Consistency Models (2018)

jepsen.io

90 points·archagon··7 comments

I remember reading Doug Terry's "Replicated Data Consistency Explained Through Baseball" in my distributed systems undergrad course which helped me understand some of the models a little better.

https://www.microsoft.com/en-us/research/wp-content/uploads/...

For any tool authors, jepsen is invaluable piece <3.

A lot of software in the messaging/storage space has undergone jepsen test and I for one am grateful to Jepsen's author Kyle Kingsbury for his incredible work in setting their expectations correct.

Thanks for sharing. Its a shame this wasn't upvoted enough to be on the frontpage for a longer time. I'm glad I read this.

A foolish consistency is the hobgoblin of little minds . . .

https://www.goodreads.com/quotes/353571-a-foolish-consistenc...

> A foolish consistency is the hobgoblin of little minds . . .

Didn't quite get the connection here. Care to elaborate? :)

In the context of comp-sci, I'd say it applies to requiring a Strict Serializable consistency model merely because it's easier to reason about it, even if some other model would be a better fit to the problem at hand.

Double plus if you use that model because you don't know of the existence of the others.

What I thought of immediately!

In the 70s and 80s it was not uncommon to run across a CS paper whose title was a play on some aphorism or poetic quotation. That playfulness is much less common these days.

Reply on news.ycombinator.com