Hybrid Metaheuristic Approach for Scheduling of Aperiodic OS Tasks - Complex System Modelling and Control Through Intelligent Soft Computations - page 840

Algorithm and Complexity

In the following paragraph we will describe our proposed algorithm for this con-
tribution for the aperiodic tasks recon guration.

Algorithm begin
t = get current time();
U =0;
For each partition
Compute( a );

Compute( b );

Compute( );

Display_parameters(

c

a; b; c );

save ( );

if (feasible) then
{
return (

a; b; c

“ Guaranteed ” );

}
else
return
( You can try by using solution 1, or,

You can try by using solution 2,
Compute(Resp k ; 1 );
Compute(Resp k ; 2 );
Generate(Resp opt
k

“

);

end

Complexity

The recon gurable schedulability in the case of aperiodic tasks, i.e., each task has

an offset Si, i , such that is strongly coNP-hard. This complexity was decreased in our
approach from Np-hard to ef

cient O(n + m) 2 guarantee algorithm when we have M

recon cient algorithm results in the dynamic

scheduling solutions. These solutions are presented by a proposed intelligent agent-
based architecture where a software agent is used to evaluate the response time, to
calculate the processor utilization factor and also to verify the satisfaction of real-
time deadlines.
On the other hand, the busy period, which is computed for every analyzed task
set and has a pseudo-polynomial complexity for U

gurations scenarios w h . This ef

≤ 1, is decreased also by the

optimization of the response times in our work.