Kubernetes HPC

We had two records, basically, like one of the end of summers or later August in one late September, where we first reached 800,000 VCPU concurrently on about 50,000 instances. And then we reached 1.2 million VCPU like still we've with 54,000 instances, just different instant shape that mean more VCPU. Are those normal numbers for HPC workloads? Yes. So it's in the high scale, but it's still like what you will expect from an HPC environment to give it some perspective is like this 21 to 30 petaflops and 75 petabytes of storage. But they haven't included the use case of cloud and Kubernet