Description
Date depot: 9 novembre 2021
Titre: Optimising Coordination in Concurrent and Geo-Distributed Systems
Directeur de thèse:
Mesaac MAKPANGOU (LIP6)
Directeur de thèse:
Marc SHAPIRO (LIP6)
Encadrant :
Julien SOPENA (LIP6)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: The objective of this PhD is develop a practical approach to optimizing concurrency control in actual, complex applications. The PhD student will aim to develop a tool that assists the application developer with implementing coordination. It shall analyze the application and its requirements, identify what coordination is necessary for correctness, analyze the performance space based on models and on real traces, and identify the trade-offs relevant to the developer’s performance objectives, under the constraint of maintaining correctness.
The research will build upon the Coordination Lattice approach, applying it to actual distributed applications with non-trivial invariants and interesting workloads. It will extend the current theory to a more comprehensive set of coordination primitives (e.g., leases or consensus), consider the additional dimensions of liveness and fault tolerance. The modelling of environmental constraints, such as workload or network latency, will use either real traces, or a statistical model, or even machine learning. The aim is to be able to predict robust real-world performance metrics.
Doctorant.e: Pandey Ayush