On Thu, 24 Jul 2008 16:36:21 -0600, Stefan de Konink <[EMAIL PROTECTED]>  
wrote:


> But there is one thing in VMs that are actually very cool. That is the
> amount of cpu cycles used. So instead of of 10 seconds of requests, you
> could actually do 50000 request, and check the CPU usage for each
> implementation.

You also gain ease of repeatability and therefore provability.  If you can  
provide a VM image than anyone can access (take a look at  
http://www.rpath.com/rbuilder/ as a simple way to build out and make  
available an image for a large list of VM clients) it's a lot harder for  
them to contend a particular benchmark than it would be otherwise.  It's  
also a lot easier to allow people to "see for yourself."

> It is not about the maximum performance on a Dell1950, because a kernel
> can make a difference. It is how much performance is required to deliver
> peak performance.

And from a business perspective, "How many servers of type X are needed to  
handle the load of site Y?" is an invaluable tool.  That's easy an easy  
equation to evaluate when you have consistent numbers to build from.

-- 
/M:D

M. David Peterson
Co-Founder & Chief Architect, 3rd&Urban, LLC
Email: [EMAIL PROTECTED] | [EMAIL PROTECTED]
Mobile: (206) 999-0588
http://3rdandUrban.com | http://amp.fm |  
http://www.oreillynet.com/pub/au/2354 |  
http://news.oreilly.com/m-david-peterson/
_______________________________________________
Cherokee mailing list
[email protected]
http://lists.octality.com/listinfo/cherokee

Reply via email to