News for hackers. Stuff that matters.
Re:A generalised solution to distributed consensus by karl_gluck on Saturday March 09, 2019 @08:52AM (#19346086)
Is anyone else surprised at the relatively mild reaction to this result here on Hacker News? Only a couple dozen comments, and many are addressing other topics.

Can we take a second to appreciate that this result will rapidly become the first thing taught in every single distributed systems class? And if this holds as a generalization of trustful distributed consensus as a field, then she has defined its Turing Machine equivalent. And it is even remarkably easy to understand! That is wild!!

Props to the authors. Personally, this is going on the top of my “to memorize” list.

Re:A generalised solution to distributed consensus by disposablename on Saturday March 09, 2019 @03:10PM (#19348191)
This is a very techincal topic, of course the vote to comment ratio is going to be very high.
Re:A generalised solution to distributed consensus by Vervious on Saturday March 09, 2019 @02:15PM (#19347872)
I disagree. This paper generalizes variants of Paxos - certainly not the class of all consensus algorithms. Moreoever, it generalizes Paxos variants, with different requirements - weaker quorums are always a tradeoff with liveness in the presense of faults.

tldr; I don't see world-changing novelty here, theoretically. There are dozens of other consensus algorithms in the world, some randomized, some not quorum based, etc. This is interesting if you're in Paxos land, trying to tune your Paxos for different fault-tolerance requirements. It's no way comparable to a computational model.

Re:A generalised solution to distributed consensus by DannyBee on Saturday March 09, 2019 @03:00PM (#19348138)
"This paper generalizes variants of Paxos - certainly not the class of all consensus algorithms."

This is not true. It generalizes quite many more than that. It happens to use PaxOS as examples.

Re:A generalised solution to distributed consensus by Vervious on Saturday March 09, 2019 @04:26PM (#19348584)
It generalizes specifically a class of consensus algorithms that write to sequences of registers in a monotonic way, such that if a "previous" register decides v, future "registers" and associated quorums will decide v.

This is not a property of all consensus algorithms. Consensus algorithms don't need to behave this way. Paxos and derivatives behave this way, but I'm not sure Raft, etc. slide in as neatly (that would require a substantial analysis).

In particular, this generalization does not capture any nondeterministic algorithm, any algorithm tolerating Byzantine faults (or various other consensus problems), any non-quorum based algorithm, Nakamoto style (which is a byzantine consensus algorithm), Ben-Orr, etc. etc.

So, my question for you: what other algorithms do you think this paper generalizes, that are useful outside of the narrow scope of optimizing Paxos?

Re:A generalised solution to distributed consensus by anigbrowl on Saturday March 09, 2019 @03:12PM (#19348198)
Not really, in fact it's something I've come to expect here. There's a loss in social status for appearing too enthusiastic about a new thing (that might not be so great) and if the new thing replaces a difficult-to-use old thing then the people who invested effort in understanding the old thing don't want to participate in making it obsolete.

Don't forget that people often just want to use an algorithm and are not necessarily that interested in the underlying mathematical structure, only in whether it works reliably and does a better job than the existing solution.

Re:A generalised solution to distributed consensus by brighton36 on Saturday March 09, 2019 @09:49AM (#19346353)
I'm very skeptical that many people will want this. Data misrepresentation is rarely (never?) due to a fault in the persistence medium. Its usually a political reason, or human error that perverts consensus.
Re:A generalised solution to distributed consensus by woah on Saturday March 09, 2019 @11:25AM (#19346837)
What? Consensus algorithms in this class are used everywhere in web programming, from MySQL to AWS. Ironic that your comment may very well have touched several consensus algorithms on its way to my eyes. The fact that you remain blissfully unaware of them is a testament to their power.
Re:A generalised solution to distributed consensus by atombender on Saturday March 09, 2019 @03:10PM (#19348189)
This has nothing to do with "data misrepresentation". It's about distributed consensus. It's not even really about storage, but about multiple parties agreeing on a single truth.
Re:A generalised solution to distributed consensus by eru on Monday March 11, 2019 @10:43AM (#19359322)
Even just something as simple as which of several replicas is supposed to be the master. Not some great external truth about the world.