Paper Info
Title
Crunching large graphs with commodity processors
Abstract
Crunching large graphs is the basis of many emerging applications, such as social network analysis and bioinformatics. Graph analytics algorithms exhibit little locality and therefore present significant performance challenges. Hardware multithreading systems (e.g., Cray XMT) show that with enough concurrency, we can tolerate long latencies. Unfortunately, this solution is not available with commodity parts. Our goal is to develop a latency-tolerant system built out of commodity parts and mostly in software. The proposed system includes a runtime that supports a large number of lightweight contexts, full-bit synchronization and a memory manager that provides a high-latency but high-bandwidth global shared memory. This paper lays out the vision for our system and justifies its feasibility with a performance analysis of the run-time for latency tolerance.
Year
Venue
Keywords
2011
HotPar
performance analysis,high-bandwidth global shared memory,commodity processor,present significant performance challenge,latency-tolerant system,commodity part,large number,proposed system,memory manager,social network analysis,large graph,network analysis,shared memory,memory management
Field
DocType
Citations 
Multithreading,Locality,Synchronization,Shared memory,Cray XMT,Computer science,Concurrency,Parallel computing,Software,Memory management,Distributed computing
Conference
9
PageRank 
References 
Authors
0.58
20
9
Name
Order
Citations
PageRank
Jacob Nelson128117.27
Brandon Myers2615.14
A. H. Hunter390.91
Preston Briggs437932.43
Luis Ceze52183125.93
Carl Ebeling61405185.32
Dan Grossman7121871.43
Simon Kahan8279209.26
Mark Oskin990676.63