William & Mary

Best Paper at PPoPP '19

xu_liu_at_ppopp_19.jpg

Professor Xu Liu's research group has been awarded Best Paper at PPoPP '19.  They also received Best Paper honors last year at PPoPP '18.

The paper "Lightweight Hardware Transactional Memory Profiling" by W&M graduate students Qingsen Wang and Pengfei Su, their adviser, Professor Xu Liu, and collaborator Milind Chabbi of Scalable Machines was selected as the best paper at the 24rd ACM SIGPLAN Annual Symposium on
Principles and Practice of Parallel Programming (PPoPP'19).  This paper describes a new, practical tool for hardware transactional memory (HTM), which is a promising technique for thread synchronization.
Besides the tool itself, this paper also describes the design of the first thorough set of HTM benchmark suite, which consists of more than 30 applications.

PPoPP is the premier forum for research in all aspects of parallel software. This year, there are only 29 papers accepted out of 152 submissions (19% acceptance rate). The Best Paper Award is given to a single paper among all submissions.