Skip to main content

IRIS-BLAS: Towards a Performance Portable and Heterogeneous BLAS Library

by Narasinga Rao Miniskar, Mohammad Alaul Haque Monil, Pedro Valero Lara, Frank Y Liu, Jeffrey S Vetter
Publication Type
Conference Paper
Book Title
2022 IEEE 29th International Conference on High Performance Computing, Data, and Analytics (HiPC)
Publication Date
Page Numbers
1 to 1
Publisher Location
New Jersey, United States of America
Conference Name
IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC)
Conference Location
Bengaluru, India
Conference Sponsor
IEEE Computer Society
Conference Date

This paper presents IRIS-BLAS, a novel heterogeneous and performance portable BLAS library. IRIS-BLAS is built on top of the IRIS runtime and multiple vendor and open-source BLAS libraries. It can transparently use all the architectures/devices available in a heterogeneous system, using the appropriate BLAS library based on the task mapping at run time. Thus, IRIS-BLAS is portable across a broad spectrum of architectures and BLAS libraries, alleviating the worry of application developers about modifying the application source code. Even though the emphasis is on portability, IRIS-BLAS provides competitive or even better performance than other state-of-the-art references. Moreover, IRIS-BLAS offers new features such as efficiently using extremely heterogeneous systems composed of multiple GPUs from different hardware vendors.