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

Request blocks from providers in response to workflow pressure. #46

Closed
yadudoc opened this Issue Dec 7, 2017 · 2 comments

Comments

Projects
None yet
1 participant
@yadudoc
Contributor

yadudoc commented Dec 7, 2017

Track available resources from each execution site, as well as workflow pressure to each site to make a determination about appropriately scaling to match workflow requirements.

This will need a few things :

  • Track outstanding tasks pending on a site
  • Implement algorithm that uses task overflow counters and timers to ensure balance of responsiveness
    vs effective measurement.
  • Integrate with provider interfaces to scale based on the task pressure algo

@yadudoc yadudoc added the enhancement label Dec 7, 2017

@yadudoc yadudoc added this to the Parsl-0.4.0 milestone Dec 7, 2017

@yadudoc yadudoc self-assigned this Dec 7, 2017

@yadudoc

This comment has been minimized.

Contributor

yadudoc commented Dec 12, 2017

The execution_providers also might need enhancement to support auto-polling of active blocks to ensure that failed blocks are re-submitted. However a better design option would be to have the DFK handle periodic polling of the providers to ensure that resources maintained at appropriate levels. Decision pending.

@yadudoc yadudoc modified the milestones: Parsl-0.4.0, Parsl-0.3.0 Dec 12, 2017

yadudoc added a commit that referenced this issue Jan 11, 2018

yadudoc added a commit that referenced this issue Jan 11, 2018

1. Updating sleep logic to sleep in 0.2s intervals till ipcontroller-…
…*json files are found

2. Fixes to scale_in and scale_out logic
3. Adding a status method to get outstanding tasks on a site #46

yadudoc added a commit that referenced this issue Jan 18, 2018

Adding FlowNoControl option for non-managed executors (for #46)
When executors with no scaling support are used, flow control may
not be necessary, and this option is by default used when only
non-managed executors are used, esp when flow diversion is not
supported.

yadudoc added a commit that referenced this issue Jan 22, 2018

Merge remote-tracking branch 'origin/flowcontrol'
    This merge include key functionality for flowcontrol(#46)
    as well as usage_tracking (#34). Documentation is pending
    and will be added straight to the master branch.
@yadudoc

This comment has been minimized.

Contributor

yadudoc commented Jan 22, 2018

This is in master now. closing.

@yadudoc yadudoc closed this Jan 22, 2018

benclifford pushed a commit that referenced this issue Aug 9, 2018

Moving infra creation for #46
If any failure happens during infra creation it is not properly
reported. This adds better reporting. @reidmcy to test.

annawoodard pushed a commit that referenced this issue Sep 24, 2018

Moving infra creation for #46
If any failure happens during infra creation it is not properly
reported. This adds better reporting. @reidmcy to test.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment