FOBS: A Lightweight Communication Protocol for Grid Computing

The advent of high-performance networks in conjunction with low-cost, powerful computational engines has made possible the development of a new set of technologies termed computational grids. These technologies are making possible the creation of very large-scale distributed computing systems by interconnecting geographically distributed computational resources via very high-performance networks. This provides tremendous computational power that can be brought to bear on large-scale problems in several domains, making feasible the development of high-performance distributed applications that, due to the low bandwidth and best-effort nature of the Internet1 environment, were heretofore infeasible. However, many problems remain before Grid computing can reach its full potential. One particularly difficult issue is that of utilizing fully the available bandwidth while being in some sense fair to competing traffic flows. It has been widely demonstrated that TCP, the communication protocol...

• Type: Conference
• Year: 2003
• Where: EUROPAR
• Authors: Phillip M. Dickens
• Keywords: Computational Grids, Distributed Application, EUROPAR 2003, Powerful Computational Engines