Push-Based Prefetching in Remote Memory Sharing System - Cloud, Grid and High Performance Computing: Emerging Applications - page 271

Chapter 17

Push-Based Prefetching in
Remote Memory Sharing System

Rui Chu
National University of Defense Technology, China

Nong Xiao
National University of Defense Technology, China

Xicheng Lu
National University of Defense Technology, China

ABSTRACT

Remote memory sharing systems aim at the goal of improving overall performance using distributed
computing nodes with surplus memory capacity. To exploit the memory resources connected by the
high-speed network, the user nodes, which are short of memory, can obtain extra space provision. The
performance of remote memory sharing is constrained with the expensive network communication cost.
In order to hide the latency of remote memory access and improve the performance, we proposed the
push-based prefetching to enable the memory providers to push the potential useful pages to the user
nodes. For each provider, it employs sequential pattern mining techniques, which adapts to the charac-
teristics of memory page access sequences, on locating useful memory pages for prefetching. We have
verified the effectiveness of the proposed method through trace-driven simulations.

INTRODUCTION

environment, and provide services for a large
number of users. The resources to be shared in
such systems include CPU cycles, storage, data,
and, as particularly discussed in this work, the
memory.
As one of the most important resources in
computer architecture, memory plays a key role
in the factors impacting the system performance.
Especially for the memory-intensive applications
that have large work sets, or the I/O-intensive appli-

The rapid developing of Internet has boosted
the bloom of network computing technology. As
typical systems, cluster computing, peer-to-peer
computing, grid computing, as well as cloud com-
puting, commonly focus on the goal of sharing
various resources distributed in a certain network