Szkaliczki, Tibor and Sobe, A and Elmenreich, W and Böszörmenyi, L (2013) Analysis of an Artificial Hormone System. In: Proceedings of the 8th Japanese-Hungarian Symposium on Discrete Mathematics and Its Applications, 2013-06-04 - 2013-06-07, Veszprém, Magyarország.

Preview	Text jh2013_ahs_v04.pdf - Accepted Version Download (195kB) | Preview
	Text Szkaliczki_473_2405077_ny.pdf Download (190kB)

Abstract

The increased complexity of modern networks and the increasingly dynamic access patterns in multimedia consumption have led to new challenges for content delivery. Dynamic networks and dynamic access patterns result in a complex system. To deliver content efficiently we introduced an artificial hormone system that is capable of handling the dynamics, is self-organizing, robust and adaptive. The content placement problem is NP complete and is closely related to several hard problems including edge-disjoint path routing, scheduling and the bin packing problem. The evaluation of self-organizing algorithms brings also a real challenge. For a first evaluation we created and ILP model of the problem. It is applied to get the exact optimum that serves as a bound in the evaluation of the solution algorithms. In this paper, we examine the convergence of the algorithm and found that the hormone levels converge to a limit at each node in the typical cases. We form a series of theorems on convergence with different conditions by starting with a specific base case and then we consider more general, practically relevant cases. The theorems can be proved by exploiting the analogy between the Markov chains and the artificial hormone system.

Update Item