-
Notifications
You must be signed in to change notification settings - Fork 178
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Remove GPU's total session runtimes! #121
Comments
Agreed! |
yes, and add better gpus options to finish projects more fast too (to wait less time on some projects) |
Agreed! 8 hours are definitely not enough as some projects tend to take more time usually than this. As aforementioned, you should bring back the old session timing for the GPUs and fix the "No runtime errors" which most of us faced recently. |
yes, that's exactly what we'd all like to see, the service getting back to how it was before when it started, 4 hour session limits and guaranteed GPU capacity for everyone, in my opinion the focus should be more on increasing the capacity of them than adding more limitations |
i agree as well this is very unfair some of my work used to take a very long time but i was still able to do them because there was good gpus i say bring back the old session system |
yes kindly remove that limit even if gpus are not guaranteed it's still much better |
i don't mind that too |
i hope they remove it soon and make gpu availability instead better |
I think it's okay to have limitation on free service. Otherwise some people will keep GPU on then it will not be available for other. Also having 8 hour per day is good option. I can do lot of thing in 8 hours. |
We all know it's a free service, but i disagree with you first of all it's a locked service meaning they get to choose who hops in and who don't, second of all no people cannot keep GPU's on anyways because there's already the 4 hour session limit and last thing is that you say you can do alot of things with it, well that's you which is personal you don't get to decide the same thing for all users having it back as before is what suits everyone and that's why the focus should be on increasing the capacity instead of such a limit i hope that's clear, have a great rest of your day! :) |
i agree! |
exactly, as i said again in my opinion the focus should be more on brining the service back to how it was back in 2021 |
mhm! in my opinion i think all of us can agree that we'd rather have gpus not being guaranteed and having the "gpu not available" error much much more than these limitations, sagemaker studio lab please hear our pleas!!! |
i don't believe they want to do it |
just stay patient man, it's not as easy as you might think we all want it to improve but give them some time |
in the meantime they probably have to keep it with this limitation, but hopefully within the next month or two it changes again to the better |
One year later, it's now 4 hours in total. |
and reduced cpu runtime. |
Is your feature request related to a problem? Please describe.
Yes, it's related to a lack of usage within the service
Describe the solution you'd like
I would like total session runtimes on GPU's to get back as they were or at least increase, since 8 hours is very low in my opinion, i know the service have had huge demand and issues recently but i'd like to at least know if there will be future plans for improving/brining the old system back if everything settles in!
Describe alternatives you've considered
Increasing total GPU runtime for at least 16-20 hours in a 24-hour period whenever GPU demand settles, i see that as a very good range of sessions
Additional context
None
The text was updated successfully, but these errors were encountered: