Add balanced mode to dispatch query #1103
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Dispatch balanced mode switch. When turned on, the dispatch query will use a formula that makes sure smaller jobs don't get starved by large jobs. This mode also disregards the layer_limit feature for performance implications. the formula takes into account the number of required cores and the time the job has been waiting on queue. Formula: rank = priority + (100 * (1 - (job.cores/job.int_min_cores))) + (age in days) This heuristic favors scenarios with a large amount of jobs and hosts and when turned off, only job priorities are taken into account, making the queue more predictable.
DiegoTavares
requested review from
bcipriano,
gregdenton,
jrray,
smith1511,
larsbijl,
IdrisMiles and
splhack
as code owners
DiegoTavares
changed the title
|
@bcipriano |
|
@bcipriano would you review this |
bcipriano
approved these changes
Apr 11, 2022
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Link the Issue(s) this Pull Request is related to.
Large studios with multiple shows and nodes can experience resource starving in smaller shows due to the priority model
Summarize your change.
Adds a scheduler_mode option to choose between
BALANCED,PRIORITY_ONLYandFIFO.The new mode (BALANCED) uses a formula that makes sure smaller jobs don't get starved by large jobs and also takes into account the time a job has been waiting in the queue. This mode also disregards the layer_limit feature for performance implications. The formula takes into account the number of required cores and the time the job has been waiting on queue.
Formula:
This heuristic favours scenarios with a large amount of jobs and hosts.