Energy Efficiency Evaluation of a Data Mining Prefetching Algorithm for Hybrid Storage Systems

This paper evaluates the energy performance of a novel data prefetching scheme in multi level storage system. This data prefetching algorithm (DM-PAS) proposes a hybrid storage system that consists of Solid State Drives (SSD), conventional Hard Disk Drives (HDD) and Tape Drives (TD). The algorithm was originally developed to meet the growing demand of high speed data fetching in the large cloud systems. Inspired by enormous download requests to the EROS (Earth Resources Observation and Science) center of the U.S. Geological Survey, the purpose of the developed algorithm was to achieve high I/O performance and reliability and tested in a real world hybrid storage system. However, as energy efficiency in the computing system context is becoming crucial day by day, energy aware algorithms offers a distinct appeal to the data center administrators. Simulations were performed to evaluate the efficacy of the tested energy-conserving strategy DM-PAS. Simulations demonstrate that utilization of DM-PAS results in reduced energy consumption of hybrid storage systems in a dynamic environment as compared with the same storage systems without using DM-PAS.

[1]  Jesús Carretero,et al.  Saving power in flash and disk hybrid storage system , 2009, 2009 IEEE International Symposium on Modeling, Analysis & Simulation of Computer and Telecommunication Systems.

[2]  Xiao Qin,et al.  HYBUD: An Energy-Efficient Architecture for Hybrid Parallel Disk Systems , 2009, 2009 Proceedings of 18th International Conference on Computer Communications and Networks.

[3]  Daniel A. Reed,et al.  Automatic ARIMA time series modeling for adaptive I/O prefetching , 2004, IEEE Transactions on Parallel and Distributed Systems.

[4]  Mais Nijim,et al.  DM-PAS: A Data Mining Prefetching Algorithm for Storage System , 2011, 2011 IEEE International Conference on High Performance Computing and Communications.

[5]  Wei Luo,et al.  Energy-Aware Prefetching for Parallel Disk Systems: Algorithms, Models, and Evaluation , 2009, 2009 Eighth IEEE International Symposium on Network Computing and Applications.

[6]  Xiao Qin,et al.  A prefetching scheme for energy conservation in parallel disk systems , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[7]  Jun Wang,et al.  A Novel Power Management for CMP Systems in Data-Intensive Environment , 2011, 2011 IEEE International Parallel & Distributed Processing Symposium.

[8]  Yao Sun,et al.  Sacrificing Reliability for Energy Saving: Is it worthwhile for disk arrays? , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[9]  Yuanyuan Zhou,et al.  Reducing Energy Consumption of Disk Storage Using Power-Aware Cache Management , 2004, 10th International Symposium on High Performance Computer Architecture (HPCA'04).