Highly parallel GEMV with register blocking method on GPU architecture

Jian Yin, Hui Yu, Weizhi Xu, Yuxuan Wang, Zhu Tian, Yingping Zhang, Bochuan ChenJVCIR 2014 · OpenReview

Abstract

Highlights•We propose a register blocking method for GEMV on GPU.•The proposed method can improve the parallelism and reuse data on chip at the same time.•Different block sizes are tested to found the best block size on a GPU platform.
BibTeX
@article{yin2014highly,
  author = {Jian Yin and Hui Yu and Weizhi Xu and Yuxuan Wang and Zhu Tian and Yingping Zhang and Bochuan Chen},
  title = {{Highly parallel GEMV with register blocking method on GPU architecture}},
  journal = {J. Vis. Commun. Image Represent.},
  year = {2014},
  url = {https://openreview.net/forum?id=zPI2Q9sTsm},
}

Loading related content...

Figure