Paper Info
Title
Dynamic Load Balancing of Matrix-Vector Multiplications on Roadrunner Compute Nodes
Abstract
Hybrid architectures that combine general purpose processors with accelerators are currently being adopted in several large-scale systems such as the petaflop Roadrunner supercomputer at Los Alamos. In this system, dual-core Opteron host processors are tightly coupled with PowerXCell 8i accelerator processors within each compute node. In this kind of hybrid architecture, an accelerated mode of operation is typically used to off-load performance hotspots in the computation to the accelerators. In this paper we explore the suitability of a variant of this acceleration mode in which the performance hotspots are actually shared between the host and the accelerators. To achieve this we have designed a new load balancing algorithm, which is optimized for the Roadrunner compute nodes, to dynamically distribute computation and associated data between the host and the accelerators at runtime. Results are presented using this approach, for sparse and dense matrix-vector multiplications, that show load-balancing can improve performance by up to 24% over solely using the accelerators.
Year
DOI
Venue
2009
10.1007/978-3-642-03869-3_19
Euro-Par
Keywords
Field
DocType
matrix-vector multiplications,general purpose processor,performance hotspots,accelerator processor,hybrid architecture,roadrunner compute nodes,accelerated mode,dense matrix-vector multiplication,dynamic load,acceleration mode,dual-core opteron host processor,los alamos,petaflop roadrunner supercomputer,distributed computing,load balance
Load management,Supercomputer,Matrix (mathematics),Block cipher mode of operation,Load balancing (computing),Computer science,Parallel computing,Roadrunner,Acceleration,Distributed computing,Computation
Conference
Volume
ISSN
Citations 
5704
0302-9743
0
PageRank 
References 
Authors
0.34
10
2
Name
Order
Citations
PageRank
José Carlos Sancho138229.97
Darren J. Kerbyson21102104.36