Use a bigger chunk size when transferring files across nodes #2738
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
|
Uffizzi Preview |
|
We can probably bump it even more? I'd expect the Erlang distribution to chunk large binaries naturally, so we don't have to worry about those concerns. I mean, in Nx we just send the whole binary tensor, and I haven't ran into issues. |
josevalim
approved these changes
Aug 9, 2024
|
@josevalim we still want to minimize the memory usage while sending the file, no? I kept it at 64kb, because bumping to 128kb didn't seem to bring further improvement :) |
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Outdated suggestions cannot be applied.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.
We transparently copy files from Livebook to remote runtime (Fly), for example, when using
Kino.Input.image. We noticed that it takes unusually long time to transfer the file. This increases file chunk size that we write across the nodes.Just as an example, I tried ~1MB .jpeg image and this change dropped the transfer time roughly from 70s to 4s.