[hyperactor] channel-level ping-pong benchmarks #906
Closed
+182
−4
Conversation
This file contains hidden or 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
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
This was referenced Aug 14, 2025
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303773788 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303773788 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#906 This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303699220 exported-using-ghexport Differential Revision: D80260732 Reviewed By: highker, vidhyav
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#906 This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303699220 exported-using-ghexport Differential Revision: D80260732 Reviewed By: highker, vidhyav
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#906 This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303699220 exported-using-ghexport Differential Revision: D80260732 Reviewed By: highker, vidhyav
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 303814001 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. Differential Revision: [D80260732](https://our.internmc.facebook.com/intern/diff/D80260732/) [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80260732 |
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 19, 2025
Summary: This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads. The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands. ghstack-source-id: 304032339 exported-using-ghexport Reviewed By: highker, vidhyav Differential Revision: D80260732
|
This pull request has been merged in 72b2ed7. |
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.
Stack from ghstack (oldest at bottom):
Message#852This is an attempt to do an apples-to-apples comparison to P1903314366, to eliminate any non-channel related overheads.
The results replicate previous findings: that our throughput is hampered by excess data copies (either outright or through growing buffers in the encoding stack), and of tokio-level network i/o overheads. Both are being addressed. These benchmarks should help to serve as validation as this work lands.
Differential Revision: D80260732