Abstract
Discovering a network topology and inferring its performances for the client/server case is a well known field of study. However, client/server model is no longer accurate when dealing with Grids, as those platforms involve coordinated transfers from multiple sources to multiple destinations. In this chapter, we first review existing work, introduce a representation of the inferred knowledge from multiple sources and multiple destinations measurements that allows to obtain a well-posed problem, algorithms in order to reconstruct such a representation, a method to probe network, and give some experimental results.
Reference42 articles.
1. A Factor Analytic Approach to Inferring Congestion Sharing Based on Flow Level Measurements
2. Bestavros, A., Byers, J., & Harfoush, K. (2001). Inference and labeling of metric-induced network topologies. (Technical Report BUCS-TR-2001-010), Boston University, Computer Science Department.
3. Inference and labeling of metric-induced network topologies
4. Bobelin, L., & Muntean, T. (2007). Metric induced network poset (MINP): A model of the network from an application point of view. In First International Conference on Networks for Grid Applications (GridNets), MetroGrid Workshop, October 2007.
5. Bobelin, L., & Muntean, T. (2008). Algorithms for network topology discovery using end-to-end measurements. In International Symposium on Parallel and Distributed Computing, 47, (pp. 267–274).