Hacker News Reader

Comment by timdorr | Hacker News Reader

Just for reference, since you don't choose your processor explicitly on GCP, but instead choose your zone with homogenous processors, here is their current processor/zone layout: https://cloud.google.com/compute/docs/regions-zones/regions-...

Since some GCP engineers are watching: Presumably we'll see some new zones to provide these processors, or will it be a limited release within existing zones? And if so, will you be moving away from homogenous zones in the future?

View on HN

Good question! We've made it possible for this whitelisted (Alpha) to let you make sure you're creating Skylakes. We'll hopefully have more details to share at NEXT in a few weeks (or immediately if you sign up!).

Disclosure: I work on Google Cloud.

Where's the form to apply for Alpha? :)

https://docs.google.com/a/google.com/forms/d/e/1FAIpQLSevvO7...

Bothers me a bit with "32 cores" listed when they are 16 core/32 Thread machines.

I try to consistently say "vCPUs", but yeah...

Given the marketing toward HPC tasks, shouldn't hyper-threading be disabled?

No? We make sure that pairs of vCPUs (hypertwins) stay together. If you'd like, you can offline your threads in the guest for a fairly similar outcome. Some of us regret that we sell an n1-standard-1 and the shared core machine types, but it's pretty nice to have a $5/month VM price point for those that want it. Most folks doing serious numerical computing end up using much larger VMs, and always crossing NUMA nodes (which is a much larger impact).

Disclosure: I work on Google Cloud (and historically focused solely on Compute Engine).

I'm curious, what is there to regret about having a shared core machine type? Some people have a use for them and appreciate the cost savings.

Disabling hyper-threading only makes sense if you're getting messed up by the OS's scheduler.

FWIW, Azure disables hyperthreading on their servers. 1 core = 1 real core.

Reply on news.ycombinator.com