Tuple as device_type input to support Heterogenous Sharding of tables across different device_typestable #2600
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
facebook-github-bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D65933148 |
faran928
pushed a commit
to faran928/torchrec
that referenced
this pull request
Dec 4, 2024
… across different device_typestable (pytorch#2600) Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard Differential Revision: D65933148
faran928
force-pushed
the
export-D65933148
branch
from
2e3aa39 to
35d3c3a
Compare
|
This pull request was exported from Phabricator. Differential Revision: D65933148 |
faran928
pushed a commit
to faran928/torchrec
that referenced
this pull request
Dec 4, 2024
… across different device_typestable (pytorch#2600) Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard Differential Revision: D65933148
faran928
pushed a commit
to faran928/torchrec
that referenced
this pull request
Dec 4, 2024
… across different device_typestable (pytorch#2600) Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard Differential Revision: D65933148
added 2 commits
December 3, 2024 18:14
Summary: Unify InferRwSequenceEmbedding Modules for GPU / CPU. There does not seem to be much difference in the implementation for InferRwSequenceEmbedding and InferCPURwSequenceEmbedding. For heterogeneous sharding, we need to merge them together into one module. Also introduced the concept of device_type_from_sharding_info to propagate the correct device for output dist. Reviewed By: jiayisuse Differential Revision: D65859663
… across different device_typestable (pytorch#2600) Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard Differential Revision: D65933148
faran928
force-pushed
the
export-D65933148
branch
from
35d3c3a to
0bf4f59
Compare
|
This pull request was exported from Phabricator. Differential Revision: D65933148 |
faran928
pushed a commit
to faran928/torchrec
that referenced
this pull request
Dec 5, 2024
… across different device_typestable (pytorch#2600) Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard Differential Revision: D65933148
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.
Summary: As we plan to support heterogenous sharding across different device types (cuda / cpu etc), we will pass device type per shard in the format of tuple for device_type_from_sharding_info where each index will represent the device_type for that particular shard
Differential Revision: D65933148