17714

Parallel Computing for the Inverse of SPD matrix

Yan Shen, Yuxing Dai, Zhiliang Zhu
Hunan University, ChangSha, P.R.China
Hunan University, 2017

@article{shen2017parallel,

   title={Parallel Computing for the Inverse of SPD matrix},

   author={Shen, Yan and Dai, Yuxing and Zhu, Zhiliang},

   year={2017}

}

In this paper, we propose a High performance Parallel Computing method for the Inverse of a symmetric positive definite (SPD) matrix. Brought in the reuse of the inverse of diagonal sub blocks technique and Combined with the newest OpenCL parallel computing framework, this methods can improve computing the inverse of SPD matrix effectively. Computing the inverse of SPD matrix can be broken down into three steps: Cholesky decomposition of SPD matrix, computing the inverse of lower triangular matrix, matrix product of triangular matrix and its transpose. Cholesky decomposition in this method can also be modified to correct a matrix which is not positive definite. By making each step parallel, we get a better performance than the current clMAGMA packages about 10x times. Our implementation has cross-platform features and can be used by CPU and GPU without modification.
No votes yet.
Please wait...

* * *

* * *

Featured events

2018
November
27-30
Hida Takayama, Japan

The Third International Workshop on GPU Computing and AI (GCA), 2018

2018
September
19-21
Nagoya University, Japan

The 5th International Conference on Power and Energy Systems Engineering (CPESE), 2018

2018
September
22-24
MediaCityUK, Salford Quays, Greater Manchester, England

The 10th International Conference on Information Management and Engineering (ICIME), 2018

2018
August
21-23
No. 1037, Luoyu Road, Hongshan District, Wuhan, China

The 4th International Conference on Control Science and Systems Engineering (ICCSSE), 2018

2018
October
29-31
Nanyang Executive Centre in Nanyang Technological University, Singapore

The 2018 International Conference on Cloud Computing and Internet of Things (CCIOT’18), 2018

HGPU group © 2010-2018 hgpu.org

All rights belong to the respective authors

Contact us: