I was talking to a friend and said there aren't _that_ many companies with the scale to truly justify FAANGesque solutions. I've only had two projects where a server couldn't just hold all hot data in 256gb. For most, services I've seen, a laptop + hard drive would suffice.<p>I'm curious how big the DBs you touch are, how much ram all your microservices use or other such metrics.
I don't know, probably need to get finops to run a report to find out. More than 256gb but scaled out. Don't need a single virtual machine with that much for most things.<p>I imagine you rarely need a lot of RAM like that on one machine except for ML and some specialist stuff. Maybe a CDN node. Not sure.
For mixed workloads, I find that memory becomes the bottleneck first for running workloads on a given physical host. This changed after NVMe SSds became available – before that it was disk I/O.<p>But pricing per GB is also a factor I have to admit, currently we are outfitting with 1.5 TB memory per CPU.