DMR: A Novel Routing Algorithm in 3D Torus Switching Fabric

Jianbo Guan (), Xicheng Lu and Zhigang Sun
Additional contact information
Jianbo Guan: National University of Defense Technology, School of Computer Science
Xicheng Lu: National University of Defense Technology, School of Computer Science
Zhigang Sun: National University of Defense Technology, School of Computer Science

A chapter in Current Trends in High Performance Computing and Its Applications, 2005, pp 299-303 from Springer

Abstract: Summary In recent years 3-D Torus network is widely considered in implementing large scale switching fabrics. Based on analysis of several problems faced in packets routing in 3-D Torus switching fabrics, a novel routing algorithm called DMR (Dimension-order-based Multi-path Routing) is proposed. The DMR algorithm is discussed in detail and it is proved that DMR can achieve high throughput by balancing traffic loads on multiple equivalent paths while maintaining packets order in one TCP flow. The performance of DMR routing algorithm is evaluated using a simulation approach and it is compared with two other representative routing algorithms: e-cube routing and random routing. The simulation results and the analysis are shown afterwards. At last a conclusion is given.

Keywords: torus; switch; routing algorithm (search for similar items in EconPapers)
Date: 2005
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-540-27912-9_35

Ordering information: This item can be ordered from
http://www.springer.com/9783540279129

DOI: 10.1007/3-540-27912-1_35

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().