4 ms·
You cannot give each VM their own core. The business model of the cloud is that multiple VMs with virtual cores run on a single real core.
by founderling 7y ago
You cannot give each VM their own core. The business model of the cloud is that multiple VMs with virtual cores run on a single real core.
- derefr 7y agoAt the low end, sure. At the medium-to-high end, each VM is bound to one or more physical cores of the host, or sometimes an entire host ("dedicated instances.") I don't know enough about the IaaS market to know what the relative revenues of low-end compute vs. medium-to-high-end compute are for your average vendor, though. Is most of the profit in the low end? I'm also curious on what the impact on margins would be if IaaS vendors decided to switch away from serving the low-end compute demand with "a few expensive high-power Intel cores per board, each multitasking many vCPUs", to serving the demand with "tons of cheap low-power ARM cores per board (per die?) with each core bound to one vCPU."
- aflag 7y agoThe low end compute must be a substantial amount of revenue.
- LASR 7y agoVery little of the cloud compute business actually involves compute. Businesses move to the cloud for lower TCO compared to solutions they would have to hire entire teams of engineers to deploy and manage. Something like AWS, the CPU cores account for a small percentage of the revenues they make. The vast majority of their EC2 instance types have dedicated CPU cores. This may not necessarily mean that VMs are pinned to cores, but just that the overall compute power available is pretty tightly segmented. But regardless, as a technical decision maker, I’ll move my organization to the cloud because of scalability, fault-tolerance, reliability & security - things that would be impossible or prohibitively expensive to do on-prem. CPU cores are rounding errors in comparison.