Phd Position F - M Multi-Tenancy In Resource-Constrained Fog - Edge Platforms For Natural Environment Observation H/F

INRIA

PhD Position F/M Multi-tenancy in resource-constrained fog/edge platforms for natural environment observation
Le descriptif de l'offre ci-dessous est en Anglais
Type de contrat : CDD

Niveau de diplôme exigé : Bac +5 ou équivalent

Fonction : Doctorant

A propos du centre ou de la direction fonctionnelle

The Inria Centre at Rennes University is one of Inria's eight centres and has more than thirty research teams. The Inria Centre 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.

Contexte et atouts du poste

This PhD thesis topic integrates in a multi-disciplinary research effort between hydrologists who are interested in monitoring, studying and modeling the behavior of natural environment, and computer scientists who are interested in developing a fog/edge platform to support the computational requirements of natural observatories. It follows previous work on the LivingFog platform which was recently deployed in the university campus in Rennes as well as in the Himalaya mountains in Nepal [1]. The PhD student will be co-supervised by a hydrologist (Laurent Longuevergne) and a computer scientist (Guillaume Pierre) considering that the expected research lies at the boundary between the two research domains. On the one hand the thesis aim to address specific needs experienced by existing natural environment observatories and inventing solutions will require a deep understanding of the expected usage by the hydrologists; on the other hand, the thesis will also aim to address broader questions relevant in the domain of fog and edge computing out of the environment monitoring use case.

Mission confiée

Nature is complicated, and it can occasionally be dangerous. For instance, a seemingly peaceful river may experience a large enough flood to threaten the local human population within minutes, either in urban environments or in remote locations. These catastrophic events often result from cascading effects where a relatively minor root cause such as a storm may result in major and hard-to-predict consequences located far from the initial incident.

To better anticipate such events, environmental scientists operate so-called environmental observatories dedicated to the short-term and long-term observation of a well-defined region of interest [1]. Long-term observation data may be useful for identifying trends such as the impact of climate change, or for building accurate models of the environment's response to specific events. On the other hand, short-term observation may help quickly identifying events when they occur, analyzing the detected events, and issuing timely and accurate warnings to relevant authorities.

Environmental observatories are composed of a number of sensors of various types placed in relevant locations. Typical observatories may exploit dozens to hundreds of sensors which communicate their produced data within the observatory to a local data logger which is in charge of buffering the data until they can be sent to a remote cloud for further analysis. Depending on local conditions, data transmission may take place periodically such as once per day. Although this is a perfectly suitable method for long-term data collection, a different approach is necessary to handle real-time data analysis. To enable quick event detection and analysis, a promising approach proposes to enhance the environmental observatories using edge/fog computing technologies to process sensor data locally within the observatory itself [2].

An interesting characteristic of fog platforms for natural environment observation is that they are expected to be fundamentally multitenant, as numerous researchers from different fields of sciencemay share the same fog platform to perform different data processing tasks based on the same set of data sources [3]. These usershave an interest in sharing the same sensor data and fog/edge execution platform, but at the same time each user has their personal research agenda. This means for instance that some users may refuse sharing data originating from their own sensors until they had the opportunity to publish their own research results first. Also, many environmental fog/edge platforms are strongly resource-constrained because they must be powered exclusively using local solar panels. Multitenancy may therefore generate issues in periods of resource scarcity due to low energy production which makes it difficult to define the relative priority and resource allocation of different applications belonging to different users.

[1] Brantley, S.L., McDowell, W.H., Dietrich, W.E., White, T.S., Kumar, P., Anderson, S.P., Chorover, J., Lohse, K.A., Bales, R.C., Richter, D.D., Grant, G., Gaillardet, J.: Designing a network of critical zone observatories to explore the living skin of the terrestrial Earth. Earth Surface Dynamics 5(4) (2017).

[2] Kazem, A., Pierre, G., Longuevergne, L.: Enhancing environmental observatories with fog computing. Frontiers in Environmental Science 13 (2025).;

[3] Nicolas, M., Mühl, E., Andermann, C., Pierre, G. LivingFog: In-Situ Environmental Data Processing for Urgent Event Detection and Analysis. Chapter in Urgent Computing across the Device-Edge-Cloud Continuum: Concepts, Applications, and Future Directions. Springer, to appear.

[4]Chih-Kai Huang, Guillaume Pierre. UnBound: Multi-Tenancy Management in Scalable Fog Meta-Federations. UCC 2024 - 17th IEEE/ACM International Conference on Utility and Cloud Computing, Dec 2024.;

[5] Bonomi, Flavio et al. (Aug. 2012). Fog Computing and Its Role in the Internet of Things. In: Proceedings of the FirstEdition of the MCC Workshop on Mobile Cloud Computing.

Principales activités

The objective of this thesis is to investigate and answer multiple questions related to the necessary multitenancy of resource-constrained fog/edge platforms for natural environment observation:

- How can multiple users of the same platform share raw dataas well as pre-processed data in such a way that they cannot negatively interfere with each other (voluntarily or not)?
- How can we dynamically allocate the limited available computational resources ofeach application in a context where the sum of demands by the different users may largelyexceedthe capacity of the platform?
- How can we manage an application's lifecycle (e.g., application initial deployment, upgrades, retirement) when multiple users depend on the same outputs to feed their respective applications and there may be inter-dependencies between applications belonging to different users?

Compétences

- A master degree in distributed systems and/or cloud/edge/fogcomputing.
- Excellent programming skills in Linux environments.
- Excellent communication and writing skills.
- Good command of English (reading, writing, listening, speaking).
- Knowledge of the following technologies is not mandatory but will be considered as a plus:

- Cloud resource scheduling
- Distributed container systems: Kubernetes, Docker Swarm.
- Single-board computers such as Raspberry PI
- Python and shell scripting
- Revision control systems: git, svn.
- Linux distributions: Debian, Ubuntu.

Avantages

- Subsidized meals
- Partial reimbursement of public transport costs
- Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
- Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
- Professional equipment available (videoconferencing, loan of computer equipment, etc.)
- Social, cultural and sports events and activities
- Access to vocational training
- Social security coverage

Rémunération

monthly gross salary 2300 euros