The role of agents in enterprise system management: a position paper

ABSTRACT

The size and complexity of enterprise computer systems are growing rapidly. As a result, system management has become increasingly difficult and expensive. In fact, management costs are typically estimated at 50%-70% of the total cost of ownership. Despite large investments in management software and personnel, enterprise computer systems are usually managed sub-optimally. This situation calls for a fundamental change in the way systems are managed.Recent studies suggest that systems manage themselves autonomously. Initial studies towards that goal reside in the fields of Autonomic Computing and Self* Systems. Many of those studies devise mechanisms for system monitoring, data filtering, problem determination, problem fixing and system adjustment and reconfiguration. The mechanisms are based, in large part, on methods from Artificial Intelligence, Statistics, Operations Research and Software Engineering.Current enterprise system management solutions --- theoretical and practical --- are centralized. However, the typical enterprise system is distributed. We claim that a centralized solution presents several risks. For instance, it introduces a management bottleneck, it incurs communication overheads, etc. We hence advocate that enterprise system management should be addressed in a distributed manner. Combining distribution and Al, we believe that an agent-based solution is appropriate. In this paper we present in more detail the suggested approach.

REFERENCES

• 1: Breitgand, D. Henis, E. Ladan-Mozes, E. and Yerushalmi, E. Root-Cause Analysis of SAN Performance Problems: an I/O Path Affine Search Approach. In IFIP/IEEE 9th International Symposium on Integrated Network Management, IM'05, Nice, France, May 2005.
• 2: Dilman, M. and Raz, D. Efficient Reactive Monitoring. IEEE Journal on Selected Areas in Communications (JSAC), Special Issue on Recent Advances in Network Management, 20(4).
• 3: A. G. Ganek , T. A. Corbi, The dawning of the autonomic computing era, IBM Systems Journal, v.42 n.1, p.5-18, January 2003
• 4: Joseph L. Hellerstein , Fan Zhang , Perwez Shahabuddin, A statistical approach to predictive detection, Computer Networks: The International Journal of Computer and Telecommunications Networking, v.35 n.1, p.77-95, Jan. 2001 [doi>10.1016/S1389-1286(00)00151-1 ]
• 5: Klaus Herrmann , Gero Muhl , Kurt Geihs, Self-Management: The Solution to Complexity or Just Another Problem?, IEEE Distributed Systems Online, v.6 n.1, p.1, January 2005 [doi>10.1109/MDSO.2005.3]
• 6: Jeffrey O. Kephart , David M. Chess, The Vision of Autonomic Computing, Computer, v.36 n.1, p.41-50, January 2003 [doi>10.1109/MC.2003.1160055 ]
• 7: Kar, G., Kochut, A., Bobroff, N. and Beaty, K. Management Issues in Storage Management Networks: Detection and Isolation of Performance Problems. In IFIP/IEEE 9th International Network Operation and Management Symposium (NOMS'04), Seol, Korea, March 2004.
• 8: Proc. of SELF-STAR: International Workshop on Self-* Properties in Complex Information Systems, 31 May - 2 June 2004, University of Bologna, http://www.cs.unibo.it/self-star/program.html
• 9: Andrew S. Tanenbaum , Maarten Van Steen, Distributed Systems: Principles and Paradigms, Prentice Hall PTR, Upper Saddle River, NJ, 2001
• 10: Turing, A. M. Intelligent Machinery, 1948, Report for National Physical Laboratory, in Machine Intelligence 7, Meltzer, B. and Michie, D., Eds. 1969.
• 11: John Von Neumann , Arthur W. Burks, Theory of Self-Reproducing Automata, University of Illinois Press, Champaign, IL, 1966
• 12: http://www-1.ibm.com/linux/whitepapers/robertFrancesGroupLinuxTCO Analysis05.pdf

INDEX TERMS

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

Additional Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
        I.2.11 Distributed Artificial Intelligence

General Terms:
Management, Performance, Reliability