Monitoring overhead in distributed systems: visualization and estimation techniques

ABSTRACT

Monitoring distributed systems involves the collection, analysis, and display of interactions among managed objects. These functions are carried out by the components of a monitoring system, such as management agents.During monitoring, resources in the system are shared between the monitoring system components and the monitored distributed system. Thus, the monitored system has to maintain its functionality with fewer resources. This eventually affects the performance of the monitored system, by causing unwanted effects such as slow response time from processes and delayed access time to devices.Furthermore, ineffcient configuration of the monitoring system components will have additional negative effects on the monitored system components. The combination of all effects of monitoring on the monitored distributed system is referred to as monitoring overhead.To use the resources in the system efficiently, one has to select a configuration with minimal monitoring overhead. However, due to the nature of distributed systems, measuring the monitoring overhead on a system is not a simple task.In this paper, we use techniques to visualize and estimate monitoring overhead. We describe our experiment and the techniques used to analyze the data and perform the measurements.

REFERENCES

• 1: {1} H. Lutfiyya, M. A. Bauer, M. Wei, and J. W. Hong. Monitoring in Distributed Application Management. 6th International Conference on Parallel and Distributed Computing and Systems, 1993.
• 2: German Goldszmidt , Yechiam Yemini , Shaula Yemini, Network management by delegation: the MAD approach, Proceedings of the 1991 conference of the Centre for Advanced Studies on Collaborative research, October 28-30, 1991, Toronto, Ontario, Canada
• 3: Jeffrey Joyce , Greg Lomow , Konrad Slind , Brian Unger, Monitoring distributed systems, ACM Transactions on Computer Systems (TOCS), v.5 n.2, p.121-150, May 1987 [doi>10.1145/13677.22723]
• 4: D. M. Ogle , K. Schwan , Richard Thomas Snodgrass, Application-Dependent Dynamic Monitoring of Distributed and Parallel Systems, IEEE Transactions on Parallel and Distributed Systems, v.4 n.7, p.762-778, July 1993 [doi>10.1109/71.238299]
• 5: Richard Snodgrass, A relational approach to monitoring complex systems, ACM Transactions on Computer Systems (TOCS), v.6 n.2, p.157-195, May 1988 [doi>10.1145/42186.42323]
• 6: William S. Cleveland, Visualizing Data, Hobart Press, 1993
• 7: {7} R. Jain. The Art of Computer Systems Performance Analysis. John Wiley and Sons, Inc., 1991.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Measurement techniques

Additional Classification:
  C. Computer Systems Organization
  C.4 PERFORMANCE OF SYSTEMS
      Subjects: Modeling techniques

  D. Software
  D.2 SOFTWARE ENGINEERING
      D.2.5 Testing and Debugging
          Subjects: Monitors

  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.4 Systems and Software
          Subjects: Distributed systems

General Terms:
Algorithms, Design, Experimentation, Measurement, Performance, Theory