Improvement of the CTI colleague network QoS by renewing its basic infrastructure Dr

Improvement of the CTI colleague network QoS by renewing its basic infrastructure

Dr. Ahmed A Alkhalifah
Assistant Professor, Computer Technology Dept. College of Telecommunication and Information,Riyadh, Saudi Arabi [email protected]

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

College of Telecommunication and Information (CTI) provides a highly technically trained cadre that would be able to find the best scientific solutions to the problems facing us and support the development process and capable of keeping pace with the fast global technological development1. As the use of CTI network and their applications increase, several deficiencies has been encountered such as network saturation, high bandwidth consumption. The current network is unable to support the new changes required by the IT rapid development and changes.

This paper is divided into two parts. We describe in the first part the global analytic that we have been made for the CTI Network by using monitoring software such as Nagios and Cacti. In the second part we describe different changes made on the network such as: installation of new core switch and new optical fiber links.
Keywords: College of Telecommunication and Information (CTI), NMS (Network Management System), Network architecture, Cisco Core switchs.
1. Introduction
College of Telecommunication and Information (CTI) systems is subject to random downtimes and performance degradation that have a tremendous impact on the quality of the services provided to employees and students in the educational workspace. To facilitate smoother operations of complex IT architectures, CTI needs a comprehensive network monitoring solution that ensures that everything stays in balance with optimal functionality. A network monitoring solution not only saves time and resources, but also reduces Data Center overhead and ensures network availability at all times.
In the following, we describe the uses of different monitoring tools to determine CTI Network defects.

Part I: Monitoring of the CTI Network by using Nagios and Cacti

This part is organized as follows. Section I presents a brief overview of CTI network technology, aSNMP based network management, and CTI network management as related work. Section II presents the design of common CTI MIB. It also presents the design and implementation of our CTI NMS. Section III presents the experimental result on the small CTI testbed.

I Overview of CTI Network

A- Infrastructure
The CTI technology is a corporate campus network serves to connect buildings including administrative, academic, college libraries, student centers, residence halls, gymnasiums, and other outlying structures, like conference centers, technology centers, and training institutes. CTI networks are normally interconnected with high speed Ethernet links operating over optical fiber such as Gigabit Ethernet as shown in Fig 1.

Fig 1. Overview of CTI High Level Network Diagram

Many tools have emerged to aid in performance monitoring of networks. The most common class of tools is based on the Simple Network Management Protocol (SNMP) 2, a protocol for sending and transmitting network performance information on IP networks. Other types of network performance monitoring tools include packet sniffers, flow monitors and application monitors. Examples of the various monitoring tools are SolarWind’s Orion SNMP monitoring platform, WireShark packet capture tool, Web metrics’GlobalWatch and Cisco’s NetFlow flow monitoring tools.

Currently, CTI network is composed of following infrastructure devices:

1) Cisco 4500 series switches (Core).
2) Cisco 3500 series switches (Distribution).
3) Cisco 2900 series switches (Access)
4) HP 1810 switches (Access)
5) Cisco 1900 series router (ISP and IPVPN)
6) Cisco 2800 series router (Voice Gateway)
7) Huawei VDI (Virtual Desktop Infrastructure System)
8) Windows 2012 Server (CCTV System)
9) Windows 2008 Server (Domain Controller)
10) Windows 2000 Server (Call Manager)
11) Aruba Network (Access Point)

Until now, Network devices such as Cisco switches have defined CTI MIBs and have an SNMP agent implemented on those devices, other devices are currently working towards defining CTI MIBs based on their SNMP supported version. Typical CTI networks will probably be composed of CTI devices from multiple vendors possibly developed using different SNMP version. However, in order for SNMP manager to manage these hybrid devices, it is desirable to have a common and standard MIB defined and implemented in all of these devices.

The current work consists on defining a common CTI MIB. We have developed a CTI network management system using Cacti and Nagios based on this common CTIMIB. This system is able to help the administrator to detect the CTI network faults, recover them, and configure the network, and monitor the performance of the network.
We have focused our development based not on any specific CTI device vendor button general CTI network management using the common CTIMIB definition. And also, we have installed CTI devices and composed the small CTI network. We have tested our proposed system on it.

In the next section, we propose a free and open source application which also focuses on system and network performance which are Cacti and Nagios.

B. CTI Network Management

Currently CTI are using the secured version of SNMP which is the Version 3.
The two applications that monitor now the system or network of CTI were CACTI and Nagios.

• Cacti is an open source web based network and system monitoring graphing solution for IT infrastructure management. Cacti allow an administrator to poll services at regular intervals to create graphs on the resulting data. Commonly, it is used to graph time-series data on metrics such as network bandwidth utilization, running processes, CPU-Memory-Disk usage6.

Nagios is a free and open source computer-software application that monitors systems, networks and infrastructure. Nagios offers monitoring and alerting services for servers, switches, applications and services. It alerts users when things go wrong and alerts them a second time when the problem has been resolved 7.

II. Design and Implementtaion

A. Design

We designed our CTI-NetMon to support the CTI network as shown in Fig 3. It is not a general network topology, but only an example of the CTI network, which we would like to manage by our proposed system. This CTI network is composed of Enterprise Network devices. We have been interested in the management of these CTI devices using SNMP. This system manages the faults, performance, and configuration of main NMS features.

Fig 2.Propose CTI Network Architecture

B. Implementation
We have established our CTI-NetMon part using LAMP (Linux-Apache-MySQL-PHP) solution. SNMP version 3 has been chosen to add cryptographic security 8 9.

The following packages were installed on POC machine.

• Apache: A Web server to display network graphs created by PHP and RRDTool.
• MySQL: A Database server to store cacti information.
• PHP: A script module to create graphs using RRDTool.
• PHP-SNMP: A PHP extension for SNMP to access data.
• NET-SNMP: A SNMP (Simple Network Management Protocol) is used to manage network.
• RRDTool: A database tool to manage and retrieve time series data like CPU load, Network Bandwidth etc.9.

III. Experiment

A. Experiment Environment Setup
We have made the SNMP Manager on POC (Proof of Concept) machine as described in Fig 3-1. We have installed a Cacti Version 0.8.8h and Nagios Core Version 4.1.1.

Fig 3-1. Proof of Concept of CTI-NetMon

The NMS Agent was configured on Network switches using SNMP version 3 as shown in Fig 4-2.

Fig 3-2. Configuration example on network switches

B. Experiment Result
To be more proactive on our daily routine, we put a four big screen (43 inches TV) on our location with speaker to hear the alarm if there is a down device or services that are having an issue.

Fig 3-3A and Fig 3-3B shows the sample Tactical Overview of Nagios Core and Nagios XI.

Fig 3-4 shows the preview of Graphs plugin which allows you to view the actual inbound and outbound traffic utilization.

Part II: Monitoring of the CTI Network by using Nagios and Cacti

This part is organized as follows. Section I presents a brief overview of the new core switches. Section II presents the design of the new transmission cables.

I. Deploy new core switches

CTI Network topology is based on traditional datacenter network infrastructure, it consists on a three-layer hierarchical model, which is called the “hierarchical inter-networking model”. It consists of core layer switches which connect to distribution layer switches, which in turn connect to access layer switches.

Figure 4-1: Basic Network Technology

Using our monitoring tools and experimental tests done by our network managers and describe previously, we noticed that the previous infrastructure suffers from:

– Network Congestion making a traffic slow-down,
– Massive use of bandwidth even outside of peak working hours
– Loss of the internal connection several times
– Problems of data storage and data recovery
– Defeat of some servers without clear cause
-Weakness of the internet connection and sometimes total disconnection
– Difficulties in integration of new applications
– Difficulties in connecting virtual machines such as VDI

Figure 4-2 represents the basic infrastructure of the CTI. This
Given the results obtained before, we decided to change a large part of our network. we started in the first phase by changing the architectural levels like the distribution of the main devices (servers, routers and switches) and in the second phase we changed completely the infrastructure of our network by replacing cables with optical fiber capable.

Figure 4-2 basic CTI network infrastructure

Figure 4-3: Old core swiths

Figure 4-4 : New core swithes
Feature of the new core switches 10 :
Cisco Catalyst 6509-E Switch
• 9 slots, 14 RU
• 32 x 40 GE ports
• 130 x 10 GE ports
• 387 x 1 GE ports
• 385 x 10/100/1000 ports
• Forwarding capacity up to 510 Mpps

II. Deploy new Fiber infrastructure

Figure 5-1 reprsents the old cabling infrastructure

We began the planning process by segmenting devices and equipment into hierarchical functions. This identifies type of operations and defines logical plant network segmentation for each level.

Our CTI campus network can greatly simplify network operation, optimize application performance, and build resilience to operate networks in deterministic order during various types of planned and unplanned outages. This paper limits the focus to construct a solid foundation and infrastructure between campus access, distribution, and core-layer systems. It covers the right set of recommendations to be applied on various types of platforms based on their roles in the network.
Before After
Net speed (Mbit/s) 12 54
Down time number Once per month No down time
Switches congestion twice per month No congestion
Recover convergence very time low high


In this paper, we concisely introduce in a first part the CTI infrastructure and encouraged the need for CTI Network Management System. We have also presented the design of common architecture of SNMP-based network management, design and implementation of CTI NMS called NetMon for the hybrid CTI networks. We also installed Cacti and Nagios in order to test our proposed system on the POC machine where we installed CTI NetMon for ourselves and presented the test result on the part of the network management. In a second part, we have described the steps taken to improve our network infrastructure, in fact we have tried to sweep the limits and drawbacks of our Network infrastructure by replacing the main devices distribution like core swithes and also by adopting a new architecture, which consisted of dividing the network into small cells and completely changing the transmission media with more consistent optic fibers.
This work is a description of a set of steps and decisions we took to improve the performance of
our network. This work could serve as a manual for other institutions that want to improve their networks


This research was supported by General Administration for Research and Studies in TVTC Corporation, member of Research Team – Dr. Mohsen DENDEN and by the System/Network Engineering Team support program supervised by the Head– Engr. Fahad Aljomah.


1 College of Telecom ; Information in Riyadh : : Brief Introduction

2 RFC3411 D. Harrington, R. Presuhn, B. Wijnen. “RFC 3411: An Architecture for Describing Simple Network Management Protocol (SNMP) Management Frameworks” IETF, Network Working Group, December,2002.

3 A Simple Network Management Protocol (SNMP), RFC1157

4 The Internet Engineering Task Force (IETF)

5 Management Information Base for Version 2 of the Simple Network Management Protocol (SNMPv2), RFC1907