-
Notifications
You must be signed in to change notification settings - Fork 592
Creating RocksDBCheckpointHandler to expose rocksdb checkpoint to python #4224
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
Conversation
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
This pull request was exported from Phabricator. Differential Revision: D75489849 |
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
234ec36
to
4e573ea
Compare
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
4e573ea
to
7862394
Compare
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
This pull request was exported from Phabricator. Differential Revision: D75489849 |
…hon (pytorch#4224) Summary: Pull Request resolved: pytorch#4224 X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
7862394
to
a9a5f8e
Compare
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
a9a5f8e
to
0f230e0
Compare
This pull request was exported from Phabricator. Differential Revision: D75489849 |
1 similar comment
This pull request was exported from Phabricator. Differential Revision: D75489849 |
0f230e0
to
ab31efd
Compare
…hon (pytorch#4224) Summary: Pull Request resolved: pytorch#4224 X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
Differential Revision: D75489819
Differential Revision: D75489827
Differential Revision: D75489841
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
ab31efd
to
5a3ae1f
Compare
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
5a3ae1f
to
eac1779
Compare
…hon (pytorch#4224) Summary: X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
This pull request was exported from Phabricator. Differential Revision: D75489849 |
eac1779
to
2a4bf51
Compare
…hon (pytorch#4224) Summary: Pull Request resolved: pytorch#4224 X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
…hon (pytorch#4224) Summary: Pull Request resolved: pytorch#4224 X-link: facebookresearch/FBGEMM#1302 Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54 Context: We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold In this diff: 1. Creating a RocksDBCheckpointHandlerWrapper class to expose the rocksdb checkpoint to python 2. Adding a checkpointhandle attribute to the KVTensorWrapper 3. Adding changes in the training loop to accommodate creation of KVTensor with checkpointHandle Reviewed By: duduyi2013 Differential Revision: D75489849
This pull request was exported from Phabricator. Differential Revision: D75489849 |
2a4bf51
to
dfa7ada
Compare
This pull request has been merged in 05763d3. |
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/1302
Design doc: https://docs.google.com/document/d/149LdAEHOLP7ei4hwVVkAFXGa4N9uLs1J7efxfBZp3dY/edit?tab=t.0#heading=h.49t3yfaqmt54
Context:
We are enabling the usage of rocksDB checkpoint feature in KVTensorWrapper. This allows us to create checkpoints of the embedding tables in SSD. Later, these checkpoints are used by the checkpointing component to create a checkpoint and upload it it to the manifold
In this diff:
Differential Revision: D75489849