PROJECT TITLE:

Adaptive Algorithms for Diagnosing Large-Scale Failures in Computer Networks - 2015

ABSTRACT:

We tend to propose a greedy algorithm, Cluster-MAX-COVERAGE (CMC), to efficiently diagnose giant-scale clustered failures. We tend to primarily address the challenge of determining faults with incomplete symptoms. CMC makes novel use of each positive and negative symptoms to output a hypothesis list with an occasional number of false negatives and false positives quickly. CMC needs reports from concerning [*fr1] as many nodes as alternative existing algorithms to see failures with one hundred % accuracy. Moreover, CMC accomplishes this gain considerably faster (typically by two orders of magnitude) than an algorithm that matches its accuracy. When there are fewer positive and negative symptoms at a reporting node, CMC performs a lot of higher than existing algorithms. We have a tendency to conjointly propose an adaptive algorithm known as Adaptive-MAX-COVERAGE (AMC) that performs efficiently throughout each freelance and clustered failures. During a series of failures that include both freelance and clustered, AMC ends up in a reduced variety of false negatives and false positives.