Inicio  /  Applied Sciences  /  Vol: 13 Par: 15 (2023)  /  Artículo
ARTÍCULO
TITULO

Improving Structured Grid-Based Sparse Matrix-Vector Multiplication and Gauss?Seidel Iteration on GPDSP

Yang Wang    
Jie Liu    
Xiaoxiong Zhu    
Qingyang Zhang    
Shengguo Li and Qinglin Wang    

Resumen

Structured grid-based sparse matrix-vector multiplication and Gauss?Seidel iterations are very important kernel functions in scientific and engineering computations, both of which are memory intensive and bandwidth-limited. GPDSP is a general purpose digital signal processor, which is a very significant embedded processor that has been introduced into high-performance computing. In this paper, we designed various optimization methods, which included a blocking method to improve data locality and increase memory access efficiency, a multicolor reordering method to develop Gauss?Seidel fine-grained parallelism, a data partitioning method designed for GPDSP memory structures, and a double buffering method to overlap computation and access memory on structured grid-based SpMV and Gauss?Seidel iterations for GPDSP. At last, we combined the above optimization methods to design a multicore vectorization algorithm. We tested the matrices generated with structured grids of different sizes on the GPDSP platform and obtained speedups of up to 41× and 47× compared to the unoptimized SpMV and Gauss?Seidel iterations, with maximum bandwidth efficiencies of 72% and 81%, respectively. The experiment results show that our algorithms could fully utilize the external memory bandwidth. We also implemented the commonly used mixed precision algorithm on the GPDSP and obtained speedups of 1.60× and 1.45× for the SpMV and Gauss?Seidel iterations, respectively.

 Artículos similares

       
 
Muhammad Saeed, Renzon Daniel Cosme Pecho, Sandeep Panchal, Sadeq K. Alhag, Laila A. Al-Shuraym, Khalid M. Al Syaad and Usman Hanif Bhutta    
The contamination of water with organic pollutants, such as dyes, has become a serious threat to the environment. Therefore, the development of a cost-effective, eco-friendly, proficient, and visible-light-driven catalyst for the treatment of organic dye... ver más
Revista: Water

 
Francesca Sasanelli, Khang Duy Ricky Le, Samuel Boon Ping Tay, Phong Tran and Johan W. Verjans    
The advent of many popular commercial forms of natural language processing tools has changed the way we can utilise digital technologies to tackle problems with big data. The objective of this review is to evaluate the current research and landscape of n... ver más
Revista: Applied Sciences

 
Shindume Lomboleni Hamukwaya, Huiying Hao, Zengying Zhao, Jingjing Dong, Tingting Zhong, Jie Xing, Liu Hao and Melvin Mununuri Mashingaidze    
The recent rapid development in perovskite solar cells (PSCs) has led to significant research interest due to their notable photovoltaic performance, currently exceeding 25% power conversion efficiency for small-area PSCs. The materials used to fabricate... ver más
Revista: Coatings

 
Malindu Sandanayake, Yanni Bouras and Zora Vrcelj    
Infrastructure advancement is a key attribute that defines the development and effective growth of a city or region. Since the introduction of the United Nations Sustainability Development Goals (UN SDGs), more construction companies are focusing on adop... ver más
Revista: Infrastructures

 
Pahmi Pahmi,Siti Niah     Pág. pp. 205 - 221
This study aims to find out students? voices towards online debate through WhatsApp as alternative media to enhance critical thinking skills during Covid-19 pandemic. Sixteen undergraduate students participated in online debate for ten sessions throughou... ver más