Performance Measurement and Management Model of Data Generation and Writing Time in Personal Computer

Abstract

This paper is aimed at doing a performance measurement of data generation and write time in personal computer by generating relatively large quantity of numeric data and writing them in the text file and along with that developing performance models for them. These models may be used for performance management and at the same time they also provide an insight into the important aspect of relationship between ‘time’ and ‘data size’ of a particular platform. In this paper, we have used the curve estimation technique for analyzing the dataset and proposed two different types of generic models which can be viewed as: (i) 'Time' ~ f('Number of Data Elements') and (ii) 'Number of Data Elements' ~ f('Time'). In total six (6) models has been proposed, three from each type. In addition to providing performance measurement models of a particular platform (hardware and software), the proposed generic models also gives us the following information: (i) how much time is required to generate a given amount of numeric data, (ii) how much time is required to generate and write a given amount of numeric data in a text file, (iii) how much time is required to write a given amount of numeric data in a text file, (iv) how much amount of numeric data can be generated in a given time, (v) how much amount of numeric data can be generated and written in a given time and (vi) how much amount of numeric data can be written in a given time.