Lan Modeling Essay
This project is to construct the model of WAN for a company on along the east coast. This company has offices in Atlanta, Philadelphia, New York and Boston. Headquarter of theoffice is located Washington D. C. which is the central network. As every office uses phone lines to connect to each other, so there is delay occur due to the unrelated traffic on the lines. Due to the problem, this company need to determine how the background traffic is affecting the FTP traffic on the network.
So, this analysis is to compare the FTP performance on the network under two scenarios. First is network without background traffic and, second is the scenario with background traffic. This WAN network contains 5 majors nodes which represent 5 location along the east coast. The network is connected by 10 Base-T. Inside each node, it contains subnet LAN network and main router of each location, except in Washington D. C. , which is headquarter, besides from LAN network and router, there is one main server located in the location as well
However, even though there are detail of subnet LAN network in each location, because this analysis is focus on WAN, so it is not necessary to consider detail in each location. Each office location will be considered as one node. So there are 5 nodes considered in this analysis. Point to Point utilization Analysis This first analysis focuses on the usage of network that compares 2 scenarios as mention earlier. The red graph show usage rate of network when there is no other back load in the network.
And blue graph represent otherwise. The result shows a distinctive different between these 2 scenarios. With back load, average utilization rate goes up to almost 11 at the 25th minute. On the other hand, without back load, the utilization rate is relatively low. So, from this simulation, back load shows a distinctive affect on the overall WAN. FTP Download Response Time Another perspective of simulation upon these 2 scenarios is by testing the FTP response time. Graph also shows a distinctive different between the 2 scenarios.
With back load (blue graph) the FTP takes almost up to 7 sec to response, while without back load, response time is only up to 2. 5 second. So, this FTP download response time simulation, it also appears that back load affects this WAN capacity. Conclusion As the results shows a distinctive affect of back load data on both simulation, so it can be concluded that the initial assumption of network delay caused by unrelated traffic on the lines it true. And it also causes a great affect on the network performance.
Business Application We apply this project to actual organization in United State which is the Royal Embassy and Consulates of Thailand in United State. The Royal Embassy of Thai is located in Washington D. C. along with other four Thailand Consulate offices which are located in New York, Chicago, Miami and Los Angeles. The Washington D. C. embassy is considered as headquarter among all offices. The projects will be applied to the setup of WAN network of Thailand Embassy with all the consulate offices. Scenario 1
As shown in figure 6, this network model shows design to have every consulate office connect directly in the embassy in Washington D. C. The result shows that point to point connection from Washington D. C. to New York, Miami and Chicago run at a similar throughput and utilization except Washington D. C. to Los Angeles that appear to have a significantly lower rate than other. So, from analyzing the physical condition of these consulates, it appears that Los Angeles has the longest distance to Washington D. C. than other location.
So, it can be concluded that distance affect the quality of network. Scenario 2 As we see in scenario 1 that distance from D. C. to LA affects the quality of network significantly. So We design to add connection point between the route from Los Angeles to Washington D. C. The connectors are located in Santafe, Oklahoma City and Nashville. So, Los Angeles connects to Santafe to Oklahoma City, Nashville to and Washington D. C. respectively in order to amplify data signal from Los Angeles to Washington D. C. Yellow graph represents the connection between Los Angeles to Washington D.
C. directly without connector (Scenario 1) that compare with other graphs in scenario 2 Dark Blue = Nashville to Washington D. C. Red = Oklahoma City to Nashville Green = Santafe to Oklahoma CityLight Blue = Los Angeles to Santafe It appears that in scenario 2, the result shows a better connection rate than in Scenario 1. The closer to Washington D. C. , the higher connection rate. But connection rate from Los Angeles to Santafe which is the connecting segment with most distances from Washington D. C. still has a relatively low rate (still higher than in scenario 1)
So, from this experiment it can be concluded that, in business application to Thailand Embassy and Consulate Office, establishing connectors in different city along the connection line can help improve the signal strength, connection quality and eliminate the distance factor that causes network quality. Even though in some segment of the area, the connection speed and connection quality is low (the segment from Los Angeles to Santafe), the overall network quality is still better than without having connector point as in Scenario 1.