Optimized Proactive Recovery in Erasure-Coded Cloud Storage Systems

Cloud data centers have started utilizing erasure coding in large-scale storage systems to ensure high reliability with limited overhead compared replication. However, recovery incurs network bandwidth consumption also play an important role the energy of centers. Heuristic proactive algorithms select all blocks from failure-predicted disk/machine and perform replication that contributes huge savings. they fail optimize selection. Optimization can further improve resource To address this issue, we propose a algorithm applies minimization on selected for by considering necessary appropriate constraints are constructed based system’s current traffic duplication information. We evaluate proposed using extensive simulations. Experiments show reduces 60% 46% heuristic approach. Also, methods reduce up 52%