Re:Introducing Cloud Spanner, a Global Database Service
by jasonwatkinspdx on Tuesday February 14, 2017 @04:23PM
(#13647331)
You're making some assumptions. For example if you use epaxos for the quorums there is no unavailability due to a leader failure and re-election. Even then, re-election is likely going to be a lot faster than any sort of fault tolerant 2PC coordinator recovery protocol. You're exaggerating the write amplification. The above offering is single region. Google says they do 5 way geo in their various papers/etc, and there's no way they're paying some cumulative 15 copy write amplification as you imply.
[ Parent
| on HN ]
Re:Introducing Cloud Spanner, a Global Database Service
by findateamfirst on Wednesday February 15, 2017 @09:16AM
(#13652109)
I've worked on a storage system that used 9. 15 doesn't seem impossible to me.
Re:Introducing Cloud Spanner, a Global Database Service
by krenoten on Tuesday February 14, 2017 @05:28PM
(#13647863)
epaxos is not general purpose replication. It leaves the problem of reconciling multiple replication streams to another layer. I've heard that they do 20+WA (distributed, before you consider on-disk WA) for much of it. shrug