Evaluating Performance of Deterministic Algorithms on a Multicore Processor of a Public Cloud

Cloud computing is considered a paradigm in which customers can access their resources directly from the Internet. It has become a very efficient way to provide computational resources in a flexible way. With the increasing offer, commercial clouds are now being used not only by commercial companies but by researchers who deal with large amount of data flows and calculation activities as well. Although there are several works that analyze the performance of computational clouds, the major focus of them are on private clouds or Amazon EC2 public cloud. The objective of this paper is to evaluate the factors, such as size of input data, operating system and hardware, which are able to influence the performance of deterministic algorithms in the commercial cloud Azure. We present the performance of these algorithms, and show how the combination of these factors influence more or less in the performance of a multicore processor of a cloud computing environment when compared with a desktop computing environment.