Computer Organization and Design RISC-V Edition

5

Large and Fast

Exploiting Memory Hierarchy

Abstract

This chapter describes how a memory hierarchy exploits locality. It shows how the principle of locality gives designers a way to overcome the long latency of memory access. It explains how strategies to exploit locality can be used at all levels of the memory hierarchy. The chapter concludes by emphasizing that memory systems are a central design issue for parallel processors, and that the growing importance of the memory hierarchy in determining system performance means that this important area will continue to be a focus of both designers and researchers for some years to come.

Keywords

cache performance; virtual memory; memory hierarchy; ARM Cortex-A53; Intel Core i7; Intel; temporal ...