PPoPP 2019
Sat 16 - Wed 20 February 2019 Washington, DC, United States
Tue 19 Feb 2019 11:20 - 11:45 at Salon 12/13 - Session 6, Best Paper Candidates Chair(s): Rudolf Eigenmann

This paper proposes GSWITCH, a pattern-based algorithmic auto-tuning system that dynamically switches between optimization variants with negligible overhead. Its novelty lies in a small set of algorithmic patterns that allow for the configurable assembly of variants of the algorithm. The fast transition of GSWITCH is based on a machine learning model trained using 644 real graphs. Moreover, GSWITCH provides a simple programming interface that conceals low-level tuning details from the user. We evaluate GSWITCH on typical graph algorithms (BFS, CC, PR, SSSP, and BC) using Nvidia Kepler and Pascal GPUs. The results show that GSWITCH runs up to 10× faster than the best configuration of the state-of-the-art programmable GPU-based graph processing libraries on 10 representative graphs. GSWITCH outperforms Gunrock on 92.4% cases of 644 graphs which is the largest dataset evaluation reported to date.

Tue 19 Feb

Hide past events
PPoPP-2019-papers
10:55 - 12:35: Main Conference - Session 6, Best Paper Candidates at Salon 12/13
Chair(s): Rudolf EigenmannUniversity of Delaware
PPoPP-2019-papers10:55 - 11:20
Talk
Qingsen WangCollege of William and Mary, Pengfei SuCollege of William and Mary, Milind ChabbiUber Technologies, Xu LiuCollege of William and Mary
DOI
PPoPP-2019-papers11:20 - 11:45
Talk
Ke Meng, Jiajia LiGeorgia Institute of Technology, Pacific Northwest National Laboratory, Guangming TanChinese Academy of Sciences(CAS), Ninghui SunState Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, University of Chinese Academy of Sciences
DOI
PPoPP-2019-papers11:45 - 12:10
Talk
Umut AcarCarnegie Mellon University, Vitaly AksenovInria & ITMO University, Arthur CharguéraudInria, Mike RaineyIndiana University, USA
DOI
PPoPP-2019-papers12:10 - 12:35
Talk
Xiuhong LiPeking University, Eric LiangPeking University, Shengen YanSenseTime, Jia LianchengPeking University, Yinghan LiSenseTime
DOI