Advertisement

Abstract

We introduce A-GAP, a protocol for continuous monitoring of network state variables with configurable accuracy. Network state variables are computed from device counters using aggregation functions, such as SUM, AVERAGE and MAX. In A-GAP, the accuracy is expressed in terms of the average error and is controlled by dynamically configuring filters in the management nodes. The protocol follows the push approach to monitoring and uses the concept of incremental aggregation on a self-stabilizing spanning tree. A-GAP is decentralized and asynchronous to achieve robustness and scalability. We provide some results from evaluating the protocol for an ISP topology (Abovenet) in several scenarios through simulation. The results show that we can effectively control the fundamental trade-off between accuracy and overhead. The protocol overhead can be reduced significantly by allowing only small error objectives.

Keywords

Span Tree Aggregation Function Accuracy Objective Filter Width Management Node 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Gonzalez Prieto, A., Stadler, R.: Distributed Real-time Monitoring with Accuracy Objectives. KTH Technical Report (December 2005)Google Scholar
  2. 2.
    Olston, C., Loo, B.T., Widom, J.: Adaptive Precision Setting for Cached Approximate Values. In: ACM SIGMOD 2001, Santa Barbara, USA (May 2001)Google Scholar
  3. 3.
    Dam, M., Stadler, R.: A Generic Protocol for Network State Aggregation. In: Radiovetenskap och Kommunication (RVK), Linkoping, Sweden, June 14-16 (2005)Google Scholar
  4. 4.
    Lim, K., Stadler, R.: SIMPSON — a SIMple Pattern Simulator for Networks (2005), http://www.comet.columbia.edu/adm/software.htm
  5. 5.
    Spring, N., Mahajan, R., Wetherall, D.: Measuring ISP topologies with Rocketfuel. In: ACM/SIGCOMM, Pittsburgh, USA (August 2002)Google Scholar

Copyright information

© IFIP International Federation for Information Processing 2006

Authors and Affiliations

  • Alberto Gonzalez Prieto
    • 1
  • Rolf Stadler
    • 1
  1. 1.School of Electrical EngineeringKTH Royal Institute of TechnologySweden

Personalised recommendations