Adaptivity and Control of Resources in Embedded Systems (ACTORS)
Researchers: Mikael Lindberg, Anton Cervin, and Karl-Erik Årzén in collaboration with the other 6 core partners.
Duration: February 2008 – January 2011
Funding: EU/IST/FP6 STREP
ACTORS is an EU/IST FP7 STREP on feedback-based resource management and data-flow modeling in embedded systems. The other partners in the project are
- Ericsson (Coordinator) – Johan Eker
- Scuola Superiore Sant'Anna di Pisa – Giorgio Buttazzo
- TU Kaiserslautern – Gerhard Fohler
- Ecole Polytechnique Fédérale de Lausanne – Marco Mattavelli
- Evidence Srl – Paolo Gai
- AKAtech SA – Vincent Noel
- Xilinx – Jörn W. Janneck
Links
ACTORS addresses design of resource-constrained software-intensive embedded systems with high requirements on adaptivity and efficiency. Three techniques have been combined:
- virtualization,
- feedback control, and
- data-flow programming models.
Virtualization techniques such as reservation-based scheduling provide spatial and temporal separation of concerns and enforce dependability and predictability. Reservations can be composed, are easier to develop and test, and provide security support.
Using feedback-based resource management, the resource allocation is based on a comparison of the actual resource utilization of, e.g., a set of activities or tasks, with the desired resource utilization. The difference is used for deciding how the resources should be dynamically allocated. Feedback control makes it possible to deal with uncertainties and variations in a controlled way and provides adaptivity to on-line changes in objectives, external conditions and use cases. By combining feedback control with resource reservations it is possible to handle incorrect reservations, reclaim and redistribute unused resources, and adjust to dynamic
changes in resource requirements.
An adaptive CPU resource management for x86 and ARM multicore platforms has been developed. The system manages CPU reservations (virtual processors) provided by SCHED_EDF, a new Linux scheduler developed within the project.
Execution efficiency and development efficiency require abstractions on a higher level than what is provided with C and threads/priorities. Data-flow models such as actor models provide the proper foundation for implementation of efficient, component based, and adaptive algorithms for both multimedia applications in consumer electronics and industrial control systems and signal processing applications. In ACTORS a design methodology has been developed that combines virtualization, feedback and actors-based dataflow programming. Three demonstrators have been developed:
- multimedia processing on cellular phone terminals,
- robotics control, and
- high-speed video processing for computer vision applications.
Publications
Vanessa Romero Segovia: "Adaptive CPU Resource Management for Multicore Platforms". Licentiate Thesis ISRN LUTFD2/TFRT--3252--SE, Department of Automatic Control, Lund University, Sweden, September 2011.
Vanessa Romero Segovia, Mikael Kralmark, Mikael Lindberg, Karl-Erik Årzén: "Processor Thermal Control Using Adaptive Bandwidth Resource Management". In 18th IFAC World Congress, Milano, Italy, August 2011.
Enrico Bini, Giorgio Buttazzo, Johan Eker, Stefan Schorr, Raphael Guerra, Gerhard Fohler, Karl-Erik Årzén, Vanessa Romero Segovia, Claudio Scordino: "Resource Management on Multicore Systems: The ACTORS Approach". IEEE Micro, 31:3, pp. 72–81, May 2011.
Karl-Erik Årzén, Vanessa Romero Segovia, Stefan Schorr, Gerhard Fohler: "Adaptive Resource Management Made Real". In Proc. 3rd Workshop on Adaptive and Reconfigurable Embedded Systems, Chicago, April 2011.
Karl-Erik Årzén, Vanessa Romero Segovia, Mikael Kralmark, Stefan Schorr, Anand Meher, Gerhard Fohler: "ACTORS Adaptive Resource Management Demo". In Proc. 3rd Workshop on Adaptive and Reconfigurable Embedded Systems, Chicago, April 2011.
Vanessa Romero Segovia, Karl-Erik Årzén, Stefan Schorr, Raphael Guerra, Gerhard Fohler, Johan Eker, Harald Gustafsson: "Adaptive Resource Management Framework for Mobile Terminals—The ACTORS Approach". In Proc. First International Workshop on Adaptive Resource Management, Stockholm, Sweden, April 2010.
Mikael Lindberg: "A Survey of Reservation-Based Scheduling". Technical Report ISRN LUTFD2/TFRT--7618--SE, Department of Automatic Control, Lund University, Sweden, October 2007.
2016-12-07