Post-Doctorant F/H Euro-3C : data management across computing continuum

A central challenge in the transition from cloud to edge computing is enabling distributed applications-originally designed for cloud platforms-to capitalize on edge opportunities while mitigating the limitations of wide-area network links. To address this, the STACK team proposed federating modular microservice instances (one per edge site or tenant) and dynamically orchestrating collaborations among them for each request through a domain-specific language (DSL), managed transparently by the Cheops runtime. Having demonstrated the effectiveness of this approach on both OpenStack and Kubernetes, we now aim to extend it to distributed storage systems, enabling simple and transparent cross-location dataset management. Within task T3.2 ‘Data Interoperability and Software Distribution Subsystems’, led by Capgemini, STACK proposes to leverage the Cheops concept and deliver a CLI for DevOps teams to manage datasets across multiple domains (i.e., distinct REST-based storage services) by abstracting data locations at the CLI level. Such a tool will allow DevOps teams to manage the full lifecycle of datasets across the Euro-3C infrastructure.

Further information at: https://inria.hal.science/hal-04522961v1 and https://gitlab.inria.fr/discovery/cheops/

Principales activités

Phase 1:

• Understanding the Euro-3C reference architecture
• Getting up to speed on the REST APIs of major storage systems and on the APIs selected by the project (e.g., CAMARA)
• Familiarization with the Cheops concepts and analysis of the proof of concept built around the Redis system

Phase 2:

• Proposal of a proof of concept within Euro-3C for managing datasets across several independent systems
• Writing scientific papers presenting the design choices and their evaluation
• Contributing to the writing of deliverables in line with the expectations of task T3.2

Phase 3: (depending on the results of the two previous phases)

• Transfer of the solution to the Euro-3C system developed by the consortium’s industrial partners
• Extending the proposal with automated loops to manage dataset lifecycle
• Integration with the “local-first” indexing system proposed by the STACK team

During all three phases, the successful candidate will participate in the project’s regular meetings, notably those of WP1 and WP3. Attendance at plenary meetings across Europe is expected also.

• Knowledge on distributed file systems/distributed storages (Ceph, S3, IPFS, Redis, etc.)
• Python, Go
• Software engineering
• Cloud/Edge

Technical skills (a plus):

• Experiment driven research (Chameleon, SLICES, enoslib, etc.)
• Kubernetes
• Camara
• Knowledge of softwarization of networks (NDN, SDN, NFV/NFC…)

Languages:

• English (required for scientific dissemination)
• French (a plus for interactions with acquisition sites)

Relational skills:

• Strong organisational and coordination skills
• Ability to work in a collaborative and interdisciplinary environment
• Autonomy and initiative in managing field campaign

• Restauration subventionnée
• Transports publics remboursés partiellement
• Congés: 7 semaines de congés annuels + 10 jours de RTT (base temps plein) + possibilité d’autorisations d’absence exceptionnelle (ex : enfants malades, déménagement)
• Possibilité de télétravail (après 6 mois d’ancienneté) et aménagement du temps de travail
• Équipements professionnels à disposition (visioconférence, prêts de matériels informatiques, etc.)
• Prestations sociales, culturelles et sportives (Association de gestion des œuvres sociales d’Inria)
• Accès à la formation professionnelle