Distributed Systems Diagnosis Using Belief Propagation
نویسنده
چکیده
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.
منابع مشابه
Mathematical Programs for Belief Propagation and Consensus
This paper develops methods of distributed Bayesian hypothesis tests for fault detection and diagnosis that are based on belief propagation and optimization in graphical models. The main challenges in developing distributed statistical estimation algorithms are i) difficulties in ensuring convergence and consensus for solutions of distributed inference problems, ii) increasing computational cos...
متن کاملSelf-healing in large-scale systems: parallel and distributed diagnostic architectures
Automated real-time problem diagnosis is a key feature of a self-healing system. However, rapidly growing size and complexity of modern distributed systems creates a challenge for traditional centralized diagnostic approaches and calls for parallel and distributed architectures. Dividing the system into subsystems controlled by separate diagnostic engines is an obvious choice; however, on top o...
متن کاملInformation-theoretic approaches to cost-efficient diagnosis
This paper provides a summary of our recent work on cost-efficient probabilistic diagnosis in Bayesian networks with applications to fault diagnosis in distributed computer systems. We focus on achieving good trade-offs between the diagnostic accuracy versus the cost of testing and computational complexity of diagnosis. We present (1) theoretical results characterizing these trade-offs, such as...
متن کاملUsing Adaptive Probing for Real-Time Problem Diagnosis in Distributed Computer Systems
In this work, we focus on cost-efficient techniques for realtime diagnosis in distributed systems that allow an adaptive, on-line selection and execution of appropriate measurements (tests). Particularly, one of our applications concerns fault diagnosis in distributed computer systems and networks by using test transactions, or probes (e.g., ”traceroute” or ”ping” commands). The key efficiency ...
متن کاملOptimization of Inter-Subnet Belief Updating in Multiply Sectioned Bayesian Networks
Recent developments show that Multiply Sectioned Bayesian Networks (MSBNs) can be used for diagnosis of natural systems as well as for model-based diagnosis of artificial systems. They can be applied to single-agent oriented reasoning systems as well as multiagent distributed reasoning systems. Belief propagation between a pair of subnets plays a central role in maintenance of global consistenc...
متن کامل