The reason we
were talking about queueing theory is that it’s part of how the hypervisor copes
with CPU oversubscription, by queueing the VMs.
VMs Queue for
free CPUs
·
Ready
Time
·
Co-Stop
time
·
Higher
utilization = higher contention
·
More
concerned about CPU busy than vCPU to logical CPU ratio
You can see
when this starts to happen by monitoring ready and Co-Stop metrics. You should typically be more worried about
CPU busy than you are the ratio of CPUs in the VMs to the logical CPUs
presented by the hardware.
Because all
this is Math, people have written programs to model this, so you can see how
busy you can run your hosts before performance becomes unacceptable.
Summary
If
there is anybody reading this who considered oversubscription to mean poor
performance then hopefully I’ve gone some way to showing you that’s not the case.
•
Oversubscription
does not equal unacceptable performance
•
Virtualisation
is expecting you to oversubscribe
– It’s the reason it exists (Don’t throw
that away. It costs you money!)
•
Take
the fear out of oversubscription through proper planning
– Plan for performance, not ratios
Look at the
metrics on your systems and use them to model the point where performance will
degrade because of utilization. You
cannot do that by looking at the ratio of vCPUs to logical CPUs but you can
with utilization figures. Thanks for following my blog series, if you have any questions please feel free to drop me a line.
If you'd like to learn more about managing VMware Capacity then why not try our Pay-to-View workshop
Phil Bell
Consultant
No comments:
Post a Comment