[GE users] How important is checkpointing support?

dangruhn Dan.Gruhn at groupw.com
Fri May 15 13:51:03 BST 2009

We have started experimenting with setting up Windows users with Linux machines running a VMware Windows XP client.  Start up is with scripting to make things look as transparent to possible to the Windows user.  Our first trial is a 4 core machine which allocates 2 processors to Windows.

During the day, only 1 slot is used for Grid jobs.  At night, all 4 are used  So far so good, but the jobs are short enough that we don't worry about checkpoint/restart.

Checkpointing would be difficult for us as our jobs can involve MySQL database operations, multiple file I/O, and multi-level script and binary file execution.

My 2 cents worth.


agay wrote:
> Chris, thanks!
> Is there any reason your friends don't use BLCR:
> https://ftg.lbl.gov/CheckpointRestart/CheckpointRestart.shtml
> there is even a SGE howto for it:
> http://gridengine.sunsource.net/howto/APSTC-TB-2004-005.pdf
> The "hidden agenda" behind my post is to check an impression that checkpointing is rarely used and usually supports only a limited class of programs.
> I would like to hear community experience on turning interactive Windows farms into night time compute clusters. Did anyone try running SGE on UNIX while Windows runs in an emulator/VM/terminal server? Do you switch the operating system twice daily? Do you work only in Windows?
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=195555
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

Dan Gruhn
Group W Inc.
8315 Lee Hwy, Suite 303
Fairfax, VA, 22031
PH: (703) 752-5831
FX: (703) 752-5851


To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].

More information about the gridengine-users mailing list