List scheduling heuristics for virtual machine mapping in cloud systems

International audienceThis article introduces the formulation of the Virtual MachinePlanning Problem in cloud computing systems. It deals with the effi-cient allocation of a set of virtual machine requests from customers intothe available pre-booked resources the broker has in a number of cloudproviders, maximizing the broker profit. Eight list scheduling heuristicsare proposed to solve the problem, by taking into account different criteriafor mapping request to available virtual machines. The experimentalevaluation analyzes the profit, makespan, and flowtime results of theproposed methods over a set of 400 problem instances that account forrealistic workloads and scenarios using real data from cloud providers