hrvatski jezikClear Cookie - decide language by browser settings

ChASE: a distributed hybrid CPU-GPU eigensolver for large-scale hermitian eigenvalue problems

Wu, Xinzhe; Davidović, Davor; Achilles, Sebastian; Di Napoli, Edoardo (2022) ChASE: a distributed hybrid CPU-GPU eigensolver for large-scale hermitian eigenvalue problems. In: PASC'22: Proceedings of the Platform for Advanced Scientific Computing Conference. New York, NY, USA, ACM, pp. 1-12 .

[img]
Preview
PDF - Published Version - article
Download (930kB) | Preview

Abstract

As modern massively parallel clusters are getting larger with beefier compute nodes, traditional parallel eigensolvers, such as direct solvers, struggle keeping the pace with the hardware evolution and being able to scale efficiently due to additional layers of communication and synchronization. This difficulty is especially important when porting traditional libraries to heterogeneous computing architectures equipped with accelerators, such as Graphics Processing Unit (GPU). Recently, there have been significant scientific contributions to the development of filter-based subspace eigensolver to compute partial eigenspectrum. The simpler structure of these type of algorithms makes for them easier to avoid the communica tion and synchronization bottlenecks typical of direct solvers. The Chebyshev Accelerated Subspace Eigensolver (ChASE) is a modern subspace eigensolver to compute partial extremal eigenpairs of large-scale Hermitian eigenproblems with the acceleration of a filter based on Chebyshev polynomials. In this work, we extend our previous work on ChASE by adding support for distributed hybrid CPU-multi-GPU computing architectures. Out tests show that ChASE achieves very good scaling performance up to 144 nodes with 526 NVIDIA A100 GPUs in total on dense eigenproblems of size up to 360k.

Item Type: Conference or workshop item published in conference proceedings (UNSPECIFIED)
Uncontrolled Keywords: Subspace iteration eigensolve ; Dense Hermitian matrix ; Chebyshev polynomial ; Distributed hybrid CPU-GPU ; Heterogeneous GPU supercomputers
Subjects: NATURAL SCIENCES > Mathematics
NATURAL SCIENCES > Mathematics > Numerical Mathematics
TECHNICAL SCIENCES > Computing > Process Computing
Divisions: Center for Informatics and Computing
Projects:
Project titleProject leaderProject codeProject type
Skalabilni algoritmi visokih performansi za buduće heterogene distribuirane računalne sustaveDavidović, DavorUIP-2020-02-4559HRZZ
Depositing User: Davor Davidović
Date Deposited: 18 Aug 2022 09:21
URI: http://fulir.irb.hr/id/eprint/7504
DOI: 10.1145/3539781.3539792

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

Contrast
Increase Font
Decrease Font
Dyslexic Font
Accessibility