Post-Doctoral Research Visit F/M [Campagne post-doctorants] - IoT systems in the intelligent compute continuum

Contract type : Fixed-term contract

Level of qualifications required : PhD or equivalent

Fonction : Post-Doctoral Research Visit

Context

The MiMove team at Inria Paris (https://mimove.inria.fr) undertakes research addressing the whole lifecycle of next-generation distributed systems, from their conception and design to their runtime support, with focus on mobile systems. We develop solutions at the intersection of distributed systems and software architectures, and in particular middleware solutions.

Assignment

The proposed postdoc research will more particularly focus on the topic of IoT systems in the intelligent edge-fog-cloud compute continuum. Designing, developing and running applications in the compute continuum requires mastering and managing the complexity of both the infrastructure (compute resources) and the available application components, in terms of distribution, heterogeneity, dynamicity and scale. To this end, the proposed research will aim at integrating intelligence in the continuum, by both facilitating and relying on AI techniques (a continuum for AI, AI for the continuum).

Main activities

In the context identified above, the postdoc researcher is expected to contribute to MiMove’s research in one of the following sub-topics:

Automated development of IoT applications leveraging domain knowledge. Developing IoT applications for the compute continuum by relying on latest programming and execution paradigms (e.g., serverless, microservices, FaaS) brings many benefits to developers, but still requires managing a potentially large number of elementary services/functions. An alternative way for application design relies on methods from the goal-oriented requirement engineering domain, where high-level specifications of applications are elaborated in terms of goals. We leverage application domain knowledge in the form of knowledge graphs and NLP methods to provide automated recommendations to the developer for goal-driven application development.

Scheduling and running IoT systems in the compute continuum. Deploying IoT applications on the resource architecture while managing the constraints, requirements and tradeoffs of resources and QoS is a key problem related to the edge-fog-cloud continuum. Extending our previous results on the placement of data streaming operators on homogeneous, fixed-location resources, we aim to account for heterogeneous resources as well as possibly volatile ones coming from mobile and resource-constrained devices. This calls for dynamic lightweight scheduling algorithms that can tackle the runtime resource variability besides the dynamic application workloads. We aim to explore autonomous predictive scheduling approaches by relying on Reinforcement Learning and following the paradigm of Intent-based networking.

Ensuring interoperability in the compute continuum. To leverage its full potential, IoT applications deployed in the edge-fog-cloud continuum will rely on the composition of resources and system services across multiple providers, which is hampered by heterogeneity (e.g., diverse APIs, protocols, data formats, topologies). This makes interoperability in the continuum an essential problem for which effective solutions are required. Building on our previous results on IoT protocol interoperability, we aim at introducing runtime interoperability and portability solutions for automated composition of resources and system services in the dynamic compute continuum. To this end, we will explore solutions based on domain knowledge and NLP.

Automated Machine Learning in the compute continuum. Automated Machine Learning (AutoML) leverages algorithms and computational capabilities to automate key aspects of the machine learning pipeline, such as feature engineering, model selection, and hyperparameter tuning. We are particularly interested in developing AutoML solutions on IoT data streams that take into account the particularities and needs of the compute continuum.

ML inference in the compute continuum. ML inference systems for industrial plants have stringent performance requirements, while thay may have to cope with potentially limited computational/network resources (such as in rural settings). Additionally, they may host multiple ML inference processes with very different characteristics and priorities (such as ensuring human safety vs. predictive equipment maintenance). We aim at developing ML inference system solutions that are tailored to the compute continuum.

Skills

The candidate should have a PhD in Computer Science with expertise – including experience in the implementation of related software prototypes – in one and possibly several of the following topics:

(Mobile) distributed systems,
Middleware architectures and protocols,
Software engineering,
Internet of Things,
Machine learning,
Knowledge representation,
Cloud/edge computing and scheduling.

Benefits package

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
Professional equipment available (videoconferencing, loan of computer equipment, etc.)
Social, cultural and sports events and activities
Access to vocational training

Remuneration

According to civil service salary scales

Apply for this position

General Information

Theme/Domain : Distributed Systems and middleware
System & Networks (BAP E)
Town/city : Paris
Inria Center : Centre Inria de Paris
Starting date : 2024-10-01
Duration of contract : 1 year, 6 months
Deadline to apply : 2024-05-19

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.

Instruction to apply

Candidate's application file:

• Cover letter highlighting the adequacy of the candidate's background with the proposed research.

• CV.

• List of publications.

• Thesis reports if the thesis has already been defended.

• For candidates who have not yet defended, an attestation from the thesis director reporting on the progress of the thesis or the composition of the jury and the probable date of defense.

• Recommendation letters.

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.

Contacts

Inria Team : MIMOVE
Recruiter :
Georgantas Nikolaos / Nikolaos.Georgantas@inria.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.