Embedded systems are getting more complex and require more intensive processing capabilities. They must be able to adapt to the rapid evolution of the high-end embedded applications that are characterized by their high computation-intensive workloads (order of TOPS: Tera Operations Per Second), and their high level of parallelism. Moreover, since the dynamism of the applications is becoming more significant, powerful computing solutions should be designed accordingly. By exploiting efficiently the dynamism, the load will be balanced between the computing resources, which will improve greatly the overall performance. To tackle the challenges of these future high-end massively-parallel dynamic embedded applications, we have designed the AHDAM architecture, which stands for "Asymmetric Homogeneous with Dynamic Allocator Manycore architecture". Its architecture permits to process applications with large data sets by efficiently hiding the processors' stall time using multithreaded processors. Besides, it exploits the parallelism of the applications at multiple levels so that they would be accelerated efficiently on dedicated resources, hence improving efficiently the overall performance. AHDAM architecture tackles the dynamism of these applications by dynamically balancing the load between its computing resources using a central controller to increase their utilization rate.The AHDAM architecture has been evaluated using a relevant embedded application from the telecommunication domain called "spectrum radio-sensing". With 136 cores running at 500 MHz, AHDAM architecture reaches a peak performance of 196 GOPS and meets the computation requirements of the application.

Les systèmes embarqués sont de plus en plus complexes et requièrent des besoins en puissance de calcul toujours plus importants. Ils doivent être capables de s'adapter à l'évolution rapide de leurs applications qui requièrent un haut niveau de performance (ordre du TOPS: Téra-opérations par seconde) et de parallélisme. Par ailleurs, la complexité des parties irrégulières étant de plus en plus importantes, des solutions de calcul performantes et adaptées doivent être mises en œuvre afin de prendre en compte leur dynamisme. Une prise en compte efficace du dynamisme réduit le déséquilibre de charge entre les ressources de calcul et améliore grandement les performances globales.Pour répondre aux besoins de ces applications de calcul intensif massivement parallèle et dynamique, nous proposons dans cette thèse l’architecture AHDAM qui signifie « Asymmetric Homogeneous with Dynamic Allocator Manycore architecture ». Cette architecture a été conçue afin de masquer efficacement la latence d’accès à la mémoire extérieure dont de nombreux accès sont nécessaires lors de la manipulation de grands volumes de données. Pour cela, des processeurs multitâches ont été utilisés. Par ailleurs, l’architecture AHDAM imbrique plusieurs niveaux de parallélisme afin de tirer partie efficacement des différentes formes de parallélisme des applications, et ainsi atteindre un haut niveau de performance. Enfin, cette architecture utilise un contrôleur centralisé pour équilibrer la charge de calcul entre ses ressources de calcul afin d’augmenter leur taux d’utilisation et supporter les applications fortement dynamiques.L’architecture AHDAM a été évaluée en portant une application de radio logicielle appelée «spectrum radio-sensing ». Avec 136 cœurs cadencés à 500 MHz, l'architecture AHDAM atteint une performance crête de 196 GOPS et répond aux exigences de l'application. / Embedded systems are getting more complex and require more intensive processing capabilities. They must be able to adapt to the rapid evolution of the high-end embedded applications that are characterized by their high computation-intensive workloads (order of TOPS: Tera Operations Per Second), and their high level of parallelism. Moreover, since the dynamism of the applications is becoming more significant, powerful computing solutions should be designed accordingly. By exploiting efficiently the dynamism, the load will be balanced between the computing resources, which will improve greatly the overall performance. To tackle the challenges of these future high-end massively-parallel dynamic embedded applications, we have designed the AHDAM architecture, which stands for “Asymmetric Homogeneous with Dynamic Allocator Manycore architecture". Its architecture permits to process applications with large data sets by efficiently hiding the processors' stall time using multithreaded processors. Besides, it exploits the parallelism of the applications at multiple levels so that they would be accelerated efficiently on dedicated resources, hence improving efficiently the overall performance. AHDAM architecture tackles the dynamism of these applications by dynamically balancing the load between its computing resources using a central controller to increase their utilization rate.The AHDAM architecture has been evaluated using a relevant embedded application from the telecommunication domain called “spectrum radio-sensing”. With 136 cores running at 500 MHz, AHDAM architecture reaches a peak performance of 196 GOPS and meets the computation requirements of the application.