Hybrid Metaheuristic Approach for Scheduling of Aperiodic OS Tasks - Complex System Modelling and Control Through Intelligent Soft Computations - page 823

enables a designer to grant that timeliness constraints associated to an application
run by the system are always met for all possible con gurations. The goal of the FC

is to ensure a deterministic respect of the timeliness constraints Gharsellaoui et al.
( 2012 ).

4.1.1 Recon guration of Sporadic Tasks

A recon guration can be decided either off-line or on-line. In the first case, the goal

is to check if several hardware platforms or several hardware con gurations can be

used to run a speci c application while preserving the timeliness constraints of

the tasks. In the second case, a recon guration might result from a system mode

change to adapt the system to the context of its execution or to handle hardware or
software faults. Sensitivity analysis aims at studying the ability to introduce more
fl

cations. In this work, we study many sensitivity analysis
(more than one task parameter can evolve).
This topic chapter focuses on the dynamic recon

flexibility in the speci

gurations of assumed mixture
of off-line and on-line workloads that should meet deadlines de

ned according to
user requirements. We propose an intelligent agent-based architecture in which a
software agent is deployed to dynamically adapt the system to its environment by
applying reconfiguration scenarios. The agent dynamically provides technical
solutions for users when the system becomes unfeasible (e.g. deadlines are vio-
lated), by sending sporadic tasks to idle times, by modifying the deadlines of tasks,
the worst case execution times (WCETs), the activation time, by tolerating some
non critical tasks m among n according to the (m, n) firm model (Hamdaoui and

Ramanathan 1995 ) and a reasonable cost, or in the worst case by removing some
non hard (soft) tasks according to prede

ned heuristic. We implement the agent to
support these services which are applied to a running example with real-life design
examples in order to demonstrate the effectiveness and the excellent performance of
the new proposed algorithm.
On the other hand, the scheduling of tasks is an essential requirement in most
real-time embedded systems, but invariably leads to unwanted CPU overheads.
This topic chapter work presents also real-time scheduling techniques for reducing
the response times of uniprocessor aperiodic tasks to be scheduled with real-time
periodic tasks. Two problems are addressed in this part: (i) the scheduling of
aperiodic tasks when they arrive in order to obtain a feasible system, and (ii) the
scheduling of periodic and aperiodic tasks to minimize their response times. In
order to improve the responsiveness to both types of problems, an ef cient

approach is proposed by using the Poisson distribution which is a discrete distri-
bution. It is often used as a model for the number of events in a speci c time period.

Instead, it uses the fixed interval of time or space in which the number of successes

is recorded. The space of feasible deadlines (D-space) is then assumed to be equal to
one time unit in our proposed approach. The effectiveness and the performance of
the designed approach is evaluated through simulation studies.