Software Defined Networking vs. Traditional Networking Performance Analysis

December 5, 2016

Analysis of the Results

This part of the dissertation presents the analysis of the results, which were obtained from the undertaken experiments, and presented in the dissertation writing service. It should be noted that the purpose of this analysis remains to identify better and more effective means of networking among SDN and traditional networking, by not just focusing on the results but also interlinking the literature review. For the fulfillment of this purpose, firstly the efficiency of delivery of Bandwidth and data transfer by SDN and traditional networking was taken into consideration.

It was perceived from the literature review assignment writing service that the SDN has a high capacity of bandwidth, and this is the reason that SDN accomplished the bandwidth greater speed than traditional networking (Carroll 2013). To understand this better, later experiment were included in the calculation of the efficacy of the bandwidth, for all five clients connected to five servers and basing it on three switches topology in 30 seconds. The results section above then approved the consideration of the literature and it was found that the performance of the SDN is better, as it takes fewer megabits per second in connecting the five clients to five servers as compared to the time taken by the traditional networking (Shalimov, ‎‎2013).

Appropriate date centers and also effective data transferring two more attributes, which were discussed in the lieu of indicating that SDN has a better positioning in the networking provision in the essay writing service. Moreover, the OpenFlows were conditioned to ensure the flow of the data in a speedy manner, along with the provision of appropriate data centers (Chauhan, 2013) ‎. While looking into the results it was found that there is a high efficacy of the data transferring to five clients connected to five servers shown by SDN, keeping in view that it is based on three switches topology in 30 seconds.

Secondly, for accessing better understanding about the data transfer on one switch topology, there was another step undertaken, in the form of average data transfer in 30 seconds, one switch topology. The rate was noted for one client contacted to one server, two clients contacted to two servers and three clients contacted to three servers. Similarly, the four clients contacted to four servers and five clients contacted to five servers for both traditional and SDN. This processes remained

The results obtained from this provided again the similar idea that the time taken by SDN to transfer the data from the client to the server, for all the considered clients and servers is less than the traditional networking. It should be noticed that this is just a single switch working in this experiment for connecting the client and the server and the fluctuating for the all the five clients and servers connected to it.

Hereby, the idea of increasing the switches to see the efficacy of these networking’s was utilized. Hence, in the next part of the experiment, two switches were introduced and then average data transfer in 30 seconds was tested. The results of this test should that the two switches would not be favorable with the SDN working this means that the data transferring was slowed down by SDN when two switches were introduced rather than using one switch. Hereby, it showed the results that the traditional networking had a higher speed of working as compared to the SDN and hence while using two switches use of traditional networking would be better.

If we look for reasons behind this failure of SDN in proving itself to be better than traditional networking, then it comes to the view that there is a high rate of wrong use of SDN and OpenFlow, which causes this slow down when there are a greater number of switches introduced (Taha, 2014 p1). Moreover, it is also provided the literature review that the SDN has the best speed in data transferring in any case which has been disapproved through these results (Gelberger, 2013).

While there could be chances of other errors in this case as well. Since it has usually been expected by utilization of effective methods to reduce such issues, the speed of the SDN could be increased. Since the difference between the speeds of SDN and traditional networking was very little hereby, this point approves that maybe there was some error in the data or results. Moreover, the decrease in speed was seen in just 1 client to 1 server and 2 clients to 2 servers, while for the rest of the clients to servers speed the results are in favor of SDN again.

This took up the expectations of the analysis and hence, it is not expected that after having such effective attributes driven, SDN would be acquiring effective protocol that is OpenFlow, which would be further increasing the efficacy of the SDN. However, the literature also adds to the idea that the traditional protocols are usually not as effective as even found the cloud computing and this makes them stand far away from achieving the speed and efficacy of software defined networking.

In the later part of the experiment, the latency or Round-trip time for the current study was measured, which means the quality of the signals and time it took to reach the server and going back by making use of through the ping command. The process for this part of the result was same as before. Which means that the rate was noted for one client contacted to one server, two clients contacted to two servers and three clients contacted to three servers? Similarly, the four clients contacted to four servers and five clients contacted to five servers for both traditional and SDN. This processes remained same for all the remaining and previous parts as well.

Monitoring Round Trip Time enabled the researcher to observe important data required for the network troubleshooting and traffic engineering. The frequent monitoring method used for monitoring is through sending probe packets from the chosen vantage points. The control over the network protocols is limited in the traditional networks, which is likely to make it difficult to monitor the selected path. The rising idea of SDN makes the network control easy. On the other hand, OpenFlow or the common SDN protocol requires no support Round Trip Time monitoring. The results for this part also asserted the point that the SDN provided the quality rates for RTT and hence it is a better alternative as compared to traditional networking (Romero, 2012).

These results approved the idea, which was provided in the literature review in regard to the (RTT) and hence, makes the SDN an approved networking device to control and produce effective RTTs. This includes the idea that the OpenFlow also stands to be an effective protocol and should be followed in accordance with its limitations because there is always some limitation in the protocols.

The literature reveals that Transmission Control Protocol/Internet Protocol (TCP) bandwidths speed are most important protocols that could be when it comes to connecting the computers. Their effectiveness helps in linking the computers in the most effective manner and here again the rate in which these protocols are distributed are the most important aspect to be considered. The TCP are hard to access effectively by making use of traditional networking and so the effectiveness of SDN also include perfect delivery of TCP through its well-managed protocols.

Considering the literature review, in this analysis, it is expected that the results for the SDN should also be underlying the same idea that the traditional networking could not catch up with the high technological speed of SDN and hence, the results and interpretation would be showing a failure of the traditional networking to hold back the SDN performance. Furthermore, this consideration should not be limited to this point but, there are again chances of using a single switch topology or two switch topologies, which were both considered during the experiment(Romero, 2012)..

The results in this regard provided with the idea that the rate of the SDN was greater than that with which the traditional networking was working for all the clients to servers, at the time only one switch topology was being used. For the two switch topology, in this case, the results did not show that the SDN would be effective for this TCP bandwidth speed, but this result was limited to only 1 client to 1 server and 2 clients to 2 servers, however, as soon as it moved to the rest of the clients to servers. This means that the efficacy is retained in this case as well.

Considering the Jitter or the latency variation, which are usually being measured through an Iperf UDP test, it remains absolute that any networking, which provides high speed of Jitters, remains the best of all. Moreover, no networking could be completed without these jitters. Hereby analyzing the quality of jitters by the SDN was also included in the experiments and the results were obtained for not just one switch topology but also for the two switch topology.

It was expected that the results obtained for the rates of SDN would surpass those which were provided by traditional networking, which is likewise all the above result and literature review considerations. However, at the very initial stage that is 1 client to 1 server, the traditional networking showed better performance as compared to SDN, while the rest of the clients to servers remained constant with favorable results for SDN to be higher in its performance.

The time in milliseconds for the UDP Jitter, Two Switch Topology, showed up to be providing the picture of the promising nature of SDN when its performance is being concerned. Here it was proven that the SDN performance was high as acquires a high-quality rate for USD jitter and traditional networking failed to perform well in this case as well. This refers to the idea that all the attributes and features of SDN are considerably better and more efficient than traditional networking. Moreover, the use of the SDN should be promoted since it would be providing better results through its shows potential performance.

Moreover, there is also a foundation of the trouble created that when making use of two switch topology there are greater chances of slowing down of the SDN. Hereby some solutions should be researched in this regard, while through this analysis, now it is quite clear that usage of the SDN, as an alternative of traditional networking would be fruitful and more productive. Moreover, the SDN remains the best opportunity for the networking purpose, for another reason that its high quality generation would add to the server’s quality provision and services.

Article Categories:

Leave a Comment

Your email address will not be published.