文件历史

作者	SHA1	备注	提交日期
Muhammed Fatih BALIN	69a532c1ab	[Feature] Gpu cache for node and edge data (#4341 ) Co-authored-by: xiny <xiny@nvidia.com>	2023-07-24 13:17:10 +08:00
Xin Yao	8d5d8962ad	[Refactor] Replace third_party/nccl with PyTorch's NCCL backend (#4989 ) * expose GeneratePermutation * add sparse_all_to_all_push * add sparse_all_to_all_pull * add unit test * handle world_size=1 * remove python nccl wrapper * remove the nccl dependency * use pinned memory to speedup D2H copy * fix lint * resolve comments * fix lint * fix ut * resolve comments	2023-03-08 12:59:10 +08:00
Xin Yao	9a00cf194f	[Feature] Import PyTorch's CUDA stream management (#4503 ) * add set_stream * add .record_stream for NDArray and HeteroGraph * refactor dgl stream Python APIs * test record_stream * add unit test for record stream * use pytorch's stream * fix lint * fix cpu build * address comments * address comments * add record stream tests for dgl.graph * record frames and update dataloder * add docstring * update frame * add backend check for record_stream * remove CUDAThreadEntry::stream * record stream for newly created formats * fix bug * fix cpp test * fix None c_void_p to c_handle	2022-09-15 16:54:52 +08:00
Xin Yao	5bef48dfda	[Performance] Optimize the use of alternative streams in dataloader (#4177 ) * fix using alternative streams * use a alternative stream for subgraph transferring * fix StreamContext when stream is None	2022-06-29 16:02:25 +08:00
nv-dlasalle	ae8dbe6d3c	[Feature][Performance] Implement NCCL wrapper for communicating NodeEmbeddings and sparse gradients. (#2825 ) * Split NCCL wrapper from sparse optimizer and sparse embedding * Add more unit tests for single node nccl * Fix unit test for tf * Switch to device histogram * Fix histgram issues * Finish migration to histogram * Handle cases with zero send/recieve data * Start on partition object * Get compiling * Updates * Add unit tests * Switch to partition object * Fix linting issues * Rename partition file * Add python doc * Fix python assert and finish doxygen comments * Remove stubs for range based partition to satisfy pylint * Wrap unit test in GPU only * Wrap explicit cuda call in ifdef * Merge with partition.py * update docstrings * Cleanup partition_op * Add Workspace object * Switch to using workspace object * Move last remainder based function out of nccl_api * Add error messages * Update docs with examples * Fix linting erros Co-authored-by: xiang song(charlie.song) <classicxsong@gmail.com>	2021-05-20 10:58:17 -07:00

5 次代码提交