ConCCL: Optimizing ML Concurrent Computation and Communication with GPU DMA Engines

Anirudha Agrawal, Shaizeen Aga, Suchita Pati, Mahzabeen Islam. ConCCL: Optimizing ML Concurrent Computation and Communication with GPU DMA Engines. In IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2025, Ghent, Belgium, May 11-13, 2025. pages 1-11, IEEE, 2025. [doi]

Authors

Anirudha Agrawal

This author has not been identified. Look up 'Anirudha Agrawal' in Google

Shaizeen Aga

This author has not been identified. Look up 'Shaizeen Aga' in Google

Suchita Pati

This author has not been identified. Look up 'Suchita Pati' in Google

Mahzabeen Islam

This author has not been identified. Look up 'Mahzabeen Islam' in Google