From 54ef7bad8ae01403577497e1e34200155d4d8cc6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tom=C3=A1=C5=A1=20Oberhuber?= <oberhuber.tomas@gmail.com>
Date: Thu, 4 Feb 2021 15:17:05 +0100
Subject: [PATCH] Added symmetric sparse matrices to SpMV benchmark.
---
src/Benchmarks/SpMV/{spmv-legacy.h => spmv.h} | 216 +++++++++++++++---
src/Benchmarks/SpMV/tnl-benchmark-spmv.h | 2 +-
src/TNL/Matrices/MatrixInfo.h | 8 +-
3 files changed, 193 insertions(+), 33 deletions(-)
rename src/Benchmarks/SpMV/{spmv-legacy.h => spmv.h} (65%)
diff --git a/src/Benchmarks/SpMV/spmv-legacy.h b/src/Benchmarks/SpMV/spmv.h
similarity index 65%
rename from src/Benchmarks/SpMV/spmv-legacy.h
rename to src/Benchmarks/SpMV/spmv.h
index 7c7e19d807..b010303672 100644
--- a/src/Benchmarks/SpMV/spmv-legacy.h
+++ b/src/Benchmarks/SpMV/spmv.h
@@ -44,11 +44,9 @@ namespace TNL {
namespace Benchmarks {
namespace SpMVLegacy {
-// Alias to match the number of template parameters with other formats
-template< typename Real, typename Device, typename Index >
-using SlicedEllpackAlias = Benchmarks::SpMV::ReferenceFormats::Legacy::SlicedEllpack< Real, Device, Index >;
-
-// Segments based sparse matrix aliases
+/////
+// General sparse matrix aliases
+//
template< typename Real, typename Device, typename Index >
using SparseMatrix_CSR_Scalar = Matrices::SparseMatrix< Real, Device, Index, Matrices::GeneralMatrix, Algorithms::Segments::CSRScalar >;
@@ -85,7 +83,49 @@ using BiEllpackSegments = Algorithms::Segments::BiEllpack< Device, Index, IndexA
template< typename Real, typename Device, typename Index >
using SparseMatrix_BiEllpack = Matrices::SparseMatrix< Real, Device, Index, Matrices::GeneralMatrix, BiEllpackSegments >;
+/////
+// Symmetric sparse matrix aliases
+//
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_CSR_Scalar = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, Algorithms::Segments::CSRScalar >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_CSR_Vector = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, Algorithms::Segments::CSRVector >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_CSR_Hybrid = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, Algorithms::Segments::CSRHybrid >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_CSR_Adaptive = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, Algorithms::Segments::CSRAdaptive >;
+
+template< typename Device, typename Index, typename IndexAllocator >
+using EllpackSegments = Algorithms::Segments::Ellpack< Device, Index, IndexAllocator >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_Ellpack = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, EllpackSegments >;
+
+template< typename Device, typename Index, typename IndexAllocator >
+using SlicedEllpackSegments = Algorithms::Segments::SlicedEllpack< Device, Index, IndexAllocator >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_SlicedEllpack = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, SlicedEllpackSegments >;
+
+template< typename Device, typename Index, typename IndexAllocator >
+using ChunkedEllpackSegments = Algorithms::Segments::ChunkedEllpack< Device, Index, IndexAllocator >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_ChunkedEllpack = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, ChunkedEllpackSegments >;
+
+template< typename Device, typename Index, typename IndexAllocator >
+using BiEllpackSegments = Algorithms::Segments::BiEllpack< Device, Index, IndexAllocator >;
+
+template< typename Real, typename Device, typename Index >
+using SymmetricSparseMatrix_BiEllpack = Matrices::SparseMatrix< Real, Device, Index, Matrices::SymmetricMatrix, BiEllpackSegments >;
+
+
+/////
// Legacy formats
+//
template< typename Real, typename Device, typename Index >
using SparseMatrixLegacy_CSR_Scalar = Benchmarks::SpMV::ReferenceFormats::Legacy::CSR< Real, Device, Index, Benchmarks::SpMV::ReferenceFormats::Legacy::CSRScalar >;
@@ -119,6 +159,9 @@ using SparseMatrixLegacy_CSR_MultiVector = Benchmarks::SpMV::ReferenceFormats::L
template< typename Real, typename Device, typename Index >
using SparseMatrixLegacy_CSR_LightWithoutAtomic = Benchmarks::SpMV::ReferenceFormats::Legacy::CSR< Real, Device, Index, Benchmarks::SpMV::ReferenceFormats::Legacy::CSRLightWithoutAtomic >;
+template< typename Real, typename Device, typename Index >
+using SlicedEllpackAlias = Benchmarks::SpMV::ReferenceFormats::Legacy::SlicedEllpack< Real, Device, Index >;
+
// Get the name (with extension) of input matrix file
std::string getMatrixFileName( const String& InputFileName )
{
@@ -186,7 +229,6 @@ benchmarkSpMVLegacy( Benchmark& benchmark,
benchmark.setMetadataColumns( Benchmark::MetadataColumns({
{ "matrix name", convertToString( inputFileName ) },
- //{ "non-zeros", convertToString( hostMatrix.getNonzeroElementsCount() ) },
{ "rows", convertToString( hostMatrix.getRows() ) },
{ "columns", convertToString( hostMatrix.getColumns() ) },
{ "matrix format", MatrixInfo< HostMatrix >::getFormat() }
@@ -195,9 +237,9 @@ benchmarkSpMVLegacy( Benchmark& benchmark,
const double datasetSize = (double) elements * ( 2 * sizeof( Real ) + sizeof( int ) ) / oneGB;
benchmark.setOperation( datasetSize );
- /***
- * Benchmark SpMV on host
- */
+ /////
+ // Benchmark SpMV on host
+ //
HostVector hostInVector( hostMatrix.getColumns() ), hostOutVector( hostMatrix.getRows() );
auto resetHostVectors = [&]() {
@@ -212,9 +254,9 @@ benchmarkSpMVLegacy( Benchmark& benchmark,
SpmvBenchmarkResult< Real, Devices::Host, int > hostBenchmarkResults( csrResultVector, hostOutVector, hostMatrix.getNonzeroElementsCount() );
benchmark.time< Devices::Host >( resetHostVectors, "CPU", spmvHost, hostBenchmarkResults );
- /***
- * Benchmark SpMV on CUDA
- */
+ /////
+ // Benchmark SpMV on CUDA
+ //
#ifdef HAVE_CUDA
cudaMatrix = hostMatrix;
CudaVector cudaInVector( hostMatrix.getColumns() ), cudaOutVector( hostMatrix.getRows() );
@@ -233,6 +275,82 @@ benchmarkSpMVLegacy( Benchmark& benchmark,
std::cout << std::endl;
}
+template< typename Real,
+ typename InputMatrix,
+ template< typename, typename, typename > class Matrix,
+ template< typename, typename, typename, typename > class Vector = Containers::Vector >
+void
+benchmarkSpMV( Benchmark& benchmark,
+ const InputMatrix& inputMatrix,
+ const TNL::Containers::Vector< Real, Devices::Host, int >& csrResultVector,
+ const String& inputFileName,
+ bool verboseMR )
+{
+ using HostMatrix = Matrix< Real, TNL::Devices::Host, int >;
+ using CudaMatrix = Matrix< Real, TNL::Devices::Cuda, int >;
+ using HostVector = Containers::Vector< Real, Devices::Host, int >;
+ using CudaVector = Containers::Vector< Real, Devices::Cuda, int >;
+
+ HostMatrix hostMatrix;
+ try
+ {
+ hostMatrix = inputMatrix;
+ }
+ catch(const std::exception& e)
+ {
+ std::cerr << "Unable to convert the matrix to the target format." << std::endl;
+ return;
+ }
+
+ benchmark.setMetadataColumns( Benchmark::MetadataColumns({
+ { "matrix name", convertToString( inputFileName ) },
+ { "rows", convertToString( hostMatrix.getRows() ) },
+ { "columns", convertToString( hostMatrix.getColumns() ) },
+ { "matrix format", MatrixInfo< HostMatrix >::getFormat() }
+ } ));
+ const int elements = hostMatrix.getNonzeroElementsCount();
+ const double datasetSize = (double) elements * ( 2 * sizeof( Real ) + sizeof( int ) ) / oneGB;
+ benchmark.setOperation( datasetSize );
+
+ /////
+ // Benchmark SpMV on host
+ //
+ HostVector hostInVector( hostMatrix.getColumns() ), hostOutVector( hostMatrix.getRows() );
+
+ auto resetHostVectors = [&]() {
+ hostInVector = 1.0;
+ hostOutVector = 0.0;
+ };
+
+ auto spmvHost = [&]() {
+ hostMatrix.vectorProduct( hostInVector, hostOutVector );
+
+ };
+ SpmvBenchmarkResult< Real, Devices::Host, int > hostBenchmarkResults( csrResultVector, hostOutVector, hostMatrix.getNonzeroElementsCount() );
+ benchmark.time< Devices::Host >( resetHostVectors, "CPU", spmvHost, hostBenchmarkResults );
+
+ /////
+ // Benchmark SpMV on CUDA
+ //
+#ifdef HAVE_CUDA
+ CudaMatrix cudaMatrix;
+ cudaMatrix = inputMatrix;
+ CudaVector cudaInVector( hostMatrix.getColumns() ), cudaOutVector( hostMatrix.getRows() );
+
+ auto resetCudaVectors = [&]() {
+ cudaInVector = 1.0;
+ cudaOutVector = 0.0;
+ };
+
+ auto spmvCuda = [&]() {
+ cudaMatrix.vectorProduct( cudaInVector, cudaOutVector );
+ };
+ SpmvBenchmarkResult< Real, Devices::Cuda, int > cudaBenchmarkResults( csrResultVector, cudaOutVector, cudaMatrix.getNonzeroElementsCount() );
+ benchmark.time< Devices::Cuda >( resetCudaVectors, "GPU", spmvCuda, cudaBenchmarkResults );
+ #endif
+ std::cout << std::endl;
+}
+
template< typename Real = double,
typename Index = int >
void
@@ -247,7 +365,7 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
// It seems that there is a problem with lambda functions identification when we create
// two instances of TNL::Matrices::SparseMatrix. The second one comes from calling of
// `benchmarkSpMV< Real, SparseMatrix_CSR_Scalar >( benchmark, hostOutVector, inputFileName, verboseMR );`
- // and simillar later in this function. Maybe splitting this function into two might help.
+ // and simillar later in this function.
#define USE_LEGACY_FORMATS
#ifdef USE_LEGACY_FORMATS
// Here we use 'int' instead of 'Index' because of compatibility with cusparse.
@@ -266,7 +384,6 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
using CudaVector = Containers::Vector< Real, Devices::Cuda, int >;
CSRHostMatrix csrHostMatrix;
- CSRCudaMatrix csrCudaMatrix;
////
// Set-up benchmark datasize
@@ -281,7 +398,6 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
//
benchmark.setMetadataColumns( Benchmark::MetadataColumns({
{ "matrix name", convertToString( inputFileName ) },
- //{ "non-zeros", convertToString( csrHostMatrix.getNumberOfNonzeroMatrixElements() ) },
{ "rows", convertToString( csrHostMatrix.getRows() ) },
{ "columns", convertToString( csrHostMatrix.getColumns() ) },
{ "matrix format", String( "CSR" ) }
@@ -307,7 +423,6 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
#ifdef HAVE_CUDA
benchmark.setMetadataColumns( Benchmark::MetadataColumns({
{ "matrix name", convertToString( inputFileName ) },
- //{ "non-zeros", convertToString( csrHostMatrix.getNumberOfNonzeroMatrixElements() ) },
{ "rows", convertToString( csrHostMatrix.getRows() ) },
{ "columns", convertToString( csrHostMatrix.getColumns() ) },
{ "matrix format", String( "cuSparse" ) }
@@ -316,6 +431,7 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
cusparseHandle_t cusparseHandle;
cusparseCreate( &cusparseHandle );
+ CSRCudaMatrix csrCudaMatrix;
csrCudaMatrix = csrHostMatrix;
// Delete the CSRhostMatrix, so it doesn't take up unnecessary space
@@ -337,25 +453,13 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
SpmvBenchmarkResult< Real, Devices::Host, int > cusparseBenchmarkResults( hostOutVector, hostOutVector, csrHostMatrix.getNonzeroElementsCount() );
benchmark.time< Devices::Cuda >( resetCusparseVectors, "GPU", spmvCusparse, cusparseBenchmarkResults );
+ csrCudaMatrix.reset();
#endif
-
- /////
- // Benchmarking TNL formats
- benchmarkSpMVLegacy< Real, SparseMatrix_CSR_Scalar >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_CSR_Vector >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_CSR_Hybrid >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_CSR_Adaptive >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_Ellpack >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SlicedEllpackAlias >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_SlicedEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_ChunkedEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
- benchmarkSpMVLegacy< Real, SparseMatrix_BiEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
-
-
- const bool withSymmetricMatrices = parameters.getParameter< bool >("with-symmetric-matrices");
+ csrHostMatrix.reset();
/////
// Benchmarking of TNL legacy formats
+ //
if( parameters.getParameter< bool >("with-legacy-matrices") )
{
using namespace Benchmarks::SpMV::ReferenceFormats;
@@ -371,12 +475,62 @@ benchmarkSpmvSynthetic( Benchmark& benchmark,
benchmarkSpMVLegacy< Real, SparseMatrixLegacy_CSR_MultiVector >( benchmark, hostOutVector, inputFileName, verboseMR );
benchmarkSpMVLegacy< Real, SparseMatrixLegacy_CSR_LightWithoutAtomic >( benchmark, hostOutVector, inputFileName, verboseMR );
benchmarkSpMVLegacy< Real, Legacy::Ellpack >( benchmark, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMVLegacy< Real, SlicedEllpackAlias >( benchmark, hostOutVector, inputFileName, verboseMR );
benchmarkSpMVLegacy< Real, Legacy::ChunkedEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
benchmarkSpMVLegacy< Real, Legacy::BiEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
}
/* AdEllpack is broken
benchmarkSpMV< Real, Matrices::AdEllpack >( benchmark, hostOutVector, inputFileName, verboseMR );
*/
+
+ /////
+ // Benchmarking TNL formats
+ //
+ using HostMatrixType = TNL::Matrices::SparseMatrix< Real, TNL::Devices::Host >;
+ HostMatrixType hostMatrix;
+ TNL::Matrices::MatrixReader< HostMatrixType >::readMtx( inputFileName, hostMatrix, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_CSR_Scalar >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_CSR_Vector >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_CSR_Hybrid >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_CSR_Adaptive >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_Ellpack >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_SlicedEllpack >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_ChunkedEllpack >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, HostMatrixType, SparseMatrix_BiEllpack >( benchmark, hostMatrix, hostOutVector, inputFileName, verboseMR );
+ hostMatrix.reset();
+
+ /////
+ // Benchmarking symmetric sparse matrices
+ //
+ if( parameters.getParameter< bool >("with-symmetric-matrices") )
+ {
+ using SymmetricInputMatrix = TNL::Matrices::SparseMatrix< Real, TNL::Devices::Host, int, TNL::Matrices::SymmetricMatrix >;
+ using InputMatrix = TNL::Matrices::SparseMatrix< Real, TNL::Devices::Host, int >;
+ SymmetricInputMatrix symmetricHostMatrix;
+ try
+ {
+ TNL::Matrices::MatrixReader< SymmetricInputMatrix >::readMtx( inputFileName, symmetricHostMatrix, verboseMR );
+ }
+ catch(const std::exception& e)
+ {
+ std::cerr << e.what() << " ... SKIPPING " << std::endl;
+ return;
+ }
+ InputMatrix hostMatrix;
+ TNL::Matrices::MatrixReader< InputMatrix >::readMtx( inputFileName, hostMatrix, verboseMR );
+ if( hostMatrix != symmetricHostMatrix )
+ {
+ std::cerr << "ERROR !!!!!! " << std::endl;
+ }
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_CSR_Scalar >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_CSR_Vector >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_CSR_Hybrid >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_CSR_Adaptive >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_Ellpack >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_SlicedEllpack >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_ChunkedEllpack >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ benchmarkSpMV< Real, SymmetricInputMatrix, SymmetricSparseMatrix_BiEllpack >( benchmark, symmetricHostMatrix, hostOutVector, inputFileName, verboseMR );
+ }
}
} // namespace SpMVLegacy
diff --git a/src/Benchmarks/SpMV/tnl-benchmark-spmv.h b/src/Benchmarks/SpMV/tnl-benchmark-spmv.h
index d4ec93934d..9a5005de73 100644
--- a/src/Benchmarks/SpMV/tnl-benchmark-spmv.h
+++ b/src/Benchmarks/SpMV/tnl-benchmark-spmv.h
@@ -18,7 +18,7 @@
#include <TNL/Devices/Cuda.h>
#include <TNL/Config/parseCommandLine.h>
-#include "spmv-legacy.h"
+#include "spmv.h"
#include <TNL/Matrices/MatrixReader.h>
using namespace TNL::Matrices;
diff --git a/src/TNL/Matrices/MatrixInfo.h b/src/TNL/Matrices/MatrixInfo.h
index 7d28956163..d84afa39a1 100644
--- a/src/TNL/Matrices/MatrixInfo.h
+++ b/src/TNL/Matrices/MatrixInfo.h
@@ -64,7 +64,13 @@ struct MatrixInfo< SparseMatrixView< Real, Device, Index, MatrixType, SegmentsVi
{
static String getDensity() { return String( "sparse" ); };
- static String getFormat() { return SegmentsView< Device, Index >::getSegmentsType(); };
+ static String getFormat()
+ {
+ if( MatrixType::isSymmetric() )
+ return TNL::String( "Symmetric " ) + SegmentsView< Device, Index >::getSegmentsType();
+ else
+ return SegmentsView< Device, Index >::getSegmentsType();
+ };
};
template< typename Real,
--
GitLab