ACM Transactions on Architecture and Code Optimization · 2012 · 28 citations · 34 references
ComputingComputer EngineeringComputer ScienceComputer ArchitectureParallel AlgorithmsParallel ComputingData ParallelismDirect Memory AccessParallel ArchitecturesParallel Computer ArchitectureParallel Computing SystemsParallel Performance AnalysisCycle-accurate ModelingData Parallel ExecutionDirect Memory Access (Dma)Direct Memory Access ArchitectureDma Performance OptimizationDouble BufferingMulti-core ComputingParallel Performance Optimization
In this paper we investigate a general approach to automate some deployment decisions for a certain class of applications on multi-core computers. We consider data-parallelizable programs that use the well-known double buffering technique to bring the data from the off-chip slow memory to the local memory of the cores via a DMA (direct memory access) mechanism. Based on the computation time and size of elementary data items as well as DMA characteristics, we derive optimal and near optimal values for the number of blocks that should be clustered in a single DMA command. We then extend the results to the case where a computation for one data item needs some data in its neighborhood. In this setting we characterize the performance of several alternative mechanisms for data sharing. Our models are validated experimentally using a cycle-accurate simulator of the Cell Broadband Engine architecture.
34
A bridging model for parallel computation
Leslie G. Valiant · Communications of the ACM · 1990 · 3.7K citations · Full text
LogP: towards a realistic model of parallel computation
David Culler, Richard M. Karp, David A. Patterson et al. · 1993 · 1.5K citations