June 24th, 2017 Toronto, ON, Canada | Held in conjunction with ISCA 2017
9:00 am - 9:10 amOpening Remarks
9:10 am - 10:10 amKeshav Pingali, The University of Texas at Austin
10:10 am - 11:00 amTechnical Session I
Session Chair: Rajiv Gupta, University of California, Riverside
Cache-Guided Scheduling: Exploiting Caches to Maximize Locality in Graph Processing
Anurag Mukkara, MIT CSAIL; Nathan Beckmann, Carnegie Mellon University; Daniel Sanchez, MIT CSAIL
An Operational Performance Model of Breadth-First Search
Sreepathi Pai, M. Amber Hassaan and Keshav Pingali, The University of Texas at Austin
11:00 am - 11:15 amBreak
11:15 am - 12:15 pmKeynote II: Fast Stochastic Algorithms for Machine Learning
Christopher De Sa, Stanford University
12:15 pm - 1:30 pmLunch
1:30 pm - 2:30 pmKeynote III: Making Parallelism Pervasive with the Swarm Architecture
Daniel Sanchez, MIT CSAIL
2:30 pm - 3:20 pmTechnical Session II
Session Chair: Xuehai Qian, University of Southern California
Near Data Processing at Runtime
Mark Sutherland, EcoCloud and École Polytechnique Fédérale de Lausanne (EPFL); Natalie Enright Jerger, University of Toronto
Exploiting Nested Parallelism to Accelerate Parallel Recursive Graph Algorithms
Fanny Nina-Paravecino and Qianqian Fang, Northeastern University; Norm Rubin, NVIDIA; David Kaeli, Northeastern University
3:20 pm - 3:30 pmBreak
3:30 pm - 4:30 pmKeynote IV: General Purpose Acceleration and the Challenges for Irregular Workloads
Tony Nowatzki, University of California, Los Angeles
4:30 pm - 5:00 pmPanel Discussion: TBD
Moderator: Xuehai Qian, University of Southern California
Panelist: Daniel Sanchez, MIT CSAIL; Keshav Pingali, The University of Texas at Austin; Tony Nowatzki, University of California, Los Angeles; Christopher De Sa, Stanford University