Did you know that majority agreement isn’t required by Paxos? In fact, most of the time, the sets of nodes required to participate in agreement (known as quorums) do not even need to intersect with each other. This is the observation which is made, proven and realised in our latest paper draft, Flexible Paxos: Quorum
Tangaroa, is a Byzantine tolerant Raft implementation in Haskell, developed by Christopher Copeland and Hongxia Zhong for a Distributed Systems class at Stanford. The authors apply many of the approaches in PBFT to Raft, allowing for the Byzantine failure of nodes. My interest in this work is how can you stop unnecessary leader elections in Raft algorithm using more
I am incredibly excited to announce that Raft Refloated will be featured in The Morning Paper, by Adrian, as part of a series of 10 posts on distributed consensus. Last month, the morning paper also featured our work on the Databox. Will the raft paper sink or swin? Let’s wait and see.
The January edition of SIGOPS Operating Systems Review is out now and thus is the aptly named “Raft Refloated: Do We Have Consensus?”. This is my first journal paper and I’m really existed to see what the community makes of it. Title: Raft Refloated: Do We Have Consensus? Authors: Heidi Howard, Malte Schwarzkopf, Anil Madhavapeddy and
“ARC: Analysis of Raft Consensus” is now available online as a UCAM technical report. http://www.cl.cam.ac.uk/techreports/UCAM-CL-TR-857.pdf Abstract The Paxos algorithm, despite being synonymous with distributed consensus for a decade, is famously difficult to reason about and implement due to its non-intuitive approach and underspecification. In response, this project implemented and evaluated a framework for constructing fault-tolerant applications,
My undergraduate dissertation “ARC: Analysis of Raft Consensus” will be submitted as a UCAM tech report. A draft is available here and I would be very grateful of any feedback. Title: ARC: Analysis of Raft ConsensusAbstract:The Paxos algorithm, despite being synonymous with distributed consensus for a decade, is famously difficult to reason about and implement due to its non-intuitive