作成者 |
|
|
|
|
本文言語 |
|
出版者 |
|
発行日 |
|
収録物名 |
|
|
巻 |
|
会議情報 |
|
出版タイプ |
|
アクセス権 |
|
概要 |
To achieve high-performance computation on recent supercomputer architecture with numerous cores, it is required to incorporate various parallelism into parallel simulation to break performance satura...tion employing only domain decomposition approach. Parallel-in-Time (PinT) is one of the most promising candidate for such issues. In this paper, the authors demonstrate that Parareal method is simple but still effective implementation of PinT and its Pipelined version shows higher performance. Performance evaluation was conducted on the K Computer with upto 8,000 nodes and found that Pipelined Parareal method with domain decomposition method achieved 222 times faster than the its sequential time-integration.続きを見る
|