Viewing profile — tommyzli

Just last week I started hacking on a k8s operator for managing postgres roles, grants and databases. I really like the continuous reconciliation for this scenario, especially havi…

I'm curious to hear more about your troubles with alloydb. It's something I just started to test out at work

the last time I gave the Postgres operator space a serious look was about a year ago, and at the time the Zalando operator was far and away the most feature complete and mature. We…

Wow, I wasn't expecting such an in-depth response! It sounds like making user profiles a reference table would solve the cross-shard join problem, but what would the performance im…

1. You'll have to define "slow" - I have a 3TB table where an index only scan takes under 1ms 2. hot_standby_feedback is absolutely safe. I've got 5 hot standbys in prod with that …

Believe it or not these numbers are actually from _after_ me and some others spent a few weeks cleaning up our heavier queries

TIL, this is really good to know! Do you know offhand if this is a new feature, or have I just always been wrong

It's not in the post, but I answered this in a separate thread. RDS doesn't let us provision as many IOPS as we need. Apparently Aurora behaves differently, but I wasn't aware of t…

Nope.

We are living life on the edge to an extent, but we have 5 hot standbys across AZs and regular backups + WAL archives to S3. May not be as durable as EBS, but it's enough for me to…

Apparently I'm living in the twilight zone because I have a vivid memory of reading the Aurora docs and seeing the same limit. Oh well, it's something to consider for the next upgr…

basically what paulryanrogers said. We thought about migrating to Citus, but I don't have a good idea of how to shard our dataset efficiently. If we were to shard by user id, then …

ants_a is correct. Also, our NVMe storage is ephemeral so you aren't recovering from a power loss anyways :)

How would that have worked with multiple replicas cascading from the new primary? Streaming replication doesn't work across versions, so would we have had to build out a tree of ne…

Thanks! We stuck with plain EC2. RDS has a limit of 80,000 provisioned IOPS and our read replicas on Postgres 9.6 would regularly hit near double that during peak