This title appears in the Scientific Report :
2022
Please use the identifier:
http://hdl.handle.net/2128/31597 in citations.
Please use the identifier: http://dx.doi.org/10.1145/3539781.3539792 in citations.
ChASE - A Distributed Hybrid CPU-GPU Eigensolver for Large-scale Hermitian Eigenvalue Problems
ChASE - A Distributed Hybrid CPU-GPU Eigensolver for Large-scale Hermitian Eigenvalue Problems
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. Thi...
Saved in:
Personal Name(s): | Wu, Xinzhe |
---|---|
Davidovic, Davor / Achilles, Sebastian / Di Napoli, Edoardo (Corresponding author) | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Published in: | S. Article No.: 9 |
Published in: |
Proceedings of the Platform for Advanced Scientific Computing Conference - ACM New York, NY, USA, 2022. - ISBN 9781450394109 - doi:10.1145/3539781.3539792 |
Imprint: |
ACM New York, NY, USA
2022
|
Physical Description: |
12 pages |
DOI: |
10.1145/3539781.3539792 |
Conference: | Platform for Advanced Scientific Computing, Basel (Switzerland), 2022-06-27 - 2022-06-29 |
Document Type: |
Contribution to a book Contribution to a conference proceedings |
Research Program: |
Simulation and Data Laboratory Quantum Materials (SDLQM) PRACE 6th Implementation Phase Project Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups |
Link: |
OpenAccess |
Publikationsportal JuSER |
Please use the identifier: http://dx.doi.org/10.1145/3539781.3539792 in citations.
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 communication 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. Our 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 $360$k. |