Blog

We released our first major stable version, Garage v1.0, almost a year ago. So what's next for Garage? A graphical user interface for cluster administration! Share your experience and feedback so that we can provide the best possible user experience.

Deuxfleurs has benefitted multiple times from European grants via the NGI project, for the developpement of Garage and Aerogramme, two pieces of software that we have developped for the needs of our association. Today, these grants are in peril, as the European Commission wishes to finance AI projects instead. We relay and sign an open letter from our friends at petites singularités, that asks that the NGI project be maintained, as it provides great assistance for the development of free software and commons on the Internet.

Deuxfleurs and IMT Atlantique are partnering to fund a PhD student to work on Garage and distributed systems theory during three years. The recruitment process is open and we are currently looking for candidates. Applications are accepted until Jan 31, 2024. Read for details.

Garage is a data storage system that is based on CRDTs internally. It does not use a consensus algorithm such as Raft, therefore maintaining consistency in a cluster has to be done by other means. Since its inception, Garage has made use of read and write quorums to guarantee read-after-write consistency, the only consistency guarantee it provides. However, as of Garage v0.9.0, this guarantee is not maintained when the composition of a cluster is updated and data is moved between storage nodes. As part of our current NLnet-funded project, we are developing a solution to this problem. This blog post proposes a high-level overview of the proposed solution.

Consensus algorithms such as Raft and Paxos, which are used in many distributed databases, have notoriously unpredictable performance in low-quality networks that suffer from latency, jitter, packet loss and/or unavailable nodes, which is why Garage does not use them and uses only CRDTs. A new paper by Antoniadis et al., Leaderless Consensus, introduces a new category of algorithms that better tolerate the frequent unavailability of a subset of nodes. However, additional research and practical work is required before these results can be put into practice. Read for more details.