Distributed Concurrent Actor Models with Akka.NET and CAF

Abstract

Many domains like deep learning, machine learning, and computational finance uses Graphical Processing Units (GPUs) for decreasing the execution time. GPUs are widely used in data centers for high performance computing where virtualization techniques are intended for optimizing the resource utilization. However, these data centers have some downsides like energy consumption and acquisition cost. We propose to take this resource virtualization a step forward by making use of "sparse" computers over the Internet that have powerful processing capabilities. The approach is inspired from the smart grid model. In this way, we want to evaluate the feasibility of introducing a service that will provide processing capabilities through renting processing resources from domestic users. This model can prove usefully for solving subdivision of the main algorithm