Name: SparseLU
Contact Person: support-compss@bsc.es
Access Level: public
License Agreement: Apache2
Platform: COMPSs
Description
The Sparse LU application computes an LU matrix factorization on a sparse blocked matrix. The matrix size (number of blocks) and the block size are parameters of the application.
As the algorithm progresses, the area of the matrix that is accessed is smaller; concretely, at each iteration, the 0th row and column of the current matrix are discarded. On the other hand, due to the sparseness of the matrix, some of its blocks might not be allocated and, therefore, no work is generated for them.
When executed with COMPSs, Sparse LU produces several types of task with different granularity and numerous dependencies between them.
Versions
There are three versions of Sparse LU, depending on the data types used to store the blocks.
Version 1
''files'', where the matrix blocks are stored in files.
Version 2
''objects'', where the matrix blocks are represented by objects.
Version 3
''arrays'', where the matrix blocks are stored in arrays.
Execution instructions
Usage:
runcompss sparseLU.files.SparseLU numberOfBlocks blockSize
runcompss sparseLU.objects.SparseLU numberOfBlocks blockSize
runcompss sparseLU.arrays.SparseLU numberOfBlocks blockSize
where:
- numberOfBlocks: Number of blocks inside each matrix
- blockSize: Size of each block
Execution Example
runcompss sparseLU.objects.SparseLU 16 4
runcompss sparseLU.files.SparseLU 16 4
runcompss sparseLU.arrays.SparseLU 16 4
Build
Option 1: Native java
cd application_sources/; javac src/main/java/sparseLU/*/*.java
cd src/main/java/; jar cf sparseLU.jar sparseLU/
cd ../../../; mv src/main/java/sparseLU.jar jar/
Option 2: Maven
cd application_sources/
mvn clean package
Click and drag the diagram to pan, double click or use the controls to zoom.
Version History
Version 1 (earliest) Created 23rd Jun 2023 at 15:03 by Raül Sirvent
Frozen
Version-1
a378b09
Creator
Additional credit
The Workflows and Distributed Computing Team (https://www.bsc.es/discover-bsc/organisation/scientific-structure/workflows-and-distributed-computing)
Submitter
Views: 1549 Downloads: 182
Created: 23rd Jun 2023 at 15:03
Last updated: 30th Oct 2023 at 14:29
None