Network Monitoring

CHAPTER I: INTRODUCTION

1.1. Introduction

The performance of communications networks can be monitored intrusively by transmitting test packets lost or duplicated may provide an indication of the performance of the links being monitored. Intrusive network monitoring, even at modest sampling rates, generates a significantly large amount of primitive data. This data needs to be summarized appropriately to address the required network performance analysis operations. These operations are generally iterative in nature, where the results of initial processing and analysis generate requirements for further processing and analysis.

Previously a group of developers from Universiti Kuala Lumpur has came up with a solution to the problem by building their own network monitoring system called Server Monitoring Real Time (SMRT). It is now in the second version. It has five main features to monitor the network which are Bandwidth, CPU Health, Port Scanner, Simple Network Management Protocol and Telnet or Remote Desktop. However, these previous system applications are monitoring based on standalone IP range but in this system, developers used the application based on Local Area Network cloud.

Therefore, developers want to enhanced this project by develop the tool of network monitoring performance based on network cloud LAN (Local Area Network) which is called LAN IP Monitor. The system should be able to monitor the computer network performance for the server in Local Area Network cloud. It consists of monitoring user activities, how much traffic can be detected and packet loss. It also can give information such as Operating System of the computer and know which computer is currently running on that time.

1.2. Purpose and Objectives

The objective of this project is to build tools that will used to monitor the computer network performance for Local Area Network. Developers used this system to monitor the performance of each computer in lab Level 4, 5 or 6 based on range of IP.

The analysis consists of:

i) How many network connection that are available and not available that can be detected.

ii) Based on the selected IP, the users can get the host details, address of the IP (IPv4 and MAC) and host status (Open, Closed and Scanned ports)

iii) The type of packets; packets that comes in and out (which can be read in character and hexadecimal packet)

iv) The graph of bandwidth detection (Time [s] VS Traffic [kB]).

v) To monitor user online activities such as what is the user applications.

vi) Error and collision detection.

The purpose of adding the module is to provide monitoring ability for the administrator of a network as well as to produce an effective network monitoring system.

1.3. Scope of Project

LAN IP Monitor is developed to monitor the network performance in Local Area Network (LAN). Currently, it is running in Lab Level 6 in UniKL and limited only in LAN. The software is using Windows and focusing more on wired platform and not over wireless connection. The scope of this project is to monitor network performance based on IP range.

1.4. Problem Statement

LAN IP Monitor is an application used by network administrators to monitor computer network performance based on IP in LAN cloud. Previous developers monitored their computer performance based on stand alone computer but in this project, the developers monitor the computer performance in LAN cloud. This tool is only commercial for Windows and not compatible with other operating system.

1.5. Project Rationale

This project is required as university purpose. The developers try to develop their own product and not depend on other products which are more expensive and this will cut cost for university.

LAN IP Monitor is an important tool to monitor network computer performance. If network problems occur, administrator can identify the source of the problems. That is why the developer feels that is it important to develop the LAN IP Monitor tool.

1.6. Limitation of Project

There are several limitations on developing the project.

* The system can only managed and handle in Local Area Network that consist of 100 to 120 computers.

* This tool cannot capture the extra sources of delay successfully.

* It is only compatible for Windows and not compatible with other operating system.

* This project is based on wired platform and not over wireless connection.

* Currently there are only two developers working on this project.

CHAPTER II: LITERATURE REVIEW

2.1. Survey of Network Performance Monitoring Tools

Various network performance monitoring tools have been developed and implemented over the last few years. These tools allow us to test different aspects of network performance and they are developed by individuals, commercial companies, non profit organizations, and government organizations. Some of them are free, open source, while some of them are not.

The following table lists network performance monitoring tools that have been developed and implemented between 1996 and 2006.

Table 1: Network performance tools developed between 1996 and 2006

Year Name of Tools

1996 Mtrg, NetNow, NetraMet, Network Probe Daemon, InterMapper, Lachesis, Optimal Networks, Digex.

1997 INS Net Perf Mgmt survey, tcpspray, Mapnet, Keynote, prtraceroute clflowd flstats, fping, tcpdpriv, NetMedic Pathchar, CAIDA Measurement Tool Taxonomy, bprobe & cprobe.

1998 NetOps, Triticom, Maple, PV-Wave, S-Plus, VisualRoute.

1999 Cheops, Ganymede, hping2, Iperf, JetMon, MeasureNet, MatLab, MTR, NeoTrace, Netflow, NetLogger, Network health, NextPoint, Nmap, Pchar, Qcheck, SAA, SafetP, Sniffit, SNMP from UCSD, Sting, ResponseNetworks, Tcpshow, Tcptrace WinTDS.

2000 Analyzer, bbftp, Big Brother, Bronc, Cricket, EdgeScape, Ethereal (now renamed Wireshark), gen_send/gen_recv,