←back to thread

Upgrading Uber's MySQL Fleet

(www.uber.com)
234 points benocodes | 2 comments | 14 Oct 24 12:08 UTC | HN request time: 0.541s | source
Show context
edf13 ◴[14 Oct 24 12:41 UTC] No.41836969[source]▶
3 million queries/second across 16k nodes seems pretty heavy on redundancy?
replies(3): >>41836989 #>>41837703 #>>41837747 #
1. sgarland ◴[14 Oct 24 14:10 UTC] No.41837703[source]▶
I was going to say, that's absolutely nothing. They state 2.1K clusters and 16K nodes; if you divide those, assuming even distribution, you get 7.6 instances/cluster. Round down because they probably rounded up for the article, so 1 primary and 6 replicas per cluster. That's still only ~1400 QPS / cluster, which isn't much at all.

I'd be interested to hear if my assumptions were wrong, or if their schema and/or queries make this more intense than it seems.

replies(1): >>41838136 #
2. pgwhalen ◴[14 Oct 24 14:57 UTC] No.41838136[source]▶
> assuming even distribution

I don't work for Uber, but this is almost certainly the assumption that is wrong. I doubt there is just a single workload duplicated 2.1K times. Additionally, different regions likely have different load.