A recursive distributed topology discovery service for grid clients

Distributed application (e.g., grid-enabled application) performance can be improved by complementing the computational resource information advertised to the clients with network information (e.g., topology and link capacity). In this way, clients may choose the jointly optimal resources. This study focuses on how to provide clients with network information. The proposed method, namely the Recursive Distributed Topology Discovery Service (RD-TDS), utilizes a selected set of distributed beacons (i.e., measurement points) for network discovery. Each RD-TDS beacon recursively probes nodes that it has not discovered during its previous exploration steps but that have been discovered by other beacons. The exploration ends when each beacon does not have any new node to probe. The RD-TDS evaluation confirms the expected advantage of such method: closely approximating the full topology discovery with a limited number of active beacons. In addition, the proposed method rationale can be easily applied to many current network exploration tools.

[1]  Balachander Krishnamurthy,et al.  Internet Measurement - Infrastructure, Traffic, and Applications , 2006 .

[2]  Rajeev Rastogi,et al.  Robust Monitoring of Link Delays and Faults in IP Networks , 2003, IEEE/ACM Transactions on Networking.

[3]  A. Kuchar Achievements of COST 266 Action and further prospects in research of advanced infrastructure for photonic networks , 2004, Proceedings of 2004 6th International Conference on Transparent Optical Networks (IEEE Cat. No.04EX804).

[4]  Stefano Giordano,et al.  The Beacon Number Problem in a Fully Distributed Topology Discovery Service , 2007, IEEE GLOBECOM 2007 - IEEE Global Telecommunications Conference.

[5]  R. Rastogi,et al.  Robust Monitoring of Link Delays and Faults , 2006 .

[6]  Francis Lepage,et al.  Application-oriented network metrology: metrics and active measurement tools , 2005, IEEE Communications Surveys & Tutorials.

[7]  Benoit Donnet,et al.  Internet topology discovery: a survey , 2007, IEEE Communications Surveys & Tutorials.

[8]  Jean-Loup Guillaume,et al.  Relevance of massively distributed explorations of the Internet topology: Qualitative results , 2006, Comput. Networks.

[9]  Shueng-Han Gary Chan,et al.  Scalable and Efficient End-to-End Network Topology Inference , 2008, IEEE Transactions on Parallel and Distributed Systems.