Filtering superfluous prefetches using density vectors

Wei Fen Lin, Steven K. Reinhardt, Doug Burger, Thomas R. Puzak

研究成果: Article同行評審

28 引文 斯高帕斯(Scopus)

摘要

A previous evaluation of scheduled region prefetching showed that this technique eliminates the bulk of main-memory stall time for applications with spatial locality. The downside to that aggressive prefetching scheme is that, even when it successfully improves performance, it increases enormously the amount of superfluous memory traffic generated by a program. In this paper, we measure the predictability of spatial locality using density vectors, bit vectors that track the block-level access pattern within a region of memory. We evaluate a number of policies that use density vector information to filter out prefetches that are unlikely to be useful. We show, that, across our benchmarks, an average of 70% of useless prefetches can be eliminated with virtually no overall performance loss front reduced coverage. Thanks to the increase in prefetch accuracy a few benchmarks show performance improvements as high as 35% over the base region prefetching scheme.

原文English
文章編號21
頁(從 - 到)124-132
頁數9
期刊Proceedings - IEEE International Conference on Computer Design: VLSI in Computers and Processors
DOIs
出版狀態Published - 2001

All Science Journal Classification (ASJC) codes

  • 硬體和架構
  • 電氣與電子工程

指紋

深入研究「Filtering superfluous prefetches using density vectors」主題。共同形成了獨特的指紋。

引用此