Presentation | 2014-04-25 Accelerating breadth-first search using Tightly Coupled Accelerator Yuki KATSUTA, Takaaki MIYAJIMA, KUHARA Takuya /, Toshihiro HANAWA, Hideharu AMANO, Taisuke BOKU, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Recent rapid increase of GPU performance makes GPGPU (General Purpose Computation using on GPUs) a mainstream of high performance computing. For a large scale target, multi-GPU systems in which a node with a few GPUs are connected through each host CPU have become popular. However, such systems suffer a large latency to communicate between GPUs attached to different hosts. In order to cope with this problem, Center of Computational Sciences, University of Tsukuba has developed a tightly couple accelerators (TCA) architecture which connects a large number of GPUs directly with dedicated switches PEACH2 (PCI-Express Adaptive Communication Hub 2). PEACH2 connects a host directly with PCI express, and forms a 2D mesh network by connecting neighboring PEACH2 switches directly also with PCI express. In this report, breadth first search (BFS) algorithm is implemented on a multi-GPU system connected with PEACH2. The preliminary evaluation results show that by using PEACH2, almost twice performance can be achieved by using 2 GPUs. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | GPU / TCA / PEACH2 / Graph500 |
Paper # | CPSY2014-4,DC2014-4 |
Date of Issue |
Conference Information | |
Committee | DC |
---|---|
Conference Date | 2014/4/18(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Dependable Computing (DC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Accelerating breadth-first search using Tightly Coupled Accelerator |
Sub Title (in English) | |
Keyword(1) | GPU |
Keyword(2) | TCA |
Keyword(3) | PEACH2 |
Keyword(4) | Graph500 |
1st Author's Name | Yuki KATSUTA |
1st Author's Affiliation | Graduate School of Science and Technology, Keio University() |
2nd Author's Name | Takaaki MIYAJIMA |
2nd Author's Affiliation | Graduate School of Science and Technology, Keio University |
3rd Author's Name | KUHARA Takuya / |
3rd Author's Affiliation | Graduate School of Science and Technology, Keio University |
4th Author's Name | Toshihiro HANAWA |
4th Author's Affiliation | Graduate School of Science and Technology, Keio University |
5th Author's Name | Hideharu AMANO |
5th Author's Affiliation | Supercomputing Division, Information Technology Center, The University of Tokyo |
6th Author's Name | Taisuke BOKU |
6th Author's Affiliation | Graduate School of Science and Technology, Keio University |
Date | 2014-04-25 |
Paper # | CPSY2014-4,DC2014-4 |
Volume (vol) | vol.114 |
Number (no) | 22 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |