Loading…

FREP: Energy proportionality for disk storage using replication

Energy saving has become a crucial concern in datacenters as several reports predict that the anticipated energy costs over a three year period will exceed hardware acquisition. In particular, saving energy for storage is of major importance as storage devices (and cooling them off) may contribute o...

Full description

Saved in:
Bibliographic Details
Published in:Journal of parallel and distributed computing 2012-08, Vol.72 (8), p.960-974
Main Authors: Kim, Jinoh, Rotem, Doron
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Energy saving has become a crucial concern in datacenters as several reports predict that the anticipated energy costs over a three year period will exceed hardware acquisition. In particular, saving energy for storage is of major importance as storage devices (and cooling them off) may contribute over 25 percent of the total energy consumed in a datacenter. Recent work introduced the concept of energy proportionality and argued that it is a more relevant metric than just energy saving as it takes into account the tradeoff between energy consumption and performance. In this paper, we present a novel approach, called FREP (Fractional Replication for Energy Proportionality), for energy management in large datacenters. FREP includes a replication strategy and basic functions to enable flexible energy management. Specifically, our method provides performance guarantees by adaptively controlling the power states of a group of disks based on observed and predicted workloads. Our experiments using a set of real and synthetic traces show that FREP dramatically reduces energy requirements with a minimal response time penalty. ► A new replication algorithm for energy-proportional storage management. ► Basic functions for load distribution, update consistency, and fault tolerance. ► A load prediction model based on past observations. ► Extensive evaluation results with real and synthetic traces
ISSN:0743-7315
1096-0848
DOI:10.1016/j.jpdc.2012.03.010