Skip to content
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

Do not materialize row partitions while doing reshuffling #5816

Closed
dchigarev opened this issue Mar 20, 2023 · 0 comments · Fixed by #5856
Closed

Do not materialize row partitions while doing reshuffling #5816

dchigarev opened this issue Mar 20, 2023 · 0 comments · Fixed by #5856
Assignees
Labels
partitions reshuffling 🔀 Issues related to partitions reshuffling Performance 🚀 Performance related issues and pull requests.

Comments

@dchigarev
Copy link
Collaborator

At the moment we're triggering .force_materialization() for row partitions in order to convert them to simple partitions that support the .split() method:

# Convert our list of block partitions to row partitions. We need to create full-axis
# row partitions since we need to send the whole partition to the split step as otherwise
# we wouldn't know how to split the block partitions that don't contain the shuffling key.
row_partitions = [
partition.force_materialization().list_of_block_partitions[0]
for partition in cls.row_partitions(partitions)
]

If we have implemented the .split() method for axis partitions we could have skipped this materialization step and thus save some time:

CPU: Intel(R) Xeon(R) Gold 6238R CPU @ 2.20GHz, 112 threads

case modin + #5780 modin + #5780 + draft impl for this tracker
repro from #5533 1.3s 1.1s
show-case from #4601 (comment) 3.9s 3.4s
@dchigarev dchigarev added Performance 🚀 Performance related issues and pull requests. partitions reshuffling 🔀 Issues related to partitions reshuffling labels Mar 20, 2023
@dchigarev dchigarev self-assigned this Mar 20, 2023
dchigarev added a commit to dchigarev/modin that referenced this issue Mar 24, 2023
RehanSD pushed a commit that referenced this issue Apr 13, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
partitions reshuffling 🔀 Issues related to partitions reshuffling Performance 🚀 Performance related issues and pull requests.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

1 participant