4 ms·
Cost, availability, and speed. Salaries more often than not, costs a hell of a lot more than compute and energy bills cost. If you have a 10k cluster, you're p
by latency-guy2 3y ago
Cost, availability, and speed.
Salaries more often than not, costs a hell of a lot more than compute and energy bills cost. If you have a 10k cluster, you're probably a mega-corp running a few teams sharing the resources running experiments in parallel to each other and themselves.
Now this is an very much so an overestimate on the actual cost of compute, the $10/hr/GPU cost is in all likelihood still significantly cheaper than the cash you pay to your researchers, and all the things that are needed to support them. Unless your team of researchers is about 10 people running very efficiently and making use of ALL the GPUs each hour of every day.