Gunrock is an open-source programmable GPU framework for graph analytics, delivering comparable performance to hardwired GPU primitives and exceeding the performance of the fastest CPU frameworks. Gunrock delivers a sophisticated load balancing and a data-centric programming model for GPU computation. Currently, Gunrock is developing higher-level primitives and higher-level programming models which can scale beyond a single node, mutable data structures, and more back ends. Read more about Gunrock in our PPoPP 2016 paper: http://escholarship.org/uc/item/6xz7z9k0.
By attending this webinar, you'll learn:
Key challenges for GPU graph analytics
How to efficiently map graph analytics to GPUs
The design decisions and architecture of the Gunrock graph analytics library
Tuesday, April 26, 2016
10:00am – 11:00am PDT
*Required. Webinar access will be emailed to you.
We value your privacy.
Can't make it? Sign up anyway to get the recording!
Prof. John Owens Ph.D.
John Owens is the Child Family Professor of Engineering and Entrepreneurship in the Department of Electrical and Computer Engineering at the University of California, Davis. He earned his Ph.D. in electrical engineering from Stanford University in 2003 and his B.S. in electrical engineering and computer sciences from the University of California, Berkeley, in 1995. At UC Davis, John leads a research group with a primary focus on GPU computing. He is an NVIDIA CUDA Fellow. John cotaught Udacity's massively open online course (MOOC) "Introduction to Parallel Programming with CUDA" in 2013, and won the 2004 Department of Energy Early Career Principal Investigator Award.