Renewable contract : Oui
Level of qualifications required : PhD or equivalent
Fonction : Post-Doctoral Research Visit
About the research centre or Inria department
The Inria Rennes - Bretagne Atlantique Centre is one of Inria's eight centres and has more than thirty research teams. The Inria Center is a major and recognized player in the field of digital sciences. It is at the heart of a rich R&D and innovation ecosystem: highly innovative PMEs, large industrial groups, competitiveness clusters, research and higher education players, laboratories of excellence, technological research institute, etc.
Context
Social advantages
- Subsidised catering service
- Partially-reimbursed public transport
- Social security
- Paid leave
- Flexible working hours
- Sports facilities
Salary - Duration
- 18 months contract
- Monthly net salary amounting to ~2160 euros
Assignment
Context
The embedded systems from the safety-critical domain industries, such as the avionics, auto-motive, space, healthcare or robotics industries, face exponential growth in terms of performance requirements, while they have to deal with strict hard real-time constraints. This constantly growing processing demand has led the processor manufacturing industry towards multi-/many-core architectures. These architectures have multiple processor elements, called cores, providing massive computing power by concurrently executing a high volume of tasks. While such architectures can successfully meet the demands for the majority of computing systems, the same can not be argued for hard real-time systems [1, 2]. Hard real-time systems have to provide timing guarantees in order to be safe, i.e. guarantee that tasks are completed before their respective deadlines and/or the total execution does not exceed a given latency requirement. Typical examples of such safety-critical systems is the Automatic Braking System (ABS) in automotive and the Fly-by-Wire control system in avionics.
In order to rigorously provide such guarantees, application deployment approaches, i.e. task map-ping/scheduling, are based on the a priori knowledge of Worst-Case Execution Time (WCET) of tasks. There is a plethora of research in WCET estimation for uni-processor systems (see [2] for a review). However, in multi-core architectures, several arbitrated resources are shared among the cores(memories and interconnects) introducing timing delays and changing the timing behavior in a non-deterministic way. Thus, the WCET varies according to the task deployment, as tasks interfere when simultaneously access the on-chip shared resources. As a result, timing analysis and deployment optimisation for multicore systems becomes very challenging [3, 4, 5, 6, 7]. This effect is particularly apparent in data-parallelisable applications, due to extensive resource sharing. We have observed that the WCET of tasks including interferences can be 750% times larger than the corresponding estimations without interferences [8] [9]. In order to extend the uni-processor deployment approaches to multi-processor architectures, the WCET of the tasks has to be over-approximated, so as to account for all possible interferences. This over-approximation practice has lead to the “one-out-of-m processors” problem [1],where the additional processing capacity is negated by the pessimism of the WCET. As a result, the sequential execution (on a single core) potentially provides better timing guarantees than any parallel execution, which seriously undermines the advantages of utilizing multi-cores. Nevertheless, recent state-of-the-art research [10] has shown that context-dependent WCETs, called interference-sensitive WCET (isWCET), reduce the pessimism in WCET leading to more efficient deployments on multi-core architectures.
In this work, we explore mechanisms to provide adaptation of the execution of the safety-critical systems so as to improve the actual run-time system performance, while still meeting the real-time constraints. This performance improvement allows the systems to provide higher Quality-of-Services or execute other best-effort applications. Our approach will efficiently use the information being available during the real execution of the system in order to improve performance, energy consumption, QoS and reliability.
References
[1] N. Kim, B. C. Ward, M. Chisholm, C.-Y. Fu, J. H. Anderson, and F. D. Smith, “Attacking the one-out-of-m multicore problem by combining hardware management with mixed-criticality provisioning,” in Real-Time and Embedded Technology and Applications Symposium (RTAS), 2016 IEEE, pp. 1–12, IEEE, 2016.
[2] R. Wilhelm, J. Engblom, A. Ermedahl, N. Holsti, S. Thesing, D. Whalley, G. Bernat, C. Ferdinand, R. Heck-mann, T. Mitra,et al., “The worst-case execution-time problem-overview of methods and survey of tools,”ACM Transactions on Embedded Computing Systems (TECS), vol. 7, no. 3, p. 36, 2008.
[3] M. Bertogna,Real-time scheduling analysis for multiprocessor platforms. PhD thesis, 2008.
[4] S. Cotton, O. Maler, J. Legriel, and S. Saidi, “Multi-criteria optimization for mapping programs to multi-processors,” in Industrial Embedded Systems (SIES), 2011 6th IEEE International Symposium on, pp. 9–17,IEEE, 2011.
[5] P. Tendulkar, P. Poplavko, I. Galanommatis, and O. Maler, “Many-core scheduling of data parallel applications using smt solvers,” in Digital System Design (DSD), 2014 17th Euromicro Conference on, pp. 615–622,IEEE, 2014.
[6] J. Legriel, C. Le Guernic, S. Cotton, and O. Maler, “Approximating the pareto front of multi-criteria optimization problems.,” in TACAS, Springer, 2010.
[7] P. M. B. de Sousa, Real-Time Scheduling on Multi-core: Theory and Practice. PhD thesis, Universidade do Porto (Portugal), 2013.
[8] A. Kritikakou, C. Rochange, M. Faug`ere, C. Pagetti, M. Roy, S. Girbal, and D. G. P ́erez, “Distributed run-time WCET controller for concurrent critical tasks in mixed-critical systems,” in Proceedings of the22nd International Conference on Real-Time Networks and Systems, p. 139, ACM, 2014.
[9] S. Skalistis and A. Simalatsar, “Near-optimal deployment of dataflow applications on many-core platforms with real-time guarantees,” in2017 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp. 752–757, IEEE, 2017.
[10] S. Skalistis, Efficient Adaptive Hard Real-time Multi-processor Systems. PhD thesis, 2017.2
Skills
- PhD or Master in Computer Science, Electrical or Computer Engineering
- Programming experience, e.g., C/C++ language, HDL languages is a plus
- Computer architecture, hardware design, embedded software development, embedded systems.
- Mostly importantly, we seek highly motivated and active researchers.
Benefits package
- Subsidized meals
- Partial reimbursement of public transport costs
Remuneration
monthly gross salary amounting to 2653 euros
Share
General Information
- Theme/Domain :
Architecture, Languages and Compilation
System & Networks (BAP E) - Town/city : Rennes
- Inria Center : CRI Rennes - Bretagne Atlantique
- Starting date : 2021-01-01
- Duration of contract : 1 year, 6 months
- Deadline to apply : 2021-01-30
Contacts
- Inria Team : CAIRN
-
Recruiter :
Kritikakou Angeliki / angeliki.kritikakou@irisa.fr
About Inria
Inria is the French national research institute dedicated to digital science and technology. It employs 2,600 people. Its 200 agile project teams, generally run jointly with academic partners, include more than 3,500 scientists and engineers working to meet the challenges of digital technology, often at the interface with other disciplines. The Institute also employs numerous talents in over forty different professions. 900 research support staff contribute to the preparation and development of scientific and entrepreneurial projects that have a worldwide impact.
Instruction to apply
Please submit online : your resume, cover letter and letters of recommendation eventually
For more information, please contact olivier.sentieys@inria.fr
Defence Security :
This position is likely to be situated in a restricted area (ZRR), as defined in Decree No. 2011-1425 relating to the protection of national scientific and technical potential (PPST).Authorisation to enter an area is granted by the director of the unit, following a favourable Ministerial decision, as defined in the decree of 3 July 2012 relating to the PPST. An unfavourable Ministerial decision in respect of a position situated in a ZRR would result in the cancellation of the appointment.
Recruitment Policy :
As part of its diversity policy, all Inria positions are accessible to people with disabilities.
Warning : you must enter your e-mail address in order to save your application to Inria. Applications must be submitted online on the Inria website. Processing of applications sent from other channels is not guaranteed.