Schedulability-driven partitioning and mapping for multi-cluster real-time systems

Abstract

We present an approach to partitioning and mapping for multicluster embedded systems consisting of time-triggered and eventtriggered clusters, interconnected via gateways. We have proposed a schedulability analysis for such systems, including a worst-case queuing delay analysis for the gateways responsible for routing inter-cluster traffic. Based on this analysis, we address design problems characteristic to multi-clusters: partitioning of the system functionality into time-triggered and event-triggered domains, and mapping of processes onto architecture nodes. We present a branch-and-bound algorithm for solving these problems. Our algorithm is able to find schedulable implementations under limited resources, achieving an efficient utilization of the system. The developed algorithms are evaluated using extensive experiments and a real-life example.

Una revisión de técnicas para la optimización del despliegue y planificación de sistemas de tiempo real distribuidos

“…Los trabajos (Pop et al, 2003b) y (Pop et al, 2004c) proponen heurísticos para desplegar tareas en computadores y planificar tareas y mensajes en sistemas de tiempo real distribuidos que combinan los paradigmas TT y ET. El modelo se basa en un grupo de computadores de clase TT conectados a través de una red de comunicaciones TTP, otro grupo de computadores de clase ET unidos mediante una red CAN y un gateway que enlaza ambas redes y por el que discurren los mensajes intercambiados entre ambos grupos.…”

Real-time scheduling of divisible loads in cluster computing environments

“…The scheduling models investigated for distributed or multiprocessor systems most often (e.g., [32,31,21,1,29,19,5,22]) assume periodic or aperiodic sequential jobs that must be allocated to a single resource and executed by their deadlines. In recent years, researchers have begun to investigate real-time scheduling of parallel applications on a cluster [41,30,18,2,4].…”

An Efficient Algorithm for Real-Time Divisible Load Scheduling

“…The scheduling models investigated for distributed or multiprocessor systems most often (e.g., [1,5,15,16,23,25,26]) assume periodic or aperiodic sequential jobs that must be allocated to a single resource and executed by their deadlines. With the evolution of cluster computing, researchers have begun to investigate real-time scheduling of parallel applications on a cluster [3,4,13,24,30].…”