Misadventures in #HPC #Cloud Migration #9

But Which VM?

Misadventures in #HPC #Cloud Migration #9

As a quick recap, in #8 of this series we discussed a Bring Your Own Scheduler approach to cloud HPC migration that retains your existing on prem scheduler. In addition though, we added a component to flex the cloud compute capacity as required in order to optimise cost. Sounds great.

One question though. Which VM type are you going to start up? 

It’s a deceptively simple question but one with a surprisingly complex answer. Or rather there isn’t an answer as such, more a mechanism by which the answer can be determined. The answer itself may be different every time the question is posed.

I’m afraid though dear reader, I am going to leave you hanging for just a little longer. for our answer to what the unnamed component from part 8 looks like in an HMx Labs solution. 

I’m sat writing this at the Folkestone terminal of Le Shuttle with another 8 hours drive ahead of me on the other side. By the time this goes live on LinkedIn I should hopefully have arrived but for now I need to rush back to the car and get it onto the train.

In the meantime, in true Blue Peter style, here’s one I prepared earlier:

High Performance Computing on the Cloud
Running HPC clusters is much easier on the cloud where there’s unlimited capacity and no scheduling problems right? That’s certainly what all the cloud vendors always tell me, but how true is it?
HPC on Cloud — Cost Optimisation & Cloud Economics
A look at the economics of running a HPC system on a public cloud infrastructure. It looks at the impact of the performance differential between different types of VM, the impact of price variation and introduces the concept of a performance normalised price to allow easier comparison of VMs.