-
Notifications
You must be signed in to change notification settings - Fork 90
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
Add BatchMultiVector class, core, kernels and tests #1371
Conversation
format! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think I will break up my review in a few pieces. This first one looks mostly at the interface and core stuff.
I fell that some of my comments are maybe too detailed and might not be critical to this PR. If you think that's the case, then those comments can be handled in another PR. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
First coarse pass with a handful of specific suggestions:
- we will be doing a lot of scheduling that is the same everywhere - looping over all batches in thread blocks, looping over all entries inside a batch. I would suggest to provide abstractions for those to a) give a single place where we can work on scheduling b) reduce the likelihood of indexing mistakes during development and c) give explicit names to the different levels of parallelism we are using here
- we don't need raw reduction implementations, we have abstractions for those already.
- with certain functions (read, write, unbatch), it seems like they are written mainly for testability purposes, I think we should find clear demarcations which functions are supposed to be used in applications, and which are mostly there for testing purposes (and decide whether we even want to expose them to users easily)
format! |
4b98377
to
a60ae1e
Compare
format! |
3fe28eb
to
62e84a2
Compare
format! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM in general. There's some potential performance improvement we can try.
Should we also have the stride for the BatchMultiVector?
Co-authored-by: Pratik Nayak <[email protected]>
Cannot use sycl::reduce_over_group for older DPCPP versions.
Co-authored-by: Yu-Hsiang Mike Tsai <[email protected]>
Co-authored-by: Yu-Hsiang Tsai<[email protected]>
Co-authored-by: Pratik Nayak <[email protected]>
Co-authored-by: Thomas Grützmacher <[email protected]> Co-authored-by: Yu-Hsiang Tsai <[email protected]> Co-authored-by: Marcel Koch <[email protected]>
Co-authored-by: Pratik Nayak <[email protected]>
Co-authored-by: Pratik Nayak <[email protected]>
Kudos, SonarCloud Quality Gate passed! 0 Bugs 83.9% Coverage The version of Java (11.0.3) you have used to run this analysis is deprecated and we will stop accepting it soon. Please update to at least Java 17. |
Release 1.7.0 to master The Ginkgo team is proud to announce the new Ginkgo minor release 1.7.0. This release brings new features such as: - Complete GPU-resident sparse direct solvers feature set and interfaces, - Improved Cholesky factorization performance, - A new MC64 reordering, - Batched iterative solver support with the BiCGSTAB solver with batched Dense and ELL matrix types, - MPI support for the SYCL backend, - Improved ParILU(T)/ParIC(T) preconditioner convergence, and more! If you face an issue, please first check our [known issues page](https://github.com/ginkgo-project/ginkgo/wiki/Known-Issues) and the [open issues list](https://github.com/ginkgo-project/ginkgo/issues) and if you do not find a solution, feel free to [open a new issue](https://github.com/ginkgo-project/ginkgo/issues/new/choose) or ask a question using the [github discussions](https://github.com/ginkgo-project/ginkgo/discussions). Supported systems and requirements: + For all platforms, CMake 3.16+ + C++14 compliant compiler + Linux and macOS + GCC: 5.5+ + clang: 3.9+ + Intel compiler: 2019+ + Apple Clang: 14.0 is tested. Earlier versions might also work. + NVHPC: 22.7+ + Cray Compiler: 14.0.1+ + CUDA module: CMake 3.18+, and CUDA 10.1+ or NVHPC 22.7+ + HIP module: ROCm 4.5+ + DPC++ module: Intel oneAPI 2022.1+ with oneMKL and oneDPL. Set the CXX compiler to `dpcpp` or `icpx`. + MPI: standard version 3.1+, ideally GPU Aware, for best performance + Windows + MinGW: GCC 5.5+ + Microsoft Visual Studio: VS 2019+ + CUDA module: CUDA 10.1+, Microsoft Visual Studio + OpenMP module: MinGW. ### Version support changes + CUDA 9.2 is no longer supported and 10.0 is untested [#1382](#1382) + Ginkgo now requires CMake version 3.16 (and 3.18 for CUDA) [#1368](#1368) ### Interface changes + `const` Factory parameters can no longer be modified through `with_*` functions, as this breaks const-correctness [#1336](#1336) [#1439](#1439) ### New Deprecations + The `device_reset` parameter of CUDA and HIP executors no longer has an effect, and its `allocation_mode` parameters have been deprecated in favor of the `Allocator` interface. [#1315](#1315) + The CMake parameter `GINKGO_BUILD_DPCPP` has been deprecated in favor of `GINKGO_BUILD_SYCL`. [#1350](#1350) + The `gko::reorder::Rcm` interface has been deprecated in favor of `gko::experimental::reorder::Rcm` based on `Permutation`. [#1418](#1418) + The Permutation class' `permute_mask` functionality. [#1415](#1415) + Multiple functions with typos (`set_complex_subpsace()`, range functions such as `conj_operaton` etc). [#1348](#1348) ### Summary of previous deprecations + `gko::lend()` is not necessary anymore. + The classes `RelativeResidualNorm` and `AbsoluteResidualNorm` are deprecated in favor of `ResidualNorm`. + The class `AmgxPgm` is deprecated in favor of `Pgm`. + Default constructors for the CSR `load_balance` and `automatical` strategies + The PolymorphicObject's move-semantic `copy_from` variant + The templated `SolverBase` class. + The class `MachineTopology` is deprecated in favor of `machine_topology`. + Logger constructors and create functions with the `executor` parameter. + The virtual, protected, Dense functions `compute_norm1_impl`, `add_scaled_impl`, etc. + Logger events for solvers and criterion without the additional `implicit_tau_sq` parameter. + The global `gko::solver::default_krylov_dim`, use instead `gko::solver::gmres_default_krylov_dim`. ### Added features + Adds a batch::BatchLinOp class that forms a base class for batched linear operators such as batched matrix formats, solver and preconditioners [#1379](#1379) + Adds a batch::MultiVector class that enables operations such as dot, norm, scale on batched vectors [#1371](#1371) + Adds a batch::Dense matrix format that stores batched dense matrices and provides gemv operations for these dense matrices. [#1413](#1413) + Adds a batch::Ell matrix format that stores batched Ell matrices and provides spmv operations for these batched Ell matrices. [#1416](#1416) [#1437](#1437) + Add a batch::Bicgstab solver (class, core, and reference kernels) that enables iterative solution of batched linear systems [#1438](#1438). + Add device kernels (CUDA, HIP, and DPCPP) for batch::Bicgstab solver. [#1443](#1443). + New MC64 reordering algorithm which optimizes the diagonal product or sum of a matrix by permuting the rows, and computes additional scaling factors for equilibriation [#1120](#1120) + New interface for (non-symmetric) permutation and scaled permutation of Dense and Csr matrices [#1415](#1415) + LU and Cholesky Factorizations can now be separated into their factors [#1432](#1432) + New symbolic LU factorization algorithm that is optimized for matrices with an almost-symmetric sparsity pattern [#1445](#1445) + Sorting kernels for SparsityCsr on all backends [#1343](#1343) + Allow passing pre-generated local solver as factory parameter for the distributed Schwarz preconditioner [#1426](#1426) + Add DPCPP kernels for Partition [#1034](#1034), and CSR's `check_diagonal_entries` and `add_scaled_identity` functionality [#1436](#1436) + Adds a helper function to create a partition based on either local sizes, or local ranges [#1227](#1227) + Add function to compute arithmetic mean of dense and distributed vectors [#1275](#1275) + Adds `icpx` compiler supports [#1350](#1350) + All backends can be built simultaneously [#1333](#1333) + Emits a CMake warning in downstream projects that use different compilers than the installed Ginkgo [#1372](#1372) + Reordering algorithms in sparse_blas benchmark [#1354](#1354) + Benchmarks gained an `-allocator` parameter to specify device allocators [#1385](#1385) + Benchmarks gained an `-input_matrix` parameter that initializes the input JSON based on the filename [#1387](#1387) + Benchmark inputs can now be reordered as a preprocessing step [#1408](#1408) ### Improvements + Significantly improve Cholesky factorization performance [#1366](#1366) + Improve parallel build performance [#1378](#1378) + Allow constrained parallel test execution using CTest resources [#1373](#1373) + Use arithmetic type more inside mixed precision ELL [#1414](#1414) + Most factory parameters of factory type no longer need to be constructed explicitly via `.on(exec)` [#1336](#1336) [#1439](#1439) + Improve ParILU(T)/ParIC(T) convergence by using more appropriate atomic operations [#1434](#1434) ### Fixes + Fix an over-allocation for OpenMP reductions [#1369](#1369) + Fix DPCPP's common-kernel reduction for empty input sizes [#1362](#1362) + Fix several typos in the API and documentation [#1348](#1348) + Fix inconsistent `Threads` between generations [#1388](#1388) + Fix benchmark median condition [#1398](#1398) + Fix HIP 5.6.0 compilation [#1411](#1411) + Fix missing destruction of rand_generator from cuda/hip [#1417](#1417) + Fix PAPI logger destruction order [#1419](#1419) + Fix TAU logger compilation [#1422](#1422) + Fix relative criterion to not iterate if the residual is already zero [#1079](#1079) + Fix memory_order invocations with C++20 changes [#1402](#1402) + Fix `check_diagonal_entries_exist` report correctly when only missing diagonal value in the last rows. [#1440](#1440) + Fix checking OpenMPI version in cross-compilation settings [#1446](#1446) + Fix false-positive deprecation warnings in Ginkgo, especially for the old Rcm (it doesn't emit deprecation warnings anymore as a result but is still considered deprecated) [#1444](#1444) ### Related PR: #1451
Release 1.7.0 to develop The Ginkgo team is proud to announce the new Ginkgo minor release 1.7.0. This release brings new features such as: - Complete GPU-resident sparse direct solvers feature set and interfaces, - Improved Cholesky factorization performance, - A new MC64 reordering, - Batched iterative solver support with the BiCGSTAB solver with batched Dense and ELL matrix types, - MPI support for the SYCL backend, - Improved ParILU(T)/ParIC(T) preconditioner convergence, and more! If you face an issue, please first check our [known issues page](https://github.com/ginkgo-project/ginkgo/wiki/Known-Issues) and the [open issues list](https://github.com/ginkgo-project/ginkgo/issues) and if you do not find a solution, feel free to [open a new issue](https://github.com/ginkgo-project/ginkgo/issues/new/choose) or ask a question using the [github discussions](https://github.com/ginkgo-project/ginkgo/discussions). Supported systems and requirements: + For all platforms, CMake 3.16+ + C++14 compliant compiler + Linux and macOS + GCC: 5.5+ + clang: 3.9+ + Intel compiler: 2019+ + Apple Clang: 14.0 is tested. Earlier versions might also work. + NVHPC: 22.7+ + Cray Compiler: 14.0.1+ + CUDA module: CMake 3.18+, and CUDA 10.1+ or NVHPC 22.7+ + HIP module: ROCm 4.5+ + DPC++ module: Intel oneAPI 2022.1+ with oneMKL and oneDPL. Set the CXX compiler to `dpcpp` or `icpx`. + MPI: standard version 3.1+, ideally GPU Aware, for best performance + Windows + MinGW: GCC 5.5+ + Microsoft Visual Studio: VS 2019+ + CUDA module: CUDA 10.1+, Microsoft Visual Studio + OpenMP module: MinGW. ### Version support changes + CUDA 9.2 is no longer supported and 10.0 is untested [#1382](#1382) + Ginkgo now requires CMake version 3.16 (and 3.18 for CUDA) [#1368](#1368) ### Interface changes + `const` Factory parameters can no longer be modified through `with_*` functions, as this breaks const-correctness [#1336](#1336) [#1439](#1439) ### New Deprecations + The `device_reset` parameter of CUDA and HIP executors no longer has an effect, and its `allocation_mode` parameters have been deprecated in favor of the `Allocator` interface. [#1315](#1315) + The CMake parameter `GINKGO_BUILD_DPCPP` has been deprecated in favor of `GINKGO_BUILD_SYCL`. [#1350](#1350) + The `gko::reorder::Rcm` interface has been deprecated in favor of `gko::experimental::reorder::Rcm` based on `Permutation`. [#1418](#1418) + The Permutation class' `permute_mask` functionality. [#1415](#1415) + Multiple functions with typos (`set_complex_subpsace()`, range functions such as `conj_operaton` etc). [#1348](#1348) ### Summary of previous deprecations + `gko::lend()` is not necessary anymore. + The classes `RelativeResidualNorm` and `AbsoluteResidualNorm` are deprecated in favor of `ResidualNorm`. + The class `AmgxPgm` is deprecated in favor of `Pgm`. + Default constructors for the CSR `load_balance` and `automatical` strategies + The PolymorphicObject's move-semantic `copy_from` variant + The templated `SolverBase` class. + The class `MachineTopology` is deprecated in favor of `machine_topology`. + Logger constructors and create functions with the `executor` parameter. + The virtual, protected, Dense functions `compute_norm1_impl`, `add_scaled_impl`, etc. + Logger events for solvers and criterion without the additional `implicit_tau_sq` parameter. + The global `gko::solver::default_krylov_dim`, use instead `gko::solver::gmres_default_krylov_dim`. ### Added features + Adds a batch::BatchLinOp class that forms a base class for batched linear operators such as batched matrix formats, solver and preconditioners [#1379](#1379) + Adds a batch::MultiVector class that enables operations such as dot, norm, scale on batched vectors [#1371](#1371) + Adds a batch::Dense matrix format that stores batched dense matrices and provides gemv operations for these dense matrices. [#1413](#1413) + Adds a batch::Ell matrix format that stores batched Ell matrices and provides spmv operations for these batched Ell matrices. [#1416](#1416) [#1437](#1437) + Add a batch::Bicgstab solver (class, core, and reference kernels) that enables iterative solution of batched linear systems [#1438](#1438). + Add device kernels (CUDA, HIP, and DPCPP) for batch::Bicgstab solver. [#1443](#1443). + New MC64 reordering algorithm which optimizes the diagonal product or sum of a matrix by permuting the rows, and computes additional scaling factors for equilibriation [#1120](#1120) + New interface for (non-symmetric) permutation and scaled permutation of Dense and Csr matrices [#1415](#1415) + LU and Cholesky Factorizations can now be separated into their factors [#1432](#1432) + New symbolic LU factorization algorithm that is optimized for matrices with an almost-symmetric sparsity pattern [#1445](#1445) + Sorting kernels for SparsityCsr on all backends [#1343](#1343) + Allow passing pre-generated local solver as factory parameter for the distributed Schwarz preconditioner [#1426](#1426) + Add DPCPP kernels for Partition [#1034](#1034), and CSR's `check_diagonal_entries` and `add_scaled_identity` functionality [#1436](#1436) + Adds a helper function to create a partition based on either local sizes, or local ranges [#1227](#1227) + Add function to compute arithmetic mean of dense and distributed vectors [#1275](#1275) + Adds `icpx` compiler supports [#1350](#1350) + All backends can be built simultaneously [#1333](#1333) + Emits a CMake warning in downstream projects that use different compilers than the installed Ginkgo [#1372](#1372) + Reordering algorithms in sparse_blas benchmark [#1354](#1354) + Benchmarks gained an `-allocator` parameter to specify device allocators [#1385](#1385) + Benchmarks gained an `-input_matrix` parameter that initializes the input JSON based on the filename [#1387](#1387) + Benchmark inputs can now be reordered as a preprocessing step [#1408](#1408) ### Improvements + Significantly improve Cholesky factorization performance [#1366](#1366) + Improve parallel build performance [#1378](#1378) + Allow constrained parallel test execution using CTest resources [#1373](#1373) + Use arithmetic type more inside mixed precision ELL [#1414](#1414) + Most factory parameters of factory type no longer need to be constructed explicitly via `.on(exec)` [#1336](#1336) [#1439](#1439) + Improve ParILU(T)/ParIC(T) convergence by using more appropriate atomic operations [#1434](#1434) ### Fixes + Fix an over-allocation for OpenMP reductions [#1369](#1369) + Fix DPCPP's common-kernel reduction for empty input sizes [#1362](#1362) + Fix several typos in the API and documentation [#1348](#1348) + Fix inconsistent `Threads` between generations [#1388](#1388) + Fix benchmark median condition [#1398](#1398) + Fix HIP 5.6.0 compilation [#1411](#1411) + Fix missing destruction of rand_generator from cuda/hip [#1417](#1417) + Fix PAPI logger destruction order [#1419](#1419) + Fix TAU logger compilation [#1422](#1422) + Fix relative criterion to not iterate if the residual is already zero [#1079](#1079) + Fix memory_order invocations with C++20 changes [#1402](#1402) + Fix `check_diagonal_entries_exist` report correctly when only missing diagonal value in the last rows. [#1440](#1440) + Fix checking OpenMPI version in cross-compilation settings [#1446](#1446) + Fix false-positive deprecation warnings in Ginkgo, especially for the old Rcm (it doesn't emit deprecation warnings anymore as a result but is still considered deprecated) [#1444](#1444) ### Related PR: #1454
This PR adds the
BatchMultiVector
class, which implements the batched multi-vector object. Following our discussion, this is now separate from theLinOp
/BatchLinOp
hierarchy. In addition, the following things are also added in this PR:batch_dim<>
functionality which is a simple wrapper overdim<>
for storing uniform batched objects.batch_initialize
functions used to initialize the all matrix formats in a fashion similar toDense<>::initialize(...)
.There are no apply kernels as this is only a Multi-vector object and hence does not lend itself to the
apply
functionality.To be general enough, a MultiVector is supported even though the solvers in
batch_develop
currently only support single vector.This PR is first in the series of the functionality that brings batched functionality to Ginkgo develop and part of the functionality has been in
batch-develop
.