Adaptive Object Placement in Object-Based Storage Systems with Minimal Blocking Probability
Dan Feng, Lingjun Qin · 2006
Object placement is an importance issue in Object- Based Storage System. A dynamic object placement algorithm is presented to minimize the blocking probability in heterogeneous storage systems. To address the online problem without a priori knowledge of workload parameters, or with changing workload, the algorithm employs an adaptive mechanism to estimate the characteristics of workloads. Due to a periodic adjustment of object placement through object replication and migration, the algorithm can dramatically reduce the overall blocking probability. The simulation results show that the performance of the algorithm exceeds that of a greedy algorithm. Furthermore, the adaptive algorithm with object replication and migration always outperforms the one with only migration especially in the situation where objects have large traffic intensities.