update margo usage to avoid runtime hangs #703
Merged
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Outdated suggestions cannot be applied.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
unifyfs_inode_extent_tandextent_tree_nodestructures, since they add useless bytes to the transfersNOTE: due to a bug in prior margo versions, we need to use 0.9.6 or later to use the timed async forwarding
Motivation and Context
At scales of 256 nodes and larger on Summit, we occasionally observe runtime hangs during various operations, particularly those involving group rpcs. These changes help to reduce any forced ordering of rpc responses, and improve diagnosis of responses never received. They also reduce the amount of data sent during synchronization of file extents.
How Has This Been Tested?
Tested on Summit up to 1024 nodes (8ppn) using writeread example (with and without lamination), with no hangs observed.
Types of changes
Checklist: