Node-Capability-Aware Replica Management for Peer-to-Peer Grids

Agneeswaran, Vijay Srinivas; D, Janakiram

doi:10.1109/tsmca.2009.2015960

Cited by 9 publications

(1 citation statement)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Agneeswaran etc. [12] establish relationship model between data replica and availability, and then set a threshold according to the need, and calculate minimum number of replica that each file block needs. The other is based on historical information, build a statistical prediction model, and thus predict the optimal layout of replica placement.…”

Section: Introductionmentioning

confidence: 99%

An activity-based replica placement method of energy-conservation

Zhao

Wang

2013

2013 International Conference on Fuzzy Theory and Its Applications (iFUZZY)

View full text Add to dashboard Cite

Aiming at the problems of increasing energy consumption of cloud storage, big environmental resources pressure, exist research on replica placement lacking of focus on energy consumption, this paper proposes a energy conservation replica placement method ECRP(Energy Conservation Replica Placement, ECRP). Based on the three main different levels of energy consumption of hardware components, the data node is divided into hot zone, warm zone and cold zone three logical zone, each zone uses different power management strategies and replica placement strategies, and after building a file activity prediction model which based on the simulated annealing BP neural network, files are stored in the data nodes on different zone according to the type of file activity, and then calculate the file activity periodically, when file activity exceeds a certain threshold, the file begins flow among the hot zone, warm zone and cold zone step by step in the two-way. Experiment results show that this method better solve the problems of replica placement and energy consumption optimization, have good energy saving effect.

show abstract

Section: Introductionmentioning

confidence: 99%