International Journal of Communication Network Security ISSN: 2231 – 1882, Volume-1, Issue-4, 2012 45 EFFICIENT PROBE STATION PLACEMENT AND PROBE SET SELECTION FOR FAULT LOCALIZATION SHAKTI KINGER 1 & B.M. PATIL 2 1,2 Computer Science Dept., Maharashtra Institute of Technology, Pune 411038, INDIA. E-mail : shakti.kinger@mitpune.edu.in & balaji.patil@mitpune.edu.in Abstract – Network fault management has been a focus of research activity with more emphasis on fault localization – zero down exact source of a failure from set of observed failures. Fault diagnosis is a central aspect of network fault management. Since faults are unavoidable in communication systems, their quick detection and isolation is essential for the robustness, reliability, and accessibility of a system. Probing technique for fault localization involves placement of probe stations (Probe stations are specially instrumented nodes from where probes can be sent to monitor the network) which affects the diagnosis capability of the probes sent by the probe stations. Probe station locations affect probing efficiency, monitoring capability, and deployment cost. We present probe station selection algorithms and aim to minimize the number of probe stations and make the monitoring robust against failures in a deterministic as well as a non-deterministic environment. We then implement algorithms that exploit interactions between probe paths to find a small collection of probes that can be used to locate faults. Small probe sets are desirable in order to minimize the costs imposed by probing, such as additional network load and data management requirements. We discuss a novel integrated approach of probe station and probe set selection for fault localization. A better placing of probe stations would produce fewer probes and probe set maintaining same diagnostic power. We provide experimental evaluation of the proposed algorithms through simulation results. Keywords — Adaptive probing, Probe station selection, Fault diagnosis, Network monitoring, Probabilistic dependency model. I. INTRODUCTION With increasing complexity in computer networks, effective network management has become even more crucial and challenging. The network management aims at ensuring networks are monitored and kept running as smoothly as possible. Network monitoring generates huge information that needs to be processed and diagnosed to detect/localize the failure. This information is generated by either monitoring tools [1,2,3,4,5] or by network entities themselves (in the form of alarms) [6,7,8,9]. Fault Management system broadly deploys two types of monitoring (1) Active Monitoring – actively send probes to gather performance data (2) Passive Monitoring – rely on network devices to send alarms, as shown in fig-1. Both approaches have their own advantages and bear their own limitations. Combined, they are used to effectively solve network management problem. Figure 1 Fault management technologies A. Active Monitoring Active monitoring deploys probing methods to gather health status and performance statistics of network entities in the managed system. The main component of probing-based techniques is a sample measurement called probe. A probe is basically a dedicated program (such as ping or traceroute) or an application entity (such as email or web access). These probes are installed, sent and their results analyzed from network nodes called as probing station. A probe is periodically sent to examine a subset of network nodes in the managed system. Once a probe is sent to the network it either successfully returns to its probing station, signifying that all the network nodes in its path are in working order, or it fails to return to its probing station, indicating that one node or more in its path are in a failure state Probing based techniques have various advantages over passive monitoring techniques, such as (1) less instrumentation (2) capability to compute end-to-end performance (3) quicker localization, etc. Developing probing based monitoring solution involves solving two major problems, namely probe station selection and probe set selection. The probe station selection problem addresses the problem of selecting minimum subset of nodes in the managed network where probe stations should be placed such that the required diagnosis capability can be achieved through probes. Probe station selection is followed by task to select optimal probes such that any failure in network can be detected and localized. Different criteria’s are imposed on probe set