Task-based, GPU-accelerated and Robust Library for Solving Dense Nonsymmetric Eigenvalue Problems
Department of Computing Science and HPC2N, Umeå University, SE-901 87 Umeå, Sweden
arXiv:2002.05024 [cs.MS], (12 Feb 2020)
@misc{myllykoski2020taskbased,
title={Task-based, GPU-accelerated and Robust Library for Solving Dense Nonsymmetric Eigenvalue Problems},
author={Mirko Myllykoski and Carl Christian Kjelgaard Mikkelsen},
year={2020},
eprint={2002.05024},
archivePrefix={arXiv},
primaryClass={cs.MS}
}
In this paper, we present the StarNEig library for solving dense nonsymmetric standard and generalized eigenvalue problems. The library is built on top of the StarPU runtime system and targets both shared and distributed memory machines. Some components of the library have support for GPU acceleration. The library is currently in an early beta state and supports only real matrices. Support for complex matrices is planned for a future release. This paper is aimed at potential users of the library. We describe the design choices and capabilities of the library, and contrast them to existing software such as ScaLAPACK. StarNEig implements a ScaLAPACK compatibility layer which should assist new users in the transition to StarNEig. We demonstrate the performance of the library with a sample of computational experiments.
February 16, 2020 by hgpu
