Fix rounding error in htex block scale in #3721
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
|
@jrueb not sure if you're still interested, but this PR affects code you introduced so you might want to look at it |
khk-globus
approved these changes
Jan 6, 2025
Comment on lines
-301
to
303
| excess_slots = math.ceil(active_slots - (active_tasks * parallelism)) | ||
| excess_blocks = math.ceil(float(excess_slots) / (tasks_per_node * nodes_per_block)) | ||
| excess_slots = math.floor(active_slots - (active_tasks * parallelism)) | ||
| excess_blocks = math.floor(float(excess_slots) / (tasks_per_node * nodes_per_block)) | ||
| excess_blocks = min(excess_blocks, active_blocks - min_blocks) |
There was a problem hiding this comment.
I'm sure it's accounted for prior to this if-branch — and it exists prior to this hunk — but the first thing that comes to mind is "is it possible for excess_blocks to become negative?)
The if-condition guarantees that active_blocks - min_blocks is at least 1, but I'm not clear on the guarantees of active_slots - (active_tasks * parallelism)
There was a problem hiding this comment.
probably many of these values are assumed to be "sensible" and this code will go wrong if not. see for example this recent issue #3726
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Outdated suggestions cannot be applied.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.
Description
PR #2196 calculates a number of blocks to scale in, in the htex strategy, rather than scaling in one block per strategy iteration. However, it rounds the wrong way: it scales in a rounded up, rather than rounded down, number of blocks.
Issue #3696 shows that then resulting in oscillating behaviour: With 14 tasks and 48 workers per block, on alternating strategy runs, the code will either scale up to the rounded up number of needed blocks (14/48 => 1), or scale down to the rounded down number of needed blocks (14/48 => 0).
This PR changes the rounding introduced in #2196 to be consistent: rounding up the number of blocks to scale up, and rounding down the number of blocks to scale down.
Changed Behaviour
HTEX scale down should oscillate less
Fixes
Fixes #3696
Type of change