Author

Ramya Danappa

Graduation Semester and Year

2020

Language

English

Document Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science and Engineering

First Advisor

Song Jiang

Abstract

There are different page replacement algorithms and yet there seems to be some drawbacks in them, making no page replacement algorithm ideal. To be one step closer to achieving the ideal algorithm it is vital to have a maximum cache hit ratio and strong consistent across different workload. In this paper we shall explore a new cache management policy called “Dynamic and Stable page replacement algorithm” uses frequency and recency importance dynamically”. The proposed page replacement algorithm has overcome the drawbacks of the LRU (Least Recently Used) algorithm in many scenarios and also overcome the drawbacks of LFU (Least Recently Used).Like LRU can be easily polluted by a scan, that is by a sequence of one-time use only page requests leading to decrease in performance and also LFU does not pay attention to recent history. The proposed algorithm has achieved consistent hit ratio by using the fundamentals of page replacement algorithm which are: low recency and high frequency of the blocks. To achieve this stability we have to integrate the design principles of LRU and LFU in the cache and the blocks have been placed blocks according to the hit on the block, the above mentioned is done using simple computations.

Keywords

LRU, Least recently used, LFU, Least frequently used

Disciplines

Computer Sciences | Physical Sciences and Mathematics

Comments

Degree granted by The University of Texas at Arlington

29096-2.zip (648 kB)

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.