Derivation of Parallel and Resilient Programs from Simulink Models

A4 Konferenspublikationer


Interna författare/redaktörer


Publikationens författare: Sergey Ostroumov, Pontus Boström, Marina Waldén
Redaktörer: Lisa O'Conner
Förläggare: Conference Publishing Services
Publiceringsår: 2015
Förläggare: IEEE
Moderpublikationens namn: Parallel, Distributed and Network-Based Processing (PDP), 2015 23rd Euromicro International Conference on
Artikelns första sida, sidnummer: 416
Artikelns sista sida, sidnummer: 420
ISBN: 978-1-4799-8492-3
ISSN: 1066-6192


Abstrakt

Modern embedded applications often require high computational power and, on the other hand, fulfilment of realtime constraints and high level of resilience. Simulink is one widely used tool for model-based development of embedded software. In this paper, we focus on the derivation of parallel programs from Simulink models and real-time resilient execution of derived implementations on a many-core platform. The main contribution is a fault-tolerance (FT) mechanism that prevents data loss when the platform is dynamically reconfigured to mask failures of individual cores. Finally, we evaluate the proposed solutions on an industrial case study using a commercially available NoC-based platform. The evaluation shows that the proposed FT mechanism has a marginal overhead.

Senast uppdaterad 2019-16-11 vid 01:17