Software — Stack — for Massively Geo-Distributed Infrastructures

logo IMT Atlantique logo inria logo LS2N

Postdoc/Junior researcher position – Data Management and Network/Computing Continuum -




Information global


Contact: - Adrien Lebre: adrien.lebre@inria.fr Duration: 12 to 18 months

Link to the offer


Within the framework of the STEEL project (PEPR Cloud France 2030), the STACK research team is seeking a postdoctoral candidate. Our objective is to focus on the storage and effective processing of data on emerging computing infrastructures

Introduction


Cloud computing has significantly increased the volume of data consumed daily, but it has also centralized somehow the storage of data. With the emergence of edge computing, data storage will become more geo-distributed to account for performance or regulatory constraints. One challenge is to maintain an up-to-date view of available content in such a dynamic distributed storage system, for instance to be able to locate the closest replica of a piece of data. Content Delivery Network (CDN) solutions could be considered due to their ability to cache content near end-users. However, they still rely on a hierarchical and centralized network model. Peer-topeer systems like IPFS are the second class of candidates, but they lack robust guarantees concerning network overheads, data availability, and performance. Our vision entails an intermediate “edge-to-edge” model, aiming to deliver a fully distributed, reliable, and highperformance storage service by leveraging a distributed fleet of edge servers.

Mission


The motivation for this position is to explore the use of a decentralized indexation system that could be used to build next-generation CDNs, and more generally to develop new distributed edge services. The successful candidate is expected to evaluate metrics such as latency, throughput, availability, and resilience in different applications scenarios involving edge-to-edge exchanges and mobility.

Expected skills


The R&D Engineer must already hold a PhD degree in Computer Science, with a solid background in Software Engineering and Distributed Systems. She/he must also have a (very) good knowledge and interest in Cloud/Edge computing. Knowledge in distributed/cloud storage systems is an asset. In addition, strong development and programming skills are highly recommended. Finally, the candidate must have good oral and written communication skills in English, with the aim to publish and present research results in high-level international journals and conferences. Autonomous, curious and strongly motivated candidates are expected.

References


[1] Confais, Bastien, Adrien Lebre, and Benoît Parrein. “An object store service for a fog/edge computing infrastructure based on ipfs and a scale-out nas.” 2017 IEEE 1st International Conference on Fog and Edge Computing (ICFEC). IEEE, 2017
[2] Lebre, Adrien, Brice Nédelec, and Alexandre Van Kempen. “AS-cast: Lock Down the Traffic of Decentralized Content Indexing at the Edge.” International Conference on Algorithms and Architectures for Parallel Processing. Cham: Springer Nature Switzerland, 2022.