Variable-size batched Gauss-Jordan elimination for block-Jacobi preconditioning on graphics processors
[EN] In this work, we address the efficient realization of block-Jacobi preconditioning on graphics processing units (GPUs). This task requires the solution of a collection of small and independent linear systems. To fully realize this implementation, we develop a variablesize batched matrix inversi...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2019 |
| País: | España |
| Institución: | Universitat Politècnica de València (UPV) |
| Repositorio: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia |
| Idioma: | inglés |
| OAI Identifier: | oai:riunet.upv.es:10251/158177 |
| Acceso en línea: | https://riunet.upv.es/handle/10251/158177 |
| Access Level: | acceso abierto |
| Palabra clave: | Batched algorithms Matrix inversion Gauss-Jordan elimination Block-Jacobi Sparse linear systems Graphics processor ARQUITECTURA Y TECNOLOGIA DE COMPUTADORES |
| Sumario: | [EN] In this work, we address the efficient realization of block-Jacobi preconditioning on graphics processing units (GPUs). This task requires the solution of a collection of small and independent linear systems. To fully realize this implementation, we develop a variablesize batched matrix inversion kernel that uses Gauss-Jordan elimination (GJE) along with a variable-size batched matrix-vector multiplication kernel that transforms the linear systems' right-hand sides into the solution vectors. Our kernels make heavy use of the increased register count and the warp-local communication associated with newer GPU architectures. Moreover, in the matrix inversion, we employ an implicit pivoting strategy that migrates the workload (i.e., operations) to the place where the data resides instead of moving the data to the executing cores. We complement the matrix inversion with extraction and insertion strategies that allow the block-Jacobi preconditioner to be set up rapidly. The experiments on NVlDlA's K40 and P100 architectures reveal that our variable-size batched matrix inversion routine outperforms the CUDA basic linear algebra subroutine (cuBLAS) library functions that provide the same (or even less) functionality. We also show that the preconditioner setup and preconditioner application cost can be somewhat offset by the faster convergence of the iterative solver. (C) 2018 Elsevier B.V. All rights reserved. |
|---|