High deadline meeting rate of non-preemptive dynamic soft real time scheduling algorithm

Hard real time systems were often implemented with preemptive scheduling, which gives priority to the highest priority task. In this paper we present a new non-preemptive scheduling of jobs meant for soft real time application. Our ultimate aim is to increase the deadline meeting rate of the Earliest Deadline First (EDF) algorithm during overload condition while maintaining the optimum performance it poses during normal load. Our approach, grouped jobs with near deadlines together using our novel algorithm and schedule the jobs within a group using another algorithm. We named the approach Group, Utilization and Deadline Tolerance EDF (gutEDF). We will present result comparing the deadline meeting rate and average response time of gutEDF and EDF under different deadline tolerance values and compare the deadline meeting ratio improvement of gutEDF and gEDF.

I. Introduction

In the real life, a real time system (RTS) is expected to act upon the environment it controls within a specified time limit. Typically, an RTS compose of a computing system which controls an environment based on data available about the controlled system. A real time computing system may control a device or process. Its input data is usually acquired by sensors within certain intervals of time which may be periodic or non-periodic. Under any circumstances, an RTS must response by triggering the respective actuators within a specified time limit. The capability of the computing system to meet these timing demands depends on various parameters which constitutes to its capacity to perform the required computation within the time limit.