Back to Search
Start Over
Adapting wave-front algorithms to efficiently utilize systems with deep communication hierarchies
- Source :
- Parallel Computing. 37:550-561
- Publication Year :
- 2011
- Publisher :
- Elsevier BV, 2011.
-
Abstract
- Large-scale systems increasingly exhibit a differential between intra-chip and inter-chip communication performance especially in hybrid systems using accelerators. Processor-cores on the same socket are able to communicate at lower latencies, and with higher bandwidths, than cores on different sockets either within the same node or between nodes. A key challenge is to efficiently use this communication hierarchy and hence optimize performance. We consider here the class of applications that contains wave-front processing. In these applications data can only be processed after their upstream neighbors have been processed. Similar dependencies result between processors in which communication is required to pass boundary data downstream and whose cost is typically impacted by the slowest communication channel in use. In this work we develop a novel hierarchical wave-front approach that reduces the use of slower communications in the hierarchy but at the cost of additional steps in the parallel computation and higher use of on-chip communications. This tradeoff is explored using a performance model. An implementation using the reverse-acceleration programming model on the petascale Roadrunner system demonstrates a 27% performance improvement at full system-scale on a kernel application. The approach is generally applicable to large-scale multi-core and accelerated systems where a differential in communication performance exists.
- Subjects :
- business.operation
Computer Networks and Communications
Computer science
Node (networking)
Distributed computing
Supercomputer
Computer Graphics and Computer-Aided Design
Theoretical Computer Science
Petascale computing
Roadrunner
Artificial Intelligence
Hardware and Architecture
Hybrid system
Programming paradigm
Upstream (networking)
business
Software
Subjects
Details
- ISSN :
- 01678191
- Volume :
- 37
- Database :
- OpenAIRE
- Journal :
- Parallel Computing
- Accession number :
- edsair.doi...........e343d9ad25b60dec40e8587693fa5831