Fault-Tolerance and Permutation Analysis of ASEN and its Variant
Journal Title: International Journal of Computer Science and Information Technologies - Year 2010, Vol 1, Issue 1
Abstract
High performance computing systems can be designed using parallel processing. The effectiveness of these parallel systems rests primarily on the communication network linking processors and memory modules. Hence, an interconnection network that provides the desired connectivity and performance at minimum cost is required. The design of a suitable interconnection network for inter-processor communication is one of the key issues of the system performance. In this paper a new multistage interconnection network IASEN (Irregular augmented shuffle exchange) has been proposed modifying existing ASEN-2 network. ASEN-2 is a regular multipath network with limited fault tolerance. The paper also discusses the permutation possibility behavior of both the networks with and without faults. It has been observed that the proposed multistage interconnection network IASEN provides much better fault-tolerance by providing more paths between any pair of source-destination as compared to ASEN-2 at the expanse of little more cost.
Authors and Affiliations
Rinkle Rani Aggarwal , Dr. Lakhwinder Kaur
Keywords
Related Articles
- EP ID EP155217
- DOI -
- Views 91
- Downloads 0
An Enhancing approach in MEDLINE & PubMed using Text Mining
Text mining is a new and exciting research area that attempts to solve the information overload problem. Information extraction using text mining has been applied to many different areas of bio-medici...
Operating System based Compliance Validation of Trusted Computing
The concept of trusted computing given was by Anderson [2]. Trust is an expectation that a device behaves in a particular manner. A trusted component, operation or process is one whose behavior is pr...
Novel Implementation of Text Mining for Reports
In this paper, we propose a text mining system to extract and use the information in radiology reports. The system consists of three main modules: medical finding extractor, report and image retriever. The medical findin...
Four Walsh Transform Sectors Feature Vectors for Image Retrieval from Image Databases
The work of this paper presents the new idea of using Walsh transform to generate the feature vector for content based image retrieval. The complex Walsh transform is conceived by multiplying Sal functions...
A Two Stage Language Independent Named Entity Recognition for Indian Languages
This paper describes about the development of a two stage hybrid Named Entity Recognition (NER) system for Indian Languages particularly for Hindi, Oriya, Bengali and Telugu. We have used both statistical Maximum Entrop...