smartdispatch
smartdispatch copied to clipboard
SMART-Lab
Add configuration file for Cedar (#168)
Coverage increased (+1.2%) to 95.725% when pulling 2a380f6c2d2b85b263257c299b74abc0cf82cb0f on bouthilx:cedar_config into 9133e15ae91e1e5b0a2b7e90f442c1ebfb623e86 on SMART-Lab:master.
Does this completly fix https://github.com/SMART-Lab/smartdispatch/issues/168 ?
@nouiz It's a branch forked from adrien_slurm, the one PR #166 is based upon. Yes it should. The question is more: are those configurations the one we should define or not. We don't have strict queues like torque clusters, so it's up to us to decide how we define/devide queues.
Coverage increased (+1.2%) to 95.738% when pulling 02823d855f9e724bdca44ede17f875b25bb84ef1 on bouthilx:cedar_config into 9133e15ae91e1e5b0a2b7e90f442c1ebfb623e86 on SMART-Lab:master.
The whole smartdispatch seems to be somewhat on hold so I can't say for sure but, yes it was planned to be merged after a couple fixes.
On Thu, Apr 12, 2018 at 11:42 AM, Jonathan Zhouhan LIN < [email protected]> wrote:
Will this PR be merged in the future?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/SMART-Lab/smartdispatch/pull/173#issuecomment-380850617, or mute the thread https://github.com/notifications/unsubscribe-auth/ABBvm9t9SKr03GCWtvZ-X9NDsZHiDvcIks5tn3XdgaJpZM4P6_v- .
Thanks for letting me know!
Zhouhan
On Thu, Apr 12, 2018 at 2:12 PM, Mathieu Germain [email protected] wrote:
The whole smartdispatch seems to be somewhat on hold so I can't say for sure but, yes it was planned to be merged after a couple fixes.
On Thu, Apr 12, 2018 at 11:42 AM, Jonathan Zhouhan LIN < [email protected]> wrote:
Will this PR be merged in the future?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <https://github.com/SMART-Lab/smartdispatch/pull/173# issuecomment-380850617>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABBvm9t9SKr03GCWtvZ- X9NDsZHiDvcIks5tn3XdgaJpZM4P6_v-> .
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/SMART-Lab/smartdispatch/pull/173#issuecomment-380896628, or mute the thread https://github.com/notifications/unsubscribe-auth/AE4ojltesSHB6QBOLFjLzf9p8a_DgCvmks5tn5kSgaJpZM4P6_v- .
-- LIN, Zhouhan （林洲汉） Department of Computer Science and Operational Research University of Montreal
Phone: +01-514-586-0551 Email: [email protected] Homepage:http://zhouhanlin.webs.com/
@hantek SmartDispatch development is on hold for the moment as Mathieu pointed out. Given new clusters and future ones are all using Slurm and given lab users seem comfortable with Slurm itself we are not sure we want to maintain a dispatcher which wraps it. Furthermore, this PR is mostly a hack of SmartDispatch's PBS support. We aren't sure it's worth investing time in finalizing the PR and maintaining this hack.
If you want, you can try my branch https://github.com/bouthilx/smartdispatch/tree/slurm, it combines cedar and graham config and should be python 2.7/3.5 compatible.
Thanks for the clarification. So is there a tutorial on how to use slurm in clusters like cedar?
Not that I'm aware of. That would be a good thing to have. We definitely need to improve the documentation for cluster usage and also add more tutorials.