Vector repacking algorithms for power-aware computing

Abstract

Abstract-In this paper we experiment with practical algorithms for the vector repacking problem and its variants. Vector repacking, like vector packing, aims to pack a set of input vectors such that the number of bins used is minimized, while minimizing the changes from the previous packing. We also consider a variant of vector repacking that stores additional copies of items with the goal of improving the performance of vector repacking algorithms. In addition, our algorithms are parameterized so that they can be effectively optimized for a variety of resource allocation applications with different input characteristics and different cost functions.

FFD Variants for Virtual Machine Placement in Cloud Computing Data Centers

“…At 20% tolerance the proposed algorithm used only 5.4% more power than optimal algorithm. 5) Power saving by live VM migration using vector based repacking: Consuegra et al in [12] proposed a vector repacking algorithm called replicas to minimize operational costs of a datacenter, that include power and migration cost. This algorithm places copies of each item/VM on different PMs that are selected by replica allocation algorithm.…”