The asymmetric multi-layer for distributed fault detection in networks with unreliable processors

Traditional centralized management solutions do not scale to present-day large-scale computer/communication networks. They suffer from certain other drawbacks too: a single point of failure and hence lack of fault tolerance, and heavy communication costs associated with the central manager. It has been recognized that the decentralization/distributed solutions can solve some of the problems associated with centralized solutions. A new theoretical multi-level paradigm called ML-ADSD (Thulasraman et al. (2003)) assumes that all nodes in the same cluster are symmetric for multi-level adaptive distributed diagnosis algorithms for fully connected networks which is not true and cannot be realized in real life. In this paper, we propose a new algorithm called asymmetric multi-level adaptive distributed diagnosis algorithms, AML-ADSD. In this new algorithm, node parameters are to be classified using a standard benchmark program. Then, each node's performance can be calculated.