University of Cambridge > Talks.cam > Computer Laboratory Systems Research Group Seminar > Grappa: Scale-out Performance on Commodity Hardware for Irregular Applications

Grappa: Scale-out Performance on Commodity Hardware for Irregular Applications

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Eiko Yoneki.

Grappa is a runtime system for commodity clusters of multicore computers that presents a massively parallel, single address space abstraction to applications. Grappa’s purpose is to enable scalable performance of irregular parallel applications, such as branch and bound optimization, SPICE circuit simulation, and graph processing. Poor data locality, imbalanced parallel work and complex communication patterns make scaling these applications difficult.

Grappa serves both as a C++ user library and as a foundation for higher level languages. Grappa tolerates delays to remote memory by multiplexing thousands of lightweight workers to each processor core, balances load via fine-grained distributed work-stealing, increases communication throughput by aggregating smaller data requests into large ones, and provides efficient synchronization and remote operations. This talk will present a detailed description of the Grappa system and performance comparisons between Grappa, GraphLab, and the Cray XMT and MTA . We find that Grappa is 2.5X faster than Graphlab and 4X faster than the Cray systems.

Bio: Mark Oskin is an Associate Professor of Computer Science and Engineering at the University of Washington. He is the recipient of both the NSF CAREER and Sloan Foundation Awards. He has graduated three PhD students, and several masters students. He has published over 50 peer-reviewed articles on a wide range of topics including runtime systems for high performance graph analytics, dataflow computing, deterministic multiprocessing, and and intelligent memory. He has co-founded two companies and is currently the VP of Engineering of Konyac Inc.

This talk is part of the Computer Laboratory Systems Research Group Seminar series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2020 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity