Article Preview
Top1. Introduction
Parallel computing have been widely studied in the field of computer science and has proven to be critical when researching high performance solutions. Modern computer systems employ multiple processors which may take part in execution in parallel to enhance the performance of an application. The benefits of parallel computing need to take into consideration the number of processors being deployed, the complexity of the system as well as the communication incurred between various processing units known as nodes. The topology of interconnection networks plays a key role in the performance of parallel computing systems. Significant progress has been made in the past decades in the design and development of efficient interconnection networks (Chhabra et al., 2009; Parhami, 2000; Kim et al., 1999; Cheng et al., 1994; Shi et al., 2005; Zhang, 2002). Hypercube architecture is a promising approach to improve the performance of multiprocessor parallel systems. Numerous variants of hypercube have been reported in the literature (Amway et al., 1991; Kumar et al., 1992; Efe, 1991; Loh et al., 2005; Khan et al., 2013). These variants include many attractive properties, including regularity, symmetry, small diameter, high connectivity and scalability. A number of approaches have been suggested and implemented to improve the overall performance of such systems (Efe, 1992; Ghose et al., 1995; Kumar, 2012). Nevertheless, no single topology claims to have all the desirable topological features. For example the Crossed Cube architecture denoted by CC is derived from the hypercube to have a smaller diameter which is almost half of its parent architecture (Efe, 1992). However, the CC makes no significant improvement in the hardware cost compared to the hypercube. Similarly, a new architecture known as Folded Crossed Hypercube (FCH) (Khan et al., 2013) network is designed by augmenting the CC network. Some extra links called complementary links are introduced which help to reduce the diameter further. The improved diameter is obtained however at a greater value of degree. The Exchange Hypercube (EH) is another example which is also derived from hypercube and has lesser hardware cost by reducing number of links in the hypercube (Loh et al., 2005). This, however, results no improvement in the diameter and the connectivity in the EH is decreased. Similarly, there are other numerous interconnection networks which are designed by improving a particular topological parameter. Some examples are the Folded Hypercube (FH) (Amway et al., 1991), Dual Cube (DC) (Li et al., 2000), Folded Dual Cube (FDC) (Adhikari et al., 2008), Meta Cube (MC) (Li et al., 2002), Folded Met cube (FMC) (Adhikari et al., 2009), and the Folded Cross Cube (FCC) (Adhikari et al., 2010).