Networking and Application Latency: When the total number of user request is equivalent to the total of traffic volume (number of threads) supported by a server, any subsequent requests will be placed in the request queue. The requests have to wait to be processed [12], [13]. This process will increase the waiting time to go. The waiting time is divided into two parts: network latency and application latency. The former refers to the waiting time for data to flow from user to server and from server to server, whereas the latter refers to the waiting time for data processing in a server