2 ms·
That desire comes from three places: 1. Minimize cognitive load by minimizing the number of things you have to learn. 2. Minimize deployment complexity and de
by api 10y ago
That desire comes from three places:
1. Minimize cognitive load by minimizing the number of things you have to learn.
2. Minimize deployment complexity and dependencies.
3. Complexity is just evil I'm general. Linear increases in complexity result in exponential increases in bugs, vulnerabilities, and failure modes. It's just combinatorics.
- mb4nck 10y agopatroni builds upon etcd and (optionally) haproxy, two rather mature pieces of infrastructure (which can both be made HA on their own if SPOFs are to be avoided). I understand where you are coming from, but having this kind of multi-server-who-is-master knowledge baked into Postgres itself will surely take another couple of releases, if it will be included at all. Probably BDR (bi-directional replication, master-master like logical replication) will be there first, but whether the question is whether it will help a lot for local scale-out workloads (as opposed to glueing two datacenters together and allow transactions on both sides).
- xyzzy_plugh 10y agoNumber 2 is the only point I see merit in. With respect to 3, isn't that an argument against increasing the complexity of postgresql?