Tackling Virtual Infrastructure Allocation in Cloud Data Centers: a GPU-Accelerated Framework

Allocating IT resources to Virtual Infrastructures (VIs) (i.e., groups of VMs, virtual switches, and their network interconnections) is a problem which belongs to a class known to be NP-hard. Several approaches designed to run on CPUs have been proposed for reducing the search space and finding suitable allocation solutions. Most algorithms, however, face scalability issues when considering current cloud data centers comprising thousands of servers. To overcome these limitations, this work offers a set of allocation algorithms designed for Graphic Processing Units (GPUs). Experimental results evaluate the scalability of the algorithms and demonstrate their ability to handle three large-scale data center topologies.