Slate

Gates, Mark; Kurzak, Jakub; Charara, Ali; YarKhan, Asim; Dongarra, Jack

doi:10.1145/3295500.3356223

Cited by 50 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SLATE (Software for Linear Algebra Targeting Exascale) 13 library is being developed to provide software for dense numerical linear algebra on current and future distributed computer systems. Currently, SLATE implements a set of parallel basic linear algebra subroutines (parallel BLAS), as well as high‐level subroutines for solving linear systems and linear least square problems.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Task‐based, GPU‐accelerated and robust library for solving dense nonsymmetric eigenvalue problems

Myllykoski

Mikkelsen

2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary 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 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.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Currently, SLATE implements a set of parallel basic linear algebra subroutines (parallel BLAS), as well as high‐level subroutines for solving linear systems and linear least square problems. The authors state that future work will include nonsymmetric eigenvalue problems 13 …”

Section: Related Workmentioning

confidence: 99%

Task‐based, GPU‐accelerated and robust library for solving dense nonsymmetric eigenvalue problems

Myllykoski

Mikkelsen

2020

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…MAGMA Templates uses the SLATE library (Gates et al, 2019) to provide dense linear algebra kernels for distributedmemory heterogeneous architectures (Abdelfattah et al, 2017;Kurzak et al, 2017Kurzak et al, , 2019a. Note that Figure 1 shows that Trilinos (Heroux et al, 2005) is the backend for the distributed-memory sparse linear algebra kernels on which MAGMA Templates depends.…”

Section: Magma Templates Software Designmentioning

confidence: 99%

“…to include support for distributed-memory systems, we designed MAGMA Templates to be a high-level thin layer set on top of multiple numerical kernels/libraries. We therefore include support for Software for Linear Algebra Targeting Exascale (SLATE) (Gates et al, 2019), Trilinos/PETSc/HYPRE, and vendor-optimized math libraries.…”

Section: Introductionmentioning

confidence: 99%

MAGMA templates for scalable linear algebra on emerging architectures

Farhan

Abdelfattah

Tomov

et al. 2020

The International Journal of High Performance Computing Applica

Self Cite

View full text Add to dashboard Cite

With the acquisition and widespread use of more resources that rely on accelerator/wide vector–based computing, there has been a strong demand for science and engineering applications to take advantage of these latest assets. This, however, has been extremely challenging due to the diversity of systems to support their extreme concurrency, complex memory hierarchies, costly data movement, and heterogeneous node architectures. To address these challenges, we design a programming model and describe its ease of use in the development of a new MAGMA Templates library that delivers high-performance scalable linear algebra portable on current and emerging architectures. MAGMA Templates derives its performance and portability by (1) building on existing state-of-the-art linear algebra libraries, like MAGMA, SLATE, Trilinos, and vendor-optimized math libraries, and (2) providing access (seamlessly to the users) to the latest algorithms and architecture-specific optimizations through a single, easy-to-use C++-based API.

show abstract

“…However, in contrast to SM-SISUBIT, there does not exist a purely GPU distributed memory linear algebra library as of this work. Currently, the stateof-the-art for GPU accelerated distributed memory dense linear algebra is the SLATE library [13]. As a hybrid GPU/CPU library, SLATE utilizes both vendor optimized CPU and GPU accelerated implementations of BLAS/LAPACK primitives to achieve its performance.…”

Section: Sisubit Implementationmentioning

confidence: 99%

Parallel Shift-Invert Spectrum Slicing on Distributed Architectures with GPU Accelerators

Williams‐Young

Yang

2020

49th International Conference on Parallel Processing - ICPP

View full text Add to dashboard Cite

The solution of large scale eigenvalue problems (EVP) is often the computational bottleneck for many scientific and engineering applications. Traditional eigensolvers, such as direct (e.g. ScaLAPACK) and Krylov subspace (e.g. Lanczos) methods, have struggled in achieving high scalability on large computing resources due to communication and synchronization bottlenecks which are inherent in their implementation. This includes a difficulty in developing well-performing ports of these algorithms to architectures which rely on the use of accelerators, such as graphics processing units (GPU), for the majority of their floating point operations. Recently, there has been significant research into the development of eigensolvers based on spectrum slicing, in particular shift-invert spectrum slicing, to alleviate the communication and synchronization bottlenecks of traditional eigensolvers. In general, spectrum slicing trades the global EVP for many smaller, independent EVPs which may be combined to assemble some desired subset of the entire eigenspectrum. The result is a method which utilizes more floating point operations than traditional eigensolvers, but in a way which allows for the expression of massive concurrency leading to an overall improvement in time-to-solution on large computing resources. In this work, we will examine the performance of parallel shift-invert spectrum slicing on modern GPU clusters using state-of-the-art linear algebra software.

show abstract

Slate

Cited by 50 publications

References 33 publications

Task‐based, GPU‐accelerated and robust library for solving dense nonsymmetric eigenvalue problems

Task‐based, GPU‐accelerated and robust library for solving dense nonsymmetric eigenvalue problems

MAGMA templates for scalable linear algebra on emerging architectures

Parallel Shift-Invert Spectrum Slicing on Distributed Architectures with GPU Accelerators

Contact Info

Product

Resources

About