IMT Atlantique, internationally recognized for the quality of its research, is a leading general engineering school under the authority of the Ministry of Industry and Digital Technologies, ranked in the 3 main international rankings (THE, SHANGAI, QS). With three campuses in Brest, Nantes and Rennes, IMT Atlantique aims to combine digital technology and energy to transform society and industry through training, research and innovation.
The Department of Automation, Production and Computer Sciences (DAPI in French) is part of the current revolution bridging the gap between the cyber and physical worlds. Our aim is to create adaptable industrial systems making the best possible use of the available resources, while remaining true to our philosophy of sustainable development. The department expertise spans a wide range of fundamental and applied skills including mathematics, operational research, computer science, robotics and automation. STACK, where the postdoc will take place, is a research group focusing on challenges related to the management and advanced usages of Utility Computing infrastructures (i.e. Cloud, Fog, Edge, and beyond).
This position is funded by the ANR SeMaFoR (Self Management of Fog Ressources) project which intends to model, design and develop a decentralized solution for self-management of Fog resources in a generic way. The Fog computing is one of the new decentralized paradigms of utility computing (the Edge could also be mentioned) that overcomes the limitations of centralized Cloud computing by leveraging smaller resources closer to the end-users and to the data sources. Hence, the Fog computing is adapted to latency-critical applications and systems (e.g., 5G, autonomous cars, virtual reality).
However, handling a pool of geo-distributed heterogeneous resources is challenging and the usual operating systems of the Cloud (e.g., OpenStack) are not easily adaptable to this purpose. The heart of SeMaFoR is to tackle Fog resources by means of a set of autonomous controllers, one for each area of the pool, where areas could be geographical or logical. This set of controllers collaborate to manage in an efficient manner the set of shared resources.
An autonomous controller follows an infinite loop, often referred as the MAPE-K model, that continuously monitors the area (i.e., the set of resources, applications and systems), take decisions on which new state to reach according to the current situation, and then plan and execute a reconfiguration plan according to the current and desired state.
The selected candidate for this position will work on the collaboration of multiple controllers when building a reconfiguration plan. Indeed, if each area knows its desired state, the reconfiguration plan is in practice not fully decoupled from other areas that may use services or resources hosted by different areas, thus requiring coordination.
In recent work, we have designed a new reconfiguration model and language, namely Concerto, that enhances the safety and efficiency of a reconfiguration. Thanks to its formal specification, we have been able to automatically synthesize correct reconfiguration plans from a partial desired state. We have used SMT solvers for this purpose. However, this work has considered a centralized synthesis of a reconfiguration plan.
The selected candidate will contribute to the decentralization of a reconfiguration plan synthesis. In particular, the candidate will study the use of constraint programmingand distributed constraint satisfaction problems (DCSP). The selected candidate will also actively participate in the SeMaFoR project meetings and will collaborate with the partners of the project.