GitLab

[Performance Improvement] Make GPU sampling and to_block use pinned memory to decrease required synchronization (#5685)

* adding LABOR sampling

* add ladies and pladies samplers

* fix compile error after rebase

* add reference for ladies sampler

* Improve ladies implementation.

* weighted labor sampling initial implementation draft
fix indentation and small bug in ladies script

* importance_sampling currently doesn't work with weights

* fix weighted importance sampling

* move labor example into its own folder

* lint fixes

* Improve documentation

* remove examples from the main PR

* fix linting by not using c++17 features

* fix documentation of labor_sampler.py

* update documentation for labor.py

* reformat the labor.py file with black

* fix linting errors

* replace exception use with if

* fix typo in error comment

* fixing win64 build for ci

* fixing weighted implementation, works now.

* fix bug in the weighted case and importance_sampling==0

* address part of the reviews

* remove unused code paths from cuda

* remove unused code path from cpu side

* remove extra features of labor making use of random seed.

* fix exclude_edges bug

* remove pcg and seed logic from cpu implementation, seed logic should still work for cuda.

* minor style change

* refactor CPU implementation, take out the importance_sampling probability computation into a function.

* improve CUDAWorkspaceAllocator

* refactor importance_sampling part out to a function

* minor optimization

* fix linting issue

* Revert "remove pcg and seed logic from cpu implementation, seed logic should still work for cuda."

This reverts commit c250e07ac6d7e13f57e79e8a2c2f098d777378c2.

* Revert "remove extra features of labor making use of random seed."

This reverts commit 7f99034353080308f4783f27d9a08bea343fb796.

* fix the documentation

* disable NIDs

* improve the documentation in the code

* use the stream argument in pcg32 instead of skipping ahead t times, can discard the use of hashmap now since it is faster this way.

* fix linting issue

* address another round of reviews

* further optimize CPU LABOR sampling implementation

* fix linting error

* update the comment

* reformat

* rename and rephrase comment

* fix formatting according to new linting specs

* fix compile error due to renaming, fix linting.

* lint

* rename DGLHeteroGraph to DGLGraph to match master

* replace other occurrences of DGLHeteroGraph to DGLGraph
Co-authored-by: Muhammed Fatih BALIN <m.f.balin@gmail.com>
Co-authored-by: Kaan Sancak <kaansnck@gmail.com>
Co-authored-by: Quan Gan <coin2028@hotmail.com>

* replace

* blabla

* balbla

* blabla
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>

* param

* brief

* note

* return

* tparam

* brief2

* file

* return2

* return

* blabla

* all
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>

* [Misc] clang-format auto fix.

* manual

* manual
Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>

Co-authored-by: Steve <ubuntu@ip-172-31-34-29.ap-northeast-1.compute.internal>

* disable warning for tensorpipe

* fix warning

* enable lint check for cuh files

* resolve comments

* rename `DLContext` to `DGLContext`

* rename `kDLGPU` to `kDLCUDA`

* replace DLTensor with DGLArray

* fix linting

* Unify DGLType and DLDataType to DGLDataType

* Fix FFI

* rename DLDeviceType to DGLDeviceType

* decouple dlpack from the core library

* fix bug

* fix lint

* fix merge

* fix build

* address comments

* rename dl_converter to dlpack_convert

* remove redundant comments

* 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

* Use an internal cuda stream for CopyDataFromTo

* small fix white space

* Fix to compile

* Make stream optional in copydata for compile

* fix lint issue

* Update cub functions to use internal stream

* Lint check

* Update CopyTo/CopyFrom/CopyFromTo to use internal stream

* Address comments

* Fix backward CUDA stream

* Avoid overloading CopyFromTo()

* Minor comment update

* Overload copydatafromto in cuda device api
Co-authored-by: xiny <xiny@nvidia.com>

* Fix a cub compile error for CUDA 11.5

* Fix comparison of integer expressions of different signedness in coo_sort.cu file

* Fix comparison of integer expressions of different signedness in cuda_compact_graph.cu file

* Remove never referenced variable in spmm.cu

* Fix comparison of integer expressions of different signedness in rowwise_pick.h file

* Fix comparison of integer expressions of different signedness in choice.cc file

* Remove never referenced variable col_data in spat_op_impl_coo.cc

* Remove never referenced variable allowed in global_uniform.cc

* Fix comparison of integer expressions of different signedness in graph.cc

* Fix comparison of integer expressions of different signedness in graph_apis.cc

* Fix the un-used ctx variable in ndarray_partition.cc file for cpu only build

* Fix comparison of integer expressions of different signedness in libra_partition.cc

* Fix comparison of integer expressions of different signedness in graph_op.cc
Co-authored-by: Triston Cao <tristonc@nvidia.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>

Co-authored-by: nv-dlasalle <63612878+nv-dlasalle@users.noreply.github.com>
Co-authored-by: Israt Nisa <neesha295@gmail.com>

* remove unnecessary induced vertices in EdgeSubgraph

* add unit test

* fix

* explain

* oops

* gpu compact graph template

* cuda compact graph draft

* fix typo

* compact graphs

* pass unit test but fail in training

* example using EdgeDataLoader on the GPU

* refactor cuda_compact_graph and cuda_to_block

* update training scripts

* fix linting

* fix linting

* fix exclude_edges for the GPU

* add --data-cpu & fix copyright

[Performance][Feature] Add `src_nodes` paramter to `to_block()` to avoid cost running unique() when available. (#2973)

* Add lhs_nodes are paremeter to to_block

* Update unit test

* Switch to simplified node conversion

* Switch lhs_nodes to be in/out parameter

* Update docs
Co-authored-by: Da Zheng <zhengda1936@gmail.com>
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

* add bruteforce impl

* add nn descent implementation

* change doc-string

* remove redundant func

* use local rng for cuda

* fix lint

* fix lint

* fix bug

* fix bug

* wrap nndescent_knn_graph into knn

* fix lint

* change function names

* add comment for dist funcs

* let the compiler do the unrolling

* use better blocksize setting

* remove redundant line

* check the return of the cub calls
Co-authored-by: Tong He <hetong007@gmail.com>

[Performance] Perform to_block on the GPU when the dataloader is created with a GPU `device`. (#3016)

* add output device for dataloading

* Update dataloader

* Get sampler device from dataloader

* Fix line length

* Update examples

* Fix to_block GPU for empty relation types

* Handle the case where the DistGraph has None for the underlying graph
Co-authored-by: Da Zheng <zhengda1936@gmail.com>

* add bruteforce impl

* add support for bruteforce-sharemem

* modify python API

* add tests

* change file path

* change python API

* fix lint

* fix test

* also check worst_dist in the last few dim

* use heap and early-stop on CPU

* fix lint

* fix lint

* add device check

* use cuda function to determine max shared mem

* use cuda to determine block info

* add memory free for tmp var

* update doc-string and add dist option

* fix lint

* add more tests
Co-authored-by: Quan (Andy) Gan <coin2028@hotmail.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>

* Add start of to_block gpu implementation

* Pull in more changes from 0.4.2 cuda_to_block

* Move more code to IdArray

* Refactor DeviceNodeMapMaker

* Updates

* get compiling

* Integrate to_block

* Fix ID allocation

* Minor fixes

* Cleanup cuda calls to use cuda_common

* Reduce kernel calls

* Lint cleanup

* Expand documentation

* Remove unused function

* Rename variables for consistency

* Add doxygen comments

* Fix file extension

* Remove raw asynccopy for deviceapi

* Remove unused function

* Fix block/tile configuration

* Add cuda_device_common.cuh

* Add basic hashtable

* Migrate part of hashtable

* Refactor to use external hashtable

* Make functions members

* Format hash table functions

* Migrate duplicate filling

* Move last function over

* Refactor with cu file

* lint c++ code

* Move context check to C++ code

* Use macro switch

* Add missing files

* Update docstring

* update docs

* Move atomic functions

* Refactor hashtable

* Fix linting

* Expand docs

* Fix mismatched argument names

* Switch doxygen comments from using @param to \param
Co-authored-by: Jinjing Zhou <VoVAllen@users.noreply.github.com>
Co-authored-by: Minjie Wang <wmjlyjemaine@gmail.com>