GPU Accelerated Planning and Placement of Edge Clouds

Future 5G communication will rely on edge computing to meet diverse Quality of Service Requirements, in particular, end-to-end latency in the order of milliseconds. Placing edge clouds requires solving a large-scale set cover problem. We show how the challenge posed by the size of the problem can be overcome through GPU accelerated Natural Evolutional Strategies (NES). We present a system that allows the evaluation of technology choices, latency requirements and locations that should be covered at the example of the United States.