Faulty Links Optimization and Load-scalability in Hypercube Networks
The nodes or processors in a hypercube networks send random point-to-point messages to other nodes in the network. We assume that a node can generate at most one message per cycle time with a global clock algorithm that drives all processors. The primary objective of this research project is to simulate the data throughput and hot spot properties of a hypercube network. In this research, we aim to develop a simple stored and forward one-bit round robin routing algorithm for transmitting and receiving messages between the nodes within the hypercube network. One of the main objectives of this research project is to demonstrate that the proposed routing algorithm not only works in an ideal situation where all the communication links between the nodes are in the operational mode but also works fine in the presences of faulty links. Furthermore, the goal is to implement the designed stored and forward one-bit round robin routing algorithm that can show some good load-scalability characteristics among the hypercube nodes. In addition, the effects of different architectures such as the dimension of the hypercube network, input parameter values, and the introduction of faulty links are evaluated in this research work.
- Syed S. Rizvi, Khaled M. Elleithy, and Aasia Riasat, “An Efficient Single Bit Store and Forward (SBSF) Routing Algorithm for Mesh-Hypercube (M-H) Networks,” The 2008 World Congress in Computer Science, Computer Engineering, and Applied Computing WORLDCOMP’08. Las Vegas, Navada. pp. 69 – 75, July 14 – 17, 2008. PDF