Network Monitoring System Team 1 Chadalavada Subrahmanya Jayaraman

Network Monitoring System Team# 1 Chadalavada Subrahmanya Jayaraman 916 -89 -1611 Alluri Vinod Harish Kumar 916 -64 -2451 Nanduri Sai Vasanth 916 -51 -7021 1

Real Time Network 2

Introduction: • Network monitoring is an essential responsibility for businesses of all sizes. This network monitoring involves multiple methods which are deployed for the purpose of maintaining the security and integrity of the network. 3

Techniques in Network Monitoring System • SNMP – Simple Network Management Protocol • RMON – Remote Monitoring • NETFLOW – Network Flow RFC 3954 4

SNMP • Simple Network Management Protocol short for SNMP is an agent/manager based model. The model contains an agent, a manager and a database which holds the information about the management and the network protocols. 5

SNMP Architecture 6

RMON • RMON short for Remote Monitoring system is for managing a network. Remote enables various monitors and console systems to exchange network monitoring data. Unlike SNMP this will send out a request for information. 7

RMON • RMON focus on IP traffic and application level traffic. There are two main components in the RMON environment. They are, ➢RMON probe ➢Management station / client 8

Net Flow RFC 3954 • Net flow is one of the network monitoring technique which has an ability to collect ip network traffic as it enters an interface through routers. • Administrator analysis the data provided by the net flow and determined things such as source and destination of traffic, class of service and also the cause of congestion. 9

Net Flow RFC 3954 • Net flow has three main components: ➢Flow catching ➢Flow collector ➢Data analyzer 10

Network Monitoring System Includes • Service Polling • Graphing • Notifications and Events 11

Service Polling • Service polling is a type of technique in network monitoring system were it checks regularly whether a device or a service is available or not. • This allows the NMS to check whether the server that is hosting the website is reachable over the network or not. 12

Graphing • This technique includes time series graphs of performance data which are useful for identifying trends and anomalies. • Such graphs are frequently used for identifying data values. 13

Notification & Events • When there is a change on the network, either its good or bad the change has to be notified. • This technique include events system which are used to notify and check some change, in a given threshold range. 14

Real Time Network Monitoring System ➢Cacti ➢Icinga ➢Smoke. Ping ➢Graphite ➢Reconnoiter 15

Cacti • Cacti is NMS designed for time series graph of performance data on a monitored network. • A graph can be drawn from multiple data sources, but requires some suitable template in cacti format. 16

Icinga • It is an NMS designed specifically for service polling notifications and report generation. • Icinga service polling is modular, each service check is handled by a separate process which is forked by Icinga monitoring system. 17

Smoke. Ping • This network monitoring system provides more specific monitoring, that is critical inside ISP’s to enable network administrator to identify faults and performance problems on network links. • Smokeping monitors latency on a network by sending periodic ICMP echo request packets to the machine at the other end of the network link. 18

Graphite • Graphite is a network monitoring system, written in python. • With the aim of being more scalable and offering more real-time graphing capabilities than existing NMS. 19

Reconnoiter • Reconnoiter is used to improve the network Monitering systems field by using faults detection and trending together to provide more information and insight than the current systems. 20

Limitations • The network monitoring systems are been currently used and have no of limitations imposed legacy tools and library that are used in construction of system. • One such limitation is inability to store large amount of data that has been collected by NMS. 21

Limitations: • This is because assumptions made about the data by the database format and due to the fixed size of the database. • NMS will only implement a subset of techniques to monitor network. • So, this makes to deploy multiple NMS to provide the whole network testing coverage. 22

Future Work • As a part of future work it aims to produce a system that we will call a Dynamic Network Monitoring System, where the emphasis is on dynamic, rather than statically, configured network models. 23