As mentioned in the abstract, the first experiment was conducted with the aim of understanding the network performance parameter characteristics by measuring the extent and load to which various network resources available to the user were being consumed when most the non-essential network applications had been turned off.
This is essential for the purpose of this experiment since the network performance parameters to be analyzed have to be measured against a certain benchmark after they have been recorded, documented and quantified.
The second experiment was performed to understand the impact of the load on the central processor of the end-user after the connection has been made, including the time elapsed in establishing a secure connection (though the application in question does not require a secure host for the establishment of a connection, due to which the characteristics differ when hosted over a mobile phone) to determine the extent of network resources being consumed when one is watching a YouTube video.
The major components of the scope of this experiment include the number of bytes sent per packet (on an average) from the client to the server or in the opposite direction, whereas trace studies are included for peer to peer conversations. Since the node congestion is also significant as mentioned in the literature review of this paper, suitable congestion measures will be parameterized based on the resources available within the network characteristics of the YouTube server.
From the trace study, the following inferences can be made of the utilization of network resources, by discussing the following processes and breaking down their salient features to be reflected by suitable network performance parameters. The most salient of these network performance parameters are listed as follows:
2.Bulk data transfer applications
3.Interactive web applications
5.Wide-area network locality