Facebook Announces Apollo, a New NoSQL Database for On-line Low Latency Storage

Post by dahjelle » Sat, Jun 14, 2014, 3:01 AM UTC

platz wrote: post #7

spiralganglion wrote: post #3

One of their supported storage primitives is CRDT-based, according to [1]. I, for one, am really interested to see how this works in practice. I've been quite excited about CRDTs, but haven't seen enough examples of them in the wild to get a sense of their drawbacks — for instance, how difficult it is to use them to model various processes or data structures. [1] https://twitter.com/adrianco/status/476843040330743809

Do you know of a resource for learning the basics CRDTs that doesn't require a PHD?

Post by msrivas » Sat, Jun 14, 2014, 3:42 AM UTC

platz wrote: post #7

spiralganglion wrote: post #3

One of their supported storage primitives is CRDT-based, according to [1]. I, for one, am really interested to see how this works in practice. I've been quite excited about CRDTs, but haven't seen enough examples of them in the wild to get a sense of their drawbacks — for instance, how difficult it is to use them to model various processes or data structures. [1] https://twitter.com/adrianco/status/476843040330743809

Do you know of a resource for learning the basics CRDTs that doesn't require a PHD?

Post by spiralganglion » Sat, Jun 14, 2014, 5:12 AM UTC

platz wrote: post #11

seiji wrote: post #9

Earlier quoted context omitted.

The name is intimidating, but the operations are simple. Basically, your storage has container types ("T"). A list, a set, a dictionary, etc. Container types can be split and added together in a distributed fashion ("R" and "D"). The "C" in CRDT stands for "Convergent and Commutative" to imply your distributed operations can obtain the same value when merged. Quick example: If you have a node with a key pointing to v…

That's helpful, thanks (I've downloaded some crdt videos to watch in the meantime). At the surface they sound like something vaguely resembling an abelian group (+/- inverses), but the conflict resolution stuff is the heart of it I'd guess.

Post by trhway » Sat, Jun 14, 2014, 6:04 AM UTC

spiralganglion wrote: post #3

One of their supported storage primitives is CRDT-based, according to [1]. I, for one, am really interested to see how this works in practice. I've been quite excited about CRDTs, but haven't seen enough examples of them in the wild to get a sense of their drawbacks — for instance, how difficult it is to use them to model various processes or data structures. [1] https://twitter.com/adrianco/status/476843040330743809

Post by noelwelsh » Sat, Jun 14, 2014, 7:06 AM UTC

spiralganglion wrote: post #16

platz wrote: post #11

Earlier quoted context omitted.

That's helpful, thanks (I've downloaded some crdt videos to watch in the meantime). At the surface they sound like something vaguely resembling an abelian group (+/- inverses), but the conflict resolution stuff is the heart of it I'd guess.

Yes, from my (limited but growing) understanding of it, they are indeed similar to abelian groups.