TY - GEN
T1 - An adaptive data retrieval scheme for reducing energy consumption in mirrored video servers
AU - Song, Minseok
PY - 2006
Y1 - 2006
N2 - We propose a new energy-aware data retrieval scheme (EDR) for mirrored video servers. We start by analytically determining the retrieval period that balances bandwidth and buffer size. We then propose a data retrieval scheme in which the period can be dynamically changed to reflect server utilization, with the aim of giving disks more chance to enter low-power mode. Simulation results show that it saves up to 36% energy compared with conventional video server operation.
AB - We propose a new energy-aware data retrieval scheme (EDR) for mirrored video servers. We start by analytically determining the retrieval period that balances bandwidth and buffer size. We then propose a data retrieval scheme in which the period can be dynamically changed to reflect server utilization, with the aim of giving disks more chance to enter low-power mode. Simulation results show that it saves up to 36% energy compared with conventional video server operation.
UR - http://www.scopus.com/inward/record.url?scp=33746606803&partnerID=8YFLogxK
U2 - 10.1007/11758501_176
DO - 10.1007/11758501_176
M3 - Conference contribution
AN - SCOPUS:33746606803
SN - 3540343792
SN - 9783540343790
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1075
EP - 1078
BT - Computational Science - ICCS 2006
PB - Springer Verlag
T2 - ICCS 2006: 6th International Conference on Computational Science
Y2 - 28 May 2006 through 31 May 2006
ER -