* If the total number of VE's in your host is more than CPUs number, and there are many tasks '''threads''' running inside each VE, and they are scheduled quite often, it's is better to give just one CPU for a single VCPU to each VE. In this case the VirtualCPU-scheduler thread memory locality will significantly reduce overhead can be significantly decreased, on SMP memory coherence and overall performance can be increased up to ~<font color=red>50-100%</font>!