A Novel Approach to Improve the System Performance by Proper Scheduling in Memory Management

Abstract

Virtual memory techniques are available in modern operating system will permits the execution of a program even if it is partially available in memory thus providing an illusion of very large memory to the user and freeing the user from the concern of large program size. The primary storage size has increased by multiple order of magnitude. With several gigabytes of primary memory, algorithms that require a periodic check of each and every memory frame are becoming less and least practical. Some basic assumptions used by the traditional page replacement algorithms were invalidated, resulting in a revival of research area. Currently the usage of the object oriented programming is much higher than the structured programming. Hence the locality of reference of user software has weakened. Due to the above problem for the page replacement methods also we have to think about new methods. In this paper mainly focused on the improvement of the processor performance. In order to achieve this, the number of page faults are tried to reduce, by using the page replacement method in the correct way. In modern operating systems mainly make use of the Least Recently Used (LRU) page replacement method, in which only the arrival time of the pages to the page frame are make used. Here we made an attempt to consider both the arrival time and the number of reference done on the page is considered. By consider these two parameter we developed a new algorithm Least Recently Used and Least Frequently Used (LRU-LFU) which will give a better performance than LRU page replacement method.