The Static Performance Effect of Hybrid- Hierarchical Interconnection by Shifted Completely Connected Network

Abstract

Massively parallel computer (MPC) systems execute many operations based on internal networks called interconnection networks. The performance of these networks is affected by their topologies.There are many topologies of interconnection networks for MPC systems, unfortunately, they faced many drawbacks. Expanding the size of the network degrades the performance of these topologies. That is why this current paper presents a hybrid-hierarchical interconnection network (HIN) topology by Shifted Completely Connected Network (SCCN) to circumvent the drawbacks of the existing topologies. An experimental evaluation involving the design and development of a hierarchical network was carried out. A two-dimensional higher level networks has been produced and its static network performance parameters evaluated through simulators. The finding of the simulations has shown some good performances compared to many previous designed networks. SCCN is better than all conventional networks in terms of diameter, cost and average distance.

Cost-effectiveness analysis of a shifted completely connected network for massively parallel computer systems

“…Based on the results of the evaluation process, the research community decides whether the network is eligible to be used or it will be subject to improvement. In previous work, we have evaluated the static network performance of SCCN in terms of node degree, wiring complexity, diameter, and many static parameters [15]. Therefore, in this paper, we will use the results of these parameters to analyze the static cost-effectiveness factor of SCCN by forming a relationship between these static network parameters in a specific formula.…”

“…Figure 5 shows the architecture of the System-Level network. The architecture of each level of this network was discussed previously in detail in previous work [15].…”

“…Analyzing the static cost-effectiveness of SCCN required particular criteria to create a relationship between some of the static network parameters which we have evaluated in previous work [15]. The parameters which will be used in analyzing the static cost-effectiveness factor of SCCN are the total number of nodes, the node degree, the network diameter, and the wiring complexity.…”

“…N odeDegree × W iringComplexity Diameter × T otalN umberof N odes (20) In previous work, we have calculated the results of the node degree, wiring complexity, and diameter of SCCN [15], by using these values and considering the total number of nodes of each level of SCCN, we can use Algorithm (3) to measure the value of SCEA of this network. Similarly, Algorithm (3) is used to calculate the SCEA of all interconnection networks which we have included in the comparison and tabulated in Tables 1, 2, The SCEA of SCCN and multiple networks is tabulated in Tables 1, 2, 3, and 4 and plotted in Figure (15).…”

“…The first concern in analyzing the topological features of a network is the static performance. If the network has good static performance, it will show perfect dynamic features [6] [15]. SCCN has been evaluated statically in previous work and it showed very good results compared to multiple networks [15].…”