Support of DCP with device order #162294
Open
Support of DCP with device order #162294
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
[ghstack-poisoned]
This was referenced Aug 30, 2025
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/162294
Note: Links to docs will display an error until the docs builds have been completed. ❌ 2 New FailuresAs of commit a7c1e62 with merge base 95a0532 ( NEW FAILURES - The following jobs have failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
pytorch-bot
bot
added
ciflow/inductor
oncall: distributed
|
Testing plan? |
Skylion007
reviewed
Sep 7, 2025
Yes, let me add tests in AP first and port it here. |
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 24, 2025
Not Ready For Review! Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 24, 2025
Not Ready For Review! Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 26, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 26, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 27, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 27, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 27, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 27, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Sep 28, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 2, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 3, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 3, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 5, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 5, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 6, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
zpcore
added a commit
that referenced
this pull request
Oct 6, 2025
***Not Ready For Review!*** ***Not Ready For Review!*** ***Not Ready For Review!*** ### Summary Introduce the `AllPermute` collective operation as mentioned in https://arxiv.org/pdf/2112.01075 "section 2.6 Collective operations". ### What is AllPermute? AllPermute can transform any 𝜏1 to 𝜏2 if their local and global shapes match. For example: Given mesh and size {X:4, Y:4, Z:16}, we have - example 1: [32{X,Y}}512, 128] -> [32{Y,X}512, 128] - example 2: [128{Y}512, 32{X}128] -> [128{X}512, 32{Y}128] - example 3: [32{X,Y}512, 128] -> [32{Z}512, 128] Note: annotation borrowed from https://arxiv.org/pdf/2112.01075 "section 2.1 Distributed array types" ### Why we need AllPermute? With AllPermute, we can eliminate some AllGather ops during redistribution. This plays an important role in reducing the memory overhead. In theory, at most one AllPermute is needed to redistribute from any 𝜏1 to 𝜏2. The `AllPermute` can be performed as the final step, or moved before the last `AllGather` to minimize the amount of data relocated between shards in the `AllPermute`. Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * __->__ #163772 * #162294 * #160903 * #160266 cc H-Huang awgu wanchaol fegin fduwjj wz337 wconstab d4l3k pragupta ezyang msaroufim dcci [ghstack-poisoned]
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):
cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta @ezyang @msaroufim