It depends on the technology used, some will have a larger overhead vs. others. A vps on Xen technology will normally score at least 95% of what the host scores (on the condition no other vps's are using up cpu/disk IO). UML ussually doesn't perform so well, in most benchmarks I've done it scored around 60-70% of what the host did. Of course this doesn't really reflect real world uses as this is scores when nothing else is using the cpu or other resources.
UML is dreadful CPU wise, you'll find in a hosting environment it does perform reasonably well, however when running a benchmark I'd expect a result of around 50%, obviously this is divided with the number of VE's you operate.
The SKAS patch does help performance slightly.
etechsupport2, what you're saying really has nothing to do with the subject of the post ...
█ Dan Kitchen | Technical Director | Razorblue
█ ddi: (+44) (0)1748 900 680 | e: [email protected]
█ UK Intensive Managed Hosting, Clusters and Colocation.
█ HP Servers, Cisco/Juniper Powered BGP Network (AS15692).