Institutional Repository [SANDBOX]
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Efficiently monitoring bandwidth and latency in IP networks

Breitbart, Y, Chan Chee-Yong, Garofalakis Minos, Rastogi Rajeev, Silberschatz Avi

Full record


URI: http://purl.tuc.gr/dl/dias/2F9D4415-EB7F-49E3-B2C4-87DD564E6E20
Year 2001
Type of Item Conference Full Paper
License
Details
Bibliographic Citation Y. Breitbart, C-Y Chan, M. Garofalakis, R. Rastogi and A. Silberschatz, "Efficiently monitoring bandwidth and latency in IP networks", in Twentieth Annual Joint Conference of the IEEE Computer and Communications Societies, Apr. 2001, pp. 933-942. doi: 10.1109/INFCOM.2001.916285 https://doi.org/10.1109/INFCOM.2001.916285
Appears in Collections

Summary

Effective monitoring of network utilization and performanceindicators is a key enabling technology for proactive and reactive resourcemanagement, flexible accounting, and intelligent planning in nextgenerationIP networks. In this paper, we address the challenging problemof efficiently monitoring bandwidth utilization and path latencies in an IPdata network. Unlike earlier approaches, our measurement architectureassumes a single point-of-control in the network (corresponding to the NetworkOperations Center) that is responsible for gathering bandwidth andlatency information using widely-deployed management tools, like SNMP,RMON/NetFlow, and explicitly-routed IP probe packets. Our goal is toidentify effective techniques for monitoring (a) bandwidth usage for a givenset of links or packet flows, and (b) path latencies for a given set of paths,while minimizing the overhead imposed by the management tools on the underlyingproduction network. We demonstrate that minimizing overheadsunder our measurement model gives rise to new combinatorial optimizationproblems, most of which prove to be ✂✁-hard.We also propose novelapproximation algorithms for these optimization problems and prove guaranteedupper bounds on their worst-case performance. Our simulation resultsvalidate our approach, demonstrating the effectiveness of our novelmonitoring algorithms over a wide range of network topologies.

Services

Statistics