Josep M. Banús

• Phd
• Universitat Rovira i Virgili

Mobile Ad Hoc Network (MANET) middleware must be aware of the underlying multi-hop topology to self-adapt and to improve its communication efficiency. For this reason, many approaches rely on specific cross-layer communications to interact with the network protocols in the kernel space. But these solutions break the strict layering of the network s...

In this paper we present a global scheduling method for shared memory multiprocessor systems that provides a fixed-priority preemptive scheduling of periodic tasks, hard aperiodic tasks and soft aperiodic tasks on a set of identical processors. The method is based on the functioning of the Dual Priority Scheduling Algorithm extended to work in a mu...

In this paper we present a modification of the Dual Priority Scheduling Algorithm to work on shared memory multiprocessor systems improving the average-case schedulability. The proposal deals with global fixedpriority preemptive scheduling of periodic tasks on identical processors. The algorithm allows to schedule hard real-time periodic tasks usin...

We present an adaptation of the dual priority scheduling algorithm to schedule both hard real-time periodic tasks and soft-aperiodic tasks in shared memory multiprocessor systems. The goal is to achieve low mean aperiodic response times while meeting all periodic task deadlines. Our proposal allows periodic and aperiodic tasks to migrate to other p...

We propose a scheme to allocate individual softaperiodic tasks in a shared memory symmetrical multiprocessor for hard-real time systems applications. The aim of this scheme is to minimize the aperiodic mean response time by running a local uniprocessor scheduler to serve periodic tasks guaranteeing their deadlines, and serving aperiodic tasks with...

We propose a scheme to allocate individual soft- aperiodic tasks in a shared memory symmetrical multiprocessor for hard-real time systems applications. The aim of this scheme is to minimize the aperiodic mean response time by running a local uniprocessor scheduler to serve periodic tasks guaranteeing their deadlines, and serving aperiodic tasks wit...

An algorithm is proposed for systems with periodic and aperiodic real-time tasks, so that the periodic ones meet their deadlines and the aperiodic ones minimise their response times. The algorithm is comparable in results with the static slack stealing algorithm but its implementation is much simpler in both memory space and computation time.

In this paper we present a modification of the Dual Priority Scheduling Algorithm to work on shared memory multiprocessor systems improving the average-case schedulability. The proposal deals with global fixed-priority preemptive scheduling of periodic tasks on identical processors. The algorithm allows to schedule hard real-time periodic tasks usi...

We propose a new off-line heuristic algorithm to allocate and assign priorities in distributed hard real-time systems when tasks have end-to-end deadlines. We have found solutions two order of magnitude faster than the simulated annealing technique applied to scheduling the same task sets, with maximal average loads around 90%.