Adaptive co-scheduling for periodic application and update transactions in real-time database systems

  • Authors:
  • Song Han;Kam-yiu Lam;Jiantao Wang;Sang H. Son;Aloysius K. Mok

  • Affiliations:
  • Department of Computer Science, University of Texas at Austin, United States;Department of Computer Science, CityU of Hong Kong, Hong Kong;Department of Computer Science, CityU of Hong Kong, Hong Kong;Department of Computer Science, University of Virginia, United States;Department of Computer Science, University of Texas at Austin, United States

  • Venue:
  • Journal of Systems and Software
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we study the co-scheduling problem of periodic application transactions and update transactions in real-time database systems for surveillance of critical events. To perform the surveillance functions effectively, it is important to meet the deadlines of the application transactions and maintain the quality of the real-time data objects for their executions. Unfortunately, these two goals are conflicting and difficult to be achieved at the same time. To address the co-scheduling problem, we propose a real-time co-scheduling algorithm, called Adaptive Earliest Deadline First Co-Scheduling (AEDF-Co). In AEDF-Co, a dynamic scheduling approach is adopted to adaptively schedule the update and application jobs based on their deadlines. The performance goal of AEDF-Co is to determine a schedule for given sets of periodic application and update transactions such that the deadline constraints of all the application transactions are satisfied and at the same time the quality of data (QoD) of the real-time data objects is maximized. Extensive simulation experiments have been performed to evaluate the performance of AEDF-Co. The results show that by adaptively adjusting the release times of update jobs and scheduling the update and application jobs dynamically based on their urgencies, AEDF-Co is effective in achieving the performance goals and maximizing the overall system performance.