Showing posts from February, 2022

Efficient Replication via Timestamp Stability (EuroSys 2021)

This paper introduces Tempo is a leaderless Paxos variant for implementing a state machine replication (SMR) protocol. Leaderless means, there is no dedicated/stable Paxos leader, instead any node can become a proposer any time (allowing many concurrent proposers) and get consensus on a proposal opportunistically. In this blog, we had discussed many examples of leaderless and multileader Paxos variants, including Mencius , MDCC , ePaxos , wPaxos , SDPaxos , Hermes , Atlas . And there are several more that we didn't get to yet. It seems like I've been proven right when I foretold my PhD students in 2016 that we will be seeing a Paxos renaissance. On the flip side, you may ask, why do we need another Paxos variant? My selfish answer is that I love Paxos variants. A more practical answer is that because this one simplifies/streamlines the leaderless SMR ideas a bit more. Recently we have been seeing a convergence of the leaderless SMR ideas with loosely synchronized clocks. This

Deep reading

How many hours does it take you to read a ~10 page research paper? — Murat Demirbas (Distributolog) (@muratdemirbas) February 8, 2022 Twenty years ago, a well-known professor in computer networking field told me that he reviews any paper in 30 minutes. Not just read the paper, but also write the review, mind you. All in 30 minutes! I said "I am slow it takes me 4 hours to read a paper". I lied. It actually took me 8+ hours to read papers, because I was a graduate student and didn't have much background and paper reading experience. Things have improved, but it still takes me 4-8 hours to read a paper and understand everything so I can write an informative conference review or blog post about the paper. (Of course, I am talking about good research papers, not content-free publication for the sake of publication papers.) Maybe because of that early encounter with that flamboyant networking professor, I always felt I am very slow in reading a paper. I checked with fellow

RAMP-TAO: Layering Atomic Transactions on Facebook’s Online TAO Data Store (VLDB'21)

This paper won the Best Industry Paper Award at VLDB 2021 . I really like the paper. It provides a good amalgamation of database and distributed system techniques! This paper is a followup to several previous paper some of the authors on this team had, including "Scalable atomic visibility with RAMP transactions" , "Highly Available Transactions", and "Coordination avoidance" . If we squint a bit more, we can say that the origins of the theme goes back to the "Probabilistically Bounded Staleness (PBS)" paper by Peter Bailis, et al. Peter found a gold mine with the PBS work. His observation that practical systems already have good freshness/consistency properties powered all of his follow-up work. This is an optimistic/opportunistic message. The theory of distributed systems and databases put most of their effort into the worst case scenario, but things are often not that bad in practice. I like this research-continuity. We should have more of th

Popular posts from this blog

Learning about distributed systems: where to start?

Hints for Distributed Systems Design

Foundational distributed systems papers

Metastable failures in the wild

Scalable OLTP in the Cloud: What’s the BIG DEAL?

SIGMOD panel: Future of Database System Architectures

The end of a myth: Distributed transactions can scale

There is plenty of room at the bottom

Distributed Transactions at Scale in Amazon DynamoDB

Dude, where's my Emacs?