Animation and exchanges in the Petri Nets community

[address@concealed]

We seek a post-doctoral researcher to join the SmartCloud project, an ANR-funded
initiative addressing critical challenges in cloud computing automation.

The position is for a period of 12 or 18 months and can be based either in Lille
or Sophia Antipolis (France), with the possibility of research stays at the
University of Bologna (Italy), depending on the candidate’s preferences. This
flexibility is intended to encourage applications from candidates who may favor
one base location or the other, including stays in Bologna/Italy. The postdoc
will collaborate with the members of the OLAS INRIA research team involved in
the SmartCloud project, who are based at the University of Bologna.

# The SmartCloud project

Modern cloud systems require constant manual intervention from support teams who
monitor thousands of components, diagnose performance issues, and coordinate
complex adaptations across distributed infrastructures. This manual approach
cannot scale with growing cloud demands and increasingly complex multi-cloud
architectures.

The SmartCloud project develops automatic adaptation frameworks that coordinate
both platform-level resource management and application-level behavioural
changes. The project focuses on creating systems that can reason about component
interdependencies, predict cascading effects, and execute coordinated
adaptations without human intervention. SmartCloud’s approach combines formal
modelling techniques with practical deployment optimisation to ensure both
correctness and efficiency.

# Research Teams

Three research teams collaborate on this project. The OLAS team at Inria
Sophia-Antipolis and the University of Bologna leads the optimisation research,
bringing expertise in the automatic deployment of component-based applications and
microservice architectures. The SPIRALS team at Inria Lille contributes
coordination mechanisms and behavioural modelling techniques, particularly
through their work on BIP coordination frameworks and OCCI cloud standards.
SCALAIR, the project's industrial partner, provides real-world cloud operations
context through their management of virtual machines and physical systems,
offering both use cases and evaluation environments.

# Research Activity

The post-doctoral researcher is expected to contribute to one or more of the
following interconnected research activities:

Developing constraint-solving and optimisation techniques for global cloud
application deployment. The work regards developing solutions that use
constraint programming and optimisation techniques to automatically generate
optimal deployment plans that respect component dependencies, resource
constraints, and domain-specific policies.

Designing heuristics for rapid local adaptations. This activity regards the
development of knowledge-based systems that rapidly select pre-computed
adaptation strategies based on monitored key performance indicators. This work
particularly addresses the coordination challenges in microservice
architectures, where uncoordinated scaling can create cascading failures
throughout the system.

Implement prototypes that automatically translate deployment specifications into
executable plans. This activity can also include developing declarative
languages for specifying deployment policies and creating mechanisms that
trigger specific deployment actions when performance thresholds breach
acceptable levels.

Develop and assess choreographic languages for adaptation. Choreographic
languages allow one to describe full distributed systems in a single artefact,
focusing on the interactions between the distributed components. In particular,
we are interested in building on the existing AIOCJ and Choral languages to make
them suitable for adaptive multi-cloud scenarios.

# Expected Candidate Profiles

We require candidates with a PhD in Computer Science, Software Engineering, or
closely related fields, demonstrating expertise in formal methods, distributed
systems, and cloud computing. Valuable technical skills for the profile sought
after include constraint programming and optimisation techniques, proficiency in
mainstream programming languages, like Java, C++, and Rust, and hands-on
experience with containerisation technologies such as Kubernetes and Docker.
Familiarity with microservice/serverless architectures, formal system modelling,
and coordination mechanisms is also a valuable asset for the position.

# Contacts and Information

Contact one of the members of the SmartCloud OLAS research unit:
Prof. Gianluigi Zavattaro (address@concealed)
Prof. Ivan Lanese (address@concealed)
Prof. Saverio Giallorenzo (address@concealed)