Title :
Analyze reuse distance for multithreaded sparse iterative solver
Author_Institution :
Sorrell Coll. of Bus., Troy Univ., Troy, AL, USA
Abstract :
Finding the behavior of application programs has been studied by many researchers to improve the performance. Especially, reuse distance, number of distinct data element between two adjacent references to a data element, has been studied by many researchers since reuse distance is independent of hardware architecture configurations. However, with the advent of multicore architectures, predicting the behavior based on reuse distance becomes much more difficult. In this paper, we explore the reuse distance of a multithreaded sparse iterative solver on multicore architecture to predict the behavior. Experiment results show that we can use reuse distance of a multithreaded sparse iterative solver on multicore architectures to predict the behavior.
Keywords :
iterative methods; multi-threading; multiprocessing systems; application program behavior; data element; multicore architectures; multithreaded sparse iterative solver; reuse distance analysis; Benchmark testing; Hardware; Instruction sets; Load modeling; Multicore processing; Sparse matrices;
Conference_Titel :
Southeastcon, 2011 Proceedings of IEEE
Conference_Location :
Nashville, TN
Print_ISBN :
978-1-61284-739-9
DOI :
10.1109/SECON.2011.5752982