8545

Scalable Multi-GPU 3-D FFT for TSUBAME 2.0 Supercomputer

Akira Nukada, Kento Sato, Satoshi Matsuoka
Tokyo Institute of Technology
International Conference on High Performance Computing, Networking, Storage and Analysis (SC ’12), 2012
@inproceedings{nukada2012scalable,

   title={Scalable Multi-GPU 3-D FFT for TSUBAME 2.0 Supercomputer},

   author={Nukada, A. and Sato, K. and Matsuoka, S.},

   booktitle={Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis},

   pages={44},

   year={2012},

   organization={IEEE Computer Society Press}

}

Download Download (PDF)   View View   Source Source   

674

views

For scalable 3-D FFT computation using multiple GPUs, efficient all-to-all communication between GPUs is the most important factor in good performance. Implementations with point-to-point MPI library functions and CUDA memory copy APIs typically exhibit very large overheads especially for small message sizes in all-to-all communications between many nodes. We propose several schemes to minimize the overheads, including employment of lower-level API of InfiniBand to effectively overlap intra- and inter-node communication, as well as auto-tuning strategies to control scheduling and determine rail assignments. As a result we achieve very good strong scalability as well as good performance, up to 4.8TFLOPS using 256 nodes of TSUBAME 2.0 Supercomputer (768 GPUs) in double precision.
VN:F [1.9.22_1171]
Rating: 0.0/5 (0 votes cast)

* * *

* * *

Follow us on Twitter

HGPU group

1946 peoples are following HGPU @twitter

Like us on Facebook

HGPU group

443 people like HGPU on Facebook

HGPU group © 2010-2016 hgpu.org

All rights belong to the respective authors

Contact us: