Abstract
SummaryIn 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 currently applies to real matrices with real and complex eigenvalues and all calculations are done using real arithmetic. 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 LAPACK and 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.
Highlights
In this paper, we present the StarNEig library[1] for solving dense nonsymmetric standard and generalized eigenvalue problems
StarNEig differs from existing libraries such as LAPACK2 and ScaLAPACK3 by relying on a modern task-based approach in a manner similar to the already well-established PLASMA library.[5]
We do not attempt to explain the details of each algorithm, rather we focus on the key steps and explain how they benefit from task parallelism
Summary
We present the StarNEig library[1] for solving dense nonsymmetric standard and generalized eigenvalue problems. StarNEig differs from existing libraries such as LAPACK2 and ScaLAPACK3 by relying on a modern task-based approach (see, eg, Reference 4) in a manner similar to the already well-established PLASMA library.[5] StarNEig is built on top of the StarPU runtime system[6] developed by the StarPU team at INRIA This allows StarNEig to target both shared memory and distributed memory machines. StarNEig can compute the eigenvectors directly from any real Schur form without suffering from floating-point overflow, that is, the implementation is robust The latter functionality does not exist in ScaLAPACK and the implementation in StarNEig is significantly faster than the LAPACK implementation in both sequential and parallel settings (parallel BLAS). The major contribution of this paper is the extended description of the library and its capabilities/limitations
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Concurrency and Computation: Practice and Experience
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
