Tasks Processing Allocation Based on Memory Usage Performance (TPA-MUP)
Journal Title: International Journal of the Computer, the Internet and Management - Year 2019, Vol 27, Issue 2
Abstract
The objective of this paper is to design a scheduler of task allocation time of simultaneous running two independence majority tasks. Objective of task allocation scheduler is to minimize two tasks memory usage while they are both processing. Task’s data that was kept in memory must be transfer or move out if this task was terminated out of CPU. New allocated task’s data must be moved to be kept in memory. Data transfer in / out between memory and disk drive consume too much time. These problems should be solved by arrange both tasks to start in difference time. Each task has its own required physical memory usage. This paper present task allocation for processing in CPU based on their physical memory usage. Prior percentage of physical memory usage equation of each task was constructed from historical memory usage data. Constraint of starting time and finish time of both tasks were assigned. Lagrange function was used to find out the minimum value of memory usage. Feasible result variable, starting time of each task, were used to be a starting time of each task. This scheduler was trained on 5 working day 12 hours working. Percentage physical memory usage was captured every twenty minutes thus there were 36 captured samples per day. TPA-MUP was test in time performance by thirty users, six users per day. The result of average time performance of two major tasks allocation under designed scheduler are decrease about 15% from ordinary FIFO scheduler. Users are appreciated in time performance of two major tasks running under TPA-MUP more than ordinary FIFO at statistical significance α 0.05.
Authors and Affiliations
Amnat Sawatnatee, Somchai Prakancharoen
Keywords
Related Articles
- EP ID EP604257
- DOI -
- Views 161
- Downloads 0
Harnessing the Power of Information Technology for Efficiency in E-Court and E-Trial
Searching Google for “E-court”, over 51 million entries were found. Similarly, searching Google for “E-Trial”, over 26 million entries were found. Therefore, it can be said that the terms “E-court” and “E-trial” are v...
Improving Institutional Performance of Study Program Based on Knowledge Management Roles and Academic Information System Mediated by Strategic Planning
The purpose of this study is to figure out the influences of knowledge management roles and academic information system on institutional performance of Study Program mediated by strategic planning. The population of this...
An Automatic Ontology Matching Using Similarity of Components and Synonym
An automatic ontology matching is a task to align identical classes from different ontologies. This work presents a new method to detect same ontological class using components within ontology including label and propert...
The Effect of Blended Learning Approach Using Facebook and Learning Experiences Outside the Classroom on an Inquiring Mind of Freshmen in Teacher Education Program at Institute of Physical Education, Chumphon Campus
The purposes of this research were: 1) to develop a blended learning model using Facebook and learning experiences outside the classroom; 2) to study the effect of the developed blended learning approach on students’ inq...
The Relationships between the Corporate Vision and Good Corporate Governance Efficiency of Listed Companies in Stock Exchange of Thailand
This research aims to study the relationships between the corporate vision and Good corporate governance efficiency of listed companies in stock exchange of Thailand. This research used a questionnaire as an instrument f...