbrb.conf tweaks, however this doesn't make much difference in my environment where we have many disk-based destinations. Seems like in large environments, the master needs to be on a processor better suited for single threaded performance."
There were a couple of responses - all complaining about the same problem....
03-31-2011 03:31 AM
CRZ,
Yup appreciate its not Netb direct however, we are talking about 2 major software suppliers, and surely world class suppliers cant just wash their hands of it...well OK I know they can, they all do. Already running 7.0.1 and its no deal.
I have a call logged with Symantec.
Thanks for info, CRZ and MVDB.
Jim
04-20-2011 09:16 AM
I know it doesn't add much to the conversation:
"There are technotes from Sybase regarding the performance issues when running Sybase ASA on that platform. The cause was due to the design of Sun Sparc and there is no work around to improve the performance.
So, if users choose UltraSparc-T as master server, a remote EMM should be suggested and that EMM server shouldn’t be Sun UltraSparc-T."
I haven't seen any of these technotes from Sybase, and I looked long and hard. I do know that the performance issues they were referencing can occur if you are using the OpCenter and Guided Application Restore screens (for Oracle) in particular.