Distinguished Professor of Computer Science and Engineering at the University of California, Riverside
Friday, November 30, 2018
11:15am - 12:30pm
2150 Torgersen Hall
The importance of iterative graph algorithms has grown due to their widespread use in graph mining and analytics. Although computations on graphs with millions of nodes and edges contain vast amounts of data level parallelism, exploiting this parallelism is challenging due to the highly irregular nature of real-world graphs and dynamic nature of parallelism. In this talk I will present our recent results that greatly improve the computation, communication, and I/O efficiencies of graph processing on a cluster, a multicore machine, and on GPUs. In comparison to prior techniques, our reliable relaxed consistency protocol achieves 2.3x performance improvement on a 16-node cluster, our dynamic shards based I/O optimization achieves up to 2.8x performance improvement on a single multicore machine, and our warp segmentation technique achieves 1.3x-2.8x performance improvement on a single GPU while our vertex refinement technique achieves 2.7x performance improvement on a multi-GPU system.
Rajiv is a Distinguished Professor of Computer Science and Engineering at the University of California, Riverside. His research interests include Compilers, Architectures, and Runtimes for Parallel Systems. He has supervised PhD dissertations of 31 students including two winners of ACM SIGPLAN Outstanding Doctoral Dissertation Award. Papers coauthored by Rajiv with his students have been selected for: inclusion in 20 Years of PLDI (1979-1999), a best paper award in PACT 2010, and a distinguished paper award in ICSE 2003. Rajiv is a Fellow of the ACM, IEEE, and AAAS. He received the NSF Presidential Young Investigator Award and UCR Doctoral Dissertation Advisor/Mentor Award. He has chaired several major conferences including FCRC, PLDI, HPCA, ASPLOS, CGO, CC, HiPEAC, and LCTES. He also served as an Associate Editor for ACM Transactions on Architecture and Code Optimization and IEEE Transactions on Computers. Rajiv served as a member of a technical advisory group on networking and information technology created by PCAST.