TY - GEN
T1 - Faulty links optimization for hypercube networks via stored and forward one-bit round robin routing algorithm
AU - Rizvi, Syed S.
AU - Sheikh, Adil
AU - Riasat, Aasia
N1 - Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2010
Y1 - 2010
N2 - Extensive research and studies have shown that the hypercube network is one of the most widely used techniques to build high performance parallel computers and it offers strong hierarchical structure with high symmetry properties. However, the performance of hypercube networks is mainly dependent on the link consistency between the nodes. Due to unexpected links failure and low connectivity, neither the hardware schemes nor point to point and multistage routing algorithms can be used without adding extra links. This paper presents a new stored and forward one-bit round robin routing algorithm that can efficiently send and receive messages between the nodes even in the presence of faulty links. In addition, we also examine and simulate the data throughput and hot spot properties of a hypercube network. Simulation results demonstrate that the proposed routing algorithm improves the overall performance of hypercube network by means of load-scalability that not only reduces the average waiting time per output queue but also increases the total data throughput even in the presence of faulty nodes.
AB - Extensive research and studies have shown that the hypercube network is one of the most widely used techniques to build high performance parallel computers and it offers strong hierarchical structure with high symmetry properties. However, the performance of hypercube networks is mainly dependent on the link consistency between the nodes. Due to unexpected links failure and low connectivity, neither the hardware schemes nor point to point and multistage routing algorithms can be used without adding extra links. This paper presents a new stored and forward one-bit round robin routing algorithm that can efficiently send and receive messages between the nodes even in the presence of faulty links. In addition, we also examine and simulate the data throughput and hot spot properties of a hypercube network. Simulation results demonstrate that the proposed routing algorithm improves the overall performance of hypercube network by means of load-scalability that not only reduces the average waiting time per output queue but also increases the total data throughput even in the presence of faulty nodes.
UR - http://www.scopus.com/inward/record.url?scp=84878956685&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84878956685&partnerID=8YFLogxK
U2 - 10.1007/978-90-481-3662-9_48
DO - 10.1007/978-90-481-3662-9_48
M3 - Conference contribution
AN - SCOPUS:84878956685
SN - 9789048136612
T3 - Novel Algorithms and Techniques in Telecommunications and Networking
SP - 281
EP - 286
BT - Novel Algorithms and Techniques in Telecommunications and Networking
PB - Kluwer Academic Publishers
T2 - 2008 International Conference on Telecommunications and Networking, TeNe 2008, Part of the International Joint Conferences on Computer, Information, and Systems Sciences, and Engineering, CISSE 2008
Y2 - 5 December 2008 through 13 December 2008
ER -