Just-Right Consistency for Scalable File Systems

RainbowFS is a collaborative research project funded by French National Agency for Research (ANR), from April 2017 to March 2022.

The aim of RainbowFS is to investigate an approach to distributed storage that ensures consistency semantics tailored to the application, while retaining scalability and availability.
Existing approaches shoehorn the application design to some predefined consistency model, but no single model is appropriate for all uses. Instead, we propose tools to co-design the application and its consistency protocol. Our approach reconciles the conflicting requirements of availability vs. safety: common-case operations are designed to be asynchronous; synchronisation is used only when strictly necessary to satisfy the application's integrity invariants.
Furthermore, we deconstruct classical consistency models into orthogonal primitives that the developer can compose efficiently, and provide a number of tools for quick, efficient and correct cloud-scale deployment and execution.
Using this methodology, we aim to develop an entreprise-grade, highly-scalable file system, exploring the rainbow of possible semantics.

The partners in the project are:

The project coordinator is Marc Shapiro from UPMC/LIP6.


dotScale 2018 - Marc Shapiro - Just-Right Consistency

Just-right consistency with Antidote

Demo of the CISE tool


  • Marc Shapiro - principal investigator
  • Ilyas Toumlilt - PhD candidate
  • Sreeja Nair - research engineer
  • Brad King - founder, chief engineering architect
  • Olivier Detour - software engineer
  • Dimitrios Vasilas - PhD candidate
  • Renaud Lachaize - assistant professor
  • Vivien Quéma — professor
  • Sébastien Monnet - professor
  • Etienne Mauffret - PhD candidate
  • Pierre Sutra - associate professor
  • Tuanir França Rezende - PhD candidate