Hybrid Metaheuristic Approach for Scheduling of Aperiodic OS Tasks - Complex System Modelling and Control Through Intelligent Soft Computations - page 822

4 Aperiodic Task Scheduling Contribution

This topic chapter deals with the problem of uniprocessor scheduling of both
periodic and sporadic/aperiodic tasks on uniprocessor. We introduce in this work an
EDF-based scheduling algorithm to optimize the response times of recon

gurable

tasks while ensuring first that all the periodic tasks meet their deadlines and second

that all the sporadic and aperiodic tasks can respect their constraints. A necessary
and sufficient schedulability test is presented, and an efficient O(n + m) guarantee
algorithm is proposed. Indeed, to obtain this goal, this system should be changed
and automatically adapted to its environment on the occurrence of random dis-
turbances such as hardware-software faults. A random disturbance is de ned in this

work as any random internal or external event allowing additions, removals or
updates of tasks at run time to adapt the system

s
implementation is dynamically changed and should meet all considered deadlines
of the current combination of tasks. Nevertheless, when an automatic recon

' s behavior. Therefore, the system '

gu-
ration scenario is applied, the deadlines of new and old tasks can be violated. We
mean by recon guration scenario in our work, the removal, update or addition of

new aperiodic tasks when they arrive at run-time without prior knowledge, in order
to save the whole system on the occurrence of hardware-software faults in a safe
state (feasible system), or also to improve its performance when random distur-
bances happen at run-time. This scheduling algorithm is used at run-time to provide
dynamic solutions when deadlines are violated after a particular recon guration

scenario. We propose an agent-based architecture where an intelligent software
agent is used to evaluate the response times, to calculate the processor utilization
factor and also to verify the satisfaction of real-time deadlines.

4.1 First Method: An EDF Based Scheduling Approach

Nowadays, due to the growing class of portable systems, such as personal com-
puting and communication devices, embedded and real-time systems contain
complex software which is increasing by the time. This complexity is growing
because many available software development models don ' t take into account the

speci c needs of embedded and systems development. The software engineering

principles for embedded system should address speci

c constraints such as hard
timing constraints, limited memory and power use, prede ned hardware platform

technology, and hardware costs. The new generations of embedded control systems
are addressing new criteria such as flexibility and agility Gharsellaoui et al. ( 2012 ).

For these reasons, there is a need to develop tools, methodologies in embedded
software engineering and recon

fl

gurable embedded control systems as an inde-
pendent discipline. By response for this requirement of developing a recon gurable

systems, many interesting academic and industrial studies have been made in recent
years. Feasibility Conditions (FC) for the dimensioning of a real-time system