What type of study is this?

This is a Quantitative Study study.

October 1, 2025ElectronicsOpen Access

Automatic Sparse Matrix Format Selection via Dynamic Labeling and Clustering on Heterogeneous CPU–GPU Systems

View Full Paper

Ask AI

Bookmark

Authors

ZSZheng ShiSouth China University of Technology YZYi ZouJingdezhen Ceramic Institute XSXianfeng SongXi'an University of Science and Technology

Discussion

Member takes

Implication

This research demonstrates improved performance in sparse matrix-vector multiplication across CPU and GPU platforms, suggesting a novel dynamic labeling approach.

Key Points

Dynamic labeling reduces label noise by using multiple labels for closely performing formats, enhancing prediction.
The proposed method achieved 99.2% accuracy in hardware selection and 98.95% in format prediction among the tested matrices.
End-to-end evaluation indicated CPUs can achieve speedups up to 3.15× and GPUs up to 1.94× over traditional formats.
Cross-round evaluations confirmed the stability and generalizability of the optimization strategy across different runs.

Cite This Study

Shi et al. (2025) studied this question.

synapsesocial.com/papers/68dd91dafe798ba2fc4994ea https://doi.org/10.3390/electronics14193895

View Full Paper

Ask AI

Bookmark

Implication

This research demonstrates improved performance in sparse matrix-vector multiplication across CPU and GPU platforms, suggesting a novel dynamic labeling approach.

Automatic Sparse Matrix Format Selection via Dynamic Labeling and Clustering on Heterogeneous CPU–GPU Systems

Discussion

Member takes

Implication

Key Points

Cite This Study

Automatic Sparse Matrix Format Selection via Dynamic Labeling and Clustering on Heterogeneous CPU–GPU Systems

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider