AN APPROACH TO IMPROVE THE EFICIENCY OF PRIORITY QUEUE IMPLEMENTATIONS

This paper is an extension to our previous work [1]. Based on their running time efficiencies, earlier we gave a comparative study of various available ways to implement priority queues. In this paper we demonstrate a way to further improve their running time efficiency. We bring an introduction to memory pooling and later describe how this concept may be used to improve the running time efficiency of priority queues. Priority queues involve a large number of system calls during memory allocation. This leads to extra time being required for node generation. However, this problem can be solved by using a already available memory pool. For the purpose of memory allocation, a pre assigned pool of memory could be used instead of making system calls on every node generation. This approach significantly reduces the time frame of program execution. Although no changes in algorithm for priority queues have been made, this approach works on the system level and is very beneficial when a very large number of data nodes have to be created. Likely so, in our previous work [1] we had shown that we already have the constant time executable algorithms available. So with this approach we go out of the box to improve the efficiency of priority queues.