scispace - formally typeset
W

Weifeng Liu

Researcher at China University of Petroleum

Publications -  45
Citations -  1441

Weifeng Liu is an academic researcher from China University of Petroleum. The author has contributed to research in topics: Sparse matrix & Computer science. The author has an hindex of 17, co-authored 37 publications receiving 1110 citations. Previous affiliations of Weifeng Liu include University of Copenhagen & Norwegian University of Science and Technology.

Papers
More filters
Proceedings ArticleDOI

CSR5: An Efficient Storage Format for Cross-Platform Sparse Matrix-Vector Multiplication

TL;DR: CSR5 (Compressed Sparse Row 5), a new storage format, which offers high-throughput SpMV on various platforms including CPUs, GPUs and Xeon Phi, is proposed for real-world applications such as a solver with only tens of iterations because of its low-overhead for format conversion.
Posted Content

CSR5: An Efficient Storage Format for Cross-Platform Sparse Matrix-Vector Multiplication

TL;DR: In this article, the authors proposed CSR5 (Compressed Sparse Row 5), a new storage format, which offers high-throughput SpMV on various platforms including CPUs, GPUs and Xeon Phi.
Proceedings ArticleDOI

An Efficient GPU General Sparse Matrix-Matrix Multiplication for Irregular Data

TL;DR: This work presents a GPU SpGEMM algorithm that particularly focuses on load balancing, memory pre-allocation for the result matrix, and parallel insert operations of the nonzero entries that is experimentally found to be the fastest GPU merge approach.
Journal ArticleDOI

A framework for general sparse matrix-matrix multiplication on GPUs and heterogeneous processors

TL;DR: This work proposes a framework for SpGEMM on GPUs and emerging CPU-GPU heterogeneous processors using the CSR format, and proposes an efficient parallel insert method for long rows of the resulting matrix and develops a heuristic-based load balancing strategy.
Book ChapterDOI

A Synchronization-Free Algorithm for Parallel Sparse Triangular Solves

TL;DR: This paper proposes a novel approach for SpTRSV in which the ordering between components is naturally enforced within the solution stage, and is an order of magnitude faster for the preprocessing stage than existing methods.