In this paper, we focus on diagnosis in distributed computer systems using end-to-end transactions, or probes. Diagnostic problem is formulated as a probabilistic inference in a bipartite noisy-OR Bayesian network. Due to general intractability of exact inference in such networks, we apply belief propagation (BP), a popular approximation technique proven successful in various applications, from image analysis to probabilistic decoding. Another attractive property of BP for our application is it natural parallelism that allows a distributed implementation of diagnosis in a distributed system to improve diagnostic speed and robustness. We derive lower bounds for diagnostic error in bipartite Bayesian networks, and particularly in noisy-OR networks, and provide promising empirical results for belief propagation on both randomly generated and realistic noisy-OR problems.
By: Irina Rish
Published in: RC23763 in 2005
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .