Analysis of Optimal File Placement for Energy-Efficient File-Sharing Cloud Storage System

Fumio Machida,Hirotake Abe,Koji Hasebe,Kazuhiko Kato

doi:10.1109/tsusc.2020.3037260

Fumio Machida, Hirotake Abe + Show 2 more

Open Access

https://doi.org/10.1109/tsusc.2020.3037260

Copy DOI

Abstract

Popular data concentration is a widely accepted storage energy-saving technique which places frequently-accessed data on a small subset of hard disks and spins-down other infrequently-accessed disks. Many previous studies use intuitive heuristic algorithms for data placement that promote the imbalance in the access frequencies across hard disks. However, the relevance and the optimality of such file placements have not been rigorously investigated. In this paper, we formally define the energy-saving file placement problem under the capacity and performance constraints as a combinatorial optimization problem and show the theory of the optimal file placement where the file access rates in the next period are given. Our analysis based on a stochastic process of disk state transitions gives the theoretical support for the common heuristic placement method. To examine the effectiveness of the optimal file placement, we experimentally evaluate the energy-efficiency of a test storage system using the file access rates generated from the real access traces from Flickr. The experimental results show that the energy consumption can be reduced by 31.8 percent with the optimal file placement compared to the evenly distributed file placement. We also conduct simulation experiments to confirm the energy-saving impacts in larger-scale storage systems.

Highlights

ENERGY-efficiency of cloud storage systems has been Popular data concentration (PDC) is to exploit the skewness of the data or file access freexplored as an important challenge for service provid- quencies, which are often observed in many application ers and enterprises owning the storage systems
While solid state drive (SSD) and Non- diction is usually made by analyzing historical workload volatile memory (NVM) are gaining popularity as energy- data, the placement method typically relies on a heuristic efficient fast storage devices, hard disk drives (HDDs) are algorithm that places data on disks in order of data access still dominant storage components used in data centers frequencies [5][6][8][12]
The results clearly show that the increased uncertainty of file access rates significantly impact on the expected energy consumption by the optimal file placement

Summary

Introduction

ENERGY-efficiency of cloud storage systems has been PDC is to exploit the skewness of the data or file access freexplored as an important challenge for service provid- quencies, which are often observed in many application ers and enterprises owning the storage systems. Present cost-effective cloud or enterprise storage data on a subset of hard disks and spinning down the disks, systems are expected to extend their capacity to store even the storage energy consumption can be reduced signifia larger volume of data. Continuous efforts to improve the en- effectiveness was validated through some simulation studergy efficiency of large-scale HDD-based storage systems ies [5][6][8] None of these studies provide the will be essential in the coming data age

Objectives

Methods

Findings

Conclusion