STH: A highly scalable and economical topology for massively parallel systems

Jahangir Alam

doi:10.17485/ijst/2011/v4i12.24

Abstract

Highly parallel systems are receiving significant attention to solve the large and complex problems. This has resulted in the emergence of many attractive interconnection network topologies. This paper introduces a new processor interconnection topology called STH (Scalable Twisted Hypercube) to counter the poor scalability of twisted hypercube. Its suitability for use as multiprocessor interconnection networks has also been explored. The various properties of the proposed topology have been analyzed and it has been compared with some other highly scalable topologies of interest on a number of interconnection networks evaluation parameters. With reduced diameter, better average distance, low traffic density, low cost, maximum number of links, high bisection width and tremendous scalability, STH is more suitable for Massively Parallel Systems. Procedures for routing and broadcasting on the proposed topology have also been discussed and a simple routing algorithm has been presented. The proposed interconnection network provides a great architectural support for parallel computing due to the concurrent existence of multiple LST(m) and TQn.

Highlights

Introduction routing algorithm andChiu & Chon (1998) proposedHigh speed parallel computing is essential for modern efficient multicasting procedure for binary hypercube.research as the demand for more and more computing Klasing (1998) discussed efficient compression of CCC power is continuously increasing
This paper introduces and analyses a new hybrid processor interconnection topology called Scalable twisted hypercube (STH) (Scalable Twisted Hypercube)
We propose the following optimal (O(m+n) in worst case) algorithm for the required message routing

Summary

The message is now routed along the path

The diameter may further be (19, 6). Let node X wants to broadcast a message on STH (m,n) suffers form high latency. It is always desirable to have a network, such a broadcasting will be performed as topology with both small degree (low cost) and small follows(Table 3): diameter (low latency). 1. X will send message to all the nodes of that LST (m) network the product of degree (d) and diameter (D) is network on which X itself resides. 2. on every twisted hypercube of STH (m,n), the Another metrics used to describe the cost (ζ) of an node received the messages will send them to all of interconnection network topology is the product of the the nodes in the same twisted hypercube network

Comparative Study

No of Node s

Findings

Total No of