The static thread pool scheme has a problem of occupying a static amount of system resources. To solve it, a dynamic thread pool scheme based on the learning for a web server is suggested, which is running on a multi-thread environment. The suggested ...
The static thread pool scheme has a problem of occupying a static amount of system resources. To solve it, a dynamic thread pool scheme based on the learning for a web server is suggested, which is running on a multi-thread environment. The suggested scheme creates the threads through the prediction of the next number of periodic requests using Auto Regressive model with the worker multi-processing module of the web serve. K-Nearest Neighbor algorithm is used to learn and set the exact number of threads in advance according to the previous requests. In this paper, the response time is decreased by modifying the number of threads dynamically, and the system resources can be used more efficiently by managing the number of threads according to the requests.