Optimizing Latency under Throughput Requirements for Streaming Applications on Cluster Execution
F.Guirado, A.Ripoll, C.Roig, E.Luque
IEEE International Conference on Cluster Computing (Cluster 2005)
Boston, Massachusetts, USA, September 27 - 30, 2005
Abstract
Parallelism in applications that act on a stream of input data can be exploited with two different approaches, spatial and temporal. In this paper we propose a new task mapping algorithm to exploit temporal parallelism efficiently when the streaming application is running in a pipeline fashion. We compare the performance of both approaches, in terms of latency and throughput for a video compression application. The results show that the pipeline execution overcomes significantly spatial parallelism and provides better scalability when the dimension of the problem is augmented.