International Journal of Computer Science Issues

More than a traditional journal...

Symmetric key cryptography is one of the most critical computing problems that need high performance computing power resources. The use of large key sizes and complex encryption/decryption algorithms to achieve unbreakable state has led to an increased time computational complexity. Traditionally, this problem is solved in the grid environment by partitioning data streams into several blocks of a predefined size. This is done while sequentially reading the data from the raw data file. The grid manager node then takes the responsibility of passing these blocks to the executer nodes where different blocks are processed separately and simultaneously. Although this technique allows parallel processing to speed up the encryption/decryption process, creating blocks by sequentially reading the data file and distributing these blocks on executers synchronously by the central manager node is a poor technique and a source of delay. In this paper, we present a novel approach that tackles this problem by allowing executers to access data file at random and asynchronously exchange the blocks among them, thereby, delay is significantly reduced and data size can be scaled up. In order to show the merit of our approach experiments have been conducted through a system-level middleware for grid computing called Alchemi. The results show a remarkable performance enhancement in our approach over traditional approaches in terms of speed.