Exploiting laxity for heterogeneous multiprocessor real-time scheduling

In this paper we study the effect of processor affinity on real-time scheduling algorithms in uniform multiprocessor environments. For this reason we compare our two previously proposed algorithms called EFDF and FLLF, with two well know algorithms, namely global EDF and global LLF. Despite its non-optimality, EDF is an appropriate algorithm to use on multiprocessors. However, the performance of the traditional algorithms such as EDF and LLF significantly degrades in overloaded situations. Moreover, they produce a relatively large number of migrations which may prove unacceptable for use on some parallel machines. It is shown that our proposed algorithms not only demonstrate a performance as good as that of EDF in non-overloaded conditions but also overcome its deficiencies in overloaded situations in many aspects. Furthermore, they impose much less overhead on the system.

show abstract

“…Additionally, FLLF employs the processor affinity concept as well. For more details on the FLLF algorithm please refer to [15].…”

Section: Approachmentioning

confidence: 99%

“…The Feasible Least Laxity First (FLLF) algorithm [15] is based on GLLF, in which we have made use of laxity in two ways: calculating the dynamic priority of tasks and performing feasibility check. Additionally, FLLF employs the processor affinity concept as well.…”

Section: Approachmentioning

confidence: 99%

Investigating the effect of processor affinity on uniform parallel machine scheduling

Khatib-Astaneh

Salmani

Yazdi

et al. 2009

2009 International Conference on Application of Information and Communication Technologies

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, most algorithms are aimed at multiprocessor systems with soft real-time constraints (Yazdi et al, 2008;Emberson and Bate, 2007;Anderson et al, 2005). Some papers have addressed both task migration and energy consumption minimization (Briao et al, 2007;Chen, 2005).…”

Section: Related Workmentioning

confidence: 99%

High performance dynamic voltage/frequency scaling algorithm for real-time dynamic load management

Coronel

Simó

2012

Journal of Systems and Software

View full text Add to dashboard Cite

Modern cyber-physical systems assume a complex and dynamic interaction between the real world and the computing system in real-time. In this context, changes in the physical environment trigger changes in the computational load to execute. On the other hand, task migration services offered by networked control systems require also management of dynamic real-time computing load in nodes. In such systems it would be difficult, if not impossible, to analyse off-line all the possible combinations of processor loads. For this reason, it is worthwhile attempting to define new flexible architectures that enable computing systems to adapt to potential changes in the environment.We assume a system composed by three main components: the first one is responsible of the manage-ment of the requests arisen when new tasks require to be executed. This management component asks to the second component about the resources available to accept the new tasks. The second component performs a feasibility analysis to determine if the new tasks can be accepted coping with its real-time constraints. A new processor speed is also computed. A third component monitors the execution of tasks applying a fixed priority scheduling policy and additionally controlling the frequency of the processor. This paper focus on the second component providing a "correct" (a task never is accepted if it is not schedulable) and "near-exact" (a task is rarely rejected if it is schedulable) algorithm that can be appli-cable in practice because its low/medium and predictable computational cost. The algorithm analyses task admission in terms of processor frequency scaling. The paper presents the details of a novel algo-rithm to analyse tasks admission and processor frequency assignment. Additionally, we perform several simulations to evaluate the comparative performance of the proposed approach. This evaluation is made in terms of energy consumption, task rejection ratios, and real computing costs. The results of simula-tions show that from the cost, execution predictability, and task acceptance points of view, the proposed algorithm mostly outperforms other constant voltage scaling algorithms.

show abstract

“…Pocos trabajos cubren la migración de tareas en el contexto de los sistemas empotrados e interconectados por red. Además muchos algoritmos están orientadosúnicamente a sistemas multiprocesador y con restricciones no estrictas de tiempo real (Yazdi et al, 2008;Emberson & Bate, 2007;Anderson et al, 2005). Algunas de estas publicaciones están relacionadas con la migración de tareas y con la minimización del consumo energético (Briao et al, 2007;Chen, 2005).…”

Section: Trabajos Relacionadosunclassified

Gestión de la carga dinámica de tareas de tiempo real con criterios de ahorro energético y su aplicación en el desarrollo de un middleware de control.

Parada¹,

Osvaldo²

View full text Add to dashboard Cite

AgradecimientosAgradezco a todos los profesores y compañeros que de una u otra forma han colaborado en el desarrollo de esta tesis. Y en especial, a mi director José Simó cuya orientación, apoyo y susánimos han sido fundamentales para la realización de este trabajo.Igualmente agradezco a los miembros del Grupo de Informática Industrial del DISCA, con especial cariño a Alfons, Paco y Pascual, por su colaboración y acogimiento durante mi formación y desarrollo de esta investigación.A mis padres, Osvaldo e Idda, a Laura, Lucas y Noa por toda la paciencia, comprensión y motivación inyectada para la culminación de este proyecto. III AbstractThe development of embedded systems in industrial sectors such as railway, aerospace and automotive are based on Critical Real-Time Embedded Systems (CRTES). These systems face new challenges and demand related to increase of dependability, intelligence, connectivity, cost-size-volume reduction and energy efficiency. In this last topic is where this thesis expects to have a higher contribution.The global energy consumption can be combined with others criteria such as schedulability, communication delays and control application correctness, which contribute to determine the dynamic code movement and on-line load balancing in a system. The main goal of this thesis is the development of mechanisms for the management and optimization of energy consumption. These mechanisms are presented in the context of a distributed real-time control system and from the perspective of control kernel middleware. Let's consider a dynamic environment where an embedded and networked system operates with the support of task migration and processor frequency scaling. Assuming that the system knows where and when it must allocate tasks, we must perform feasibility analyses when each task arrives and departs on the affected embedded units. This guarantees that the temporal requirements of the system will be accomplished during the task allocation phase or delegation of tasks. Additionally, a new processor speed (frequency scaling) should be also computed to enable the system to adapt itself to the new computational workload and reduce energy consumption. And in this last is where the proposed algorithms in this work have their higher relevance. Although some authors have carried out these two phases (feasibility analysis and frequency scaling computation) separately, these analyses are strongly related and in some cases can be performed together.In this thesis, we present novel algorithm that perform feasibility analyses and compute new processor static frequencies based on dynamic voltage and frequency scaling techniques (DVFS). The frequency obtained as result of applying the algorithm proposed is the minimum processor frequency that minimizes CPU energy consumption while guaranteeing the fulfilment of real-time system constraints. The algorithm uses fixed priority scheduling schemes with deadlines less than, or equal to, the period of the tasks. Other propose of this algorithm is the use on-line during th...

show abstract

Exploiting laxity for heterogeneous multiprocessor real-time scheduling

Cited by 4 publications

References 8 publications

Investigating the effect of processor affinity on uniform parallel machine scheduling

Investigating the effect of processor affinity on uniform parallel machine scheduling

High performance dynamic voltage/frequency scaling algorithm for real-time dynamic load management

Gestión de la carga dinámica de tareas de tiempo real con criterios de ahorro energético y su aplicación en el desarrollo de un middleware de control.

Contact Info

Product

Resources

About