I'm sorry you find the concept confusing. The explanation may not be as clear as you'd like, but it's just an artifact of how the virtualization software that they use works. From a customer perspective, I'm very pleased with this arrangement over walling off a certain number of cores because it means that while my VM still has a dedicated amount of memory at all times, it can consume even *more* CPU power than just two cores during the times it's processing something and needs the power and there's no penalty for doing so.
I don't believe it would be any simpler to gauge 'load' based off of dedicated cores than it is off dedicated priority, but that's just my view. If a node gets CPU bound, they're going to address it pretty quickly anyway.
At any rate, I hope you find what you're looking for. I'm pretty pleased with KH though.
I don't believe it would be any simpler to gauge 'load' based off of dedicated cores than it is off dedicated priority, but that's just my view. If a node gets CPU bound, they're going to address it pretty quickly anyway.
At any rate, I hope you find what you're looking for. I'm pretty pleased with KH though.