3 ms·
This article leaves out the new generation of distributed consensus algorithms used by databases like Spanner and my employer FaunaDB. You can read a compariso
by jchanimal 9y ago
This article leaves out the new generation of distributed consensus algorithms used by databases like Spanner and my employer FaunaDB.
You can read a comparison between Spanner's atomic-clock based approach and FaunaDB's consensus algorithm here: https://fauna.com/blog/distributed-consistency-at-scale-spanner-vs-calvin https://fauna.com/blog/distributed-consistency-at-scale-span...
And if you want to look at how these consistency tradeoffs show up in application programming, check out the bit about serialized indexes in the second half of this blog post: https://fauna.com/blog/consistent-transactions-in-a-globally-distributed-database https://fauna.com/blog/consistent-transactions-in-a-globally...
- zzzcpan 9y agoYou are leaving out CockroachDB from that group as well. But it's not really a new generation of consensus algorithms. There is still Paxos, 2PC, etc., just more granularity. The trade offs article discusses still apply though, you still trade something in each case for consistency, including latency. It's a shame these "NewSQL" databases attack AP databases, claiming that most projects don't need them. While in reality latency is super important and consistency is something of an afterthought, even for hardcore RDBMS users.