We have a set Xen domUs for production and we have been providing different QOS for different clients and needs. We are able to limit the number of VCPUs and the network traffic.

Now, we would like to be able to have something like the AWS Micro-Instance, with CPU bursts of a limited time with a fixed CPU top limit.

So, my question here is, How could we achieve that? What is the option or options from the CreditSheduler? Does it even got anything to do with the CreditSheduler?

Thanks a lot!!


The answer is more of a physical to virutal hardware mapping. What you need to be able to do is to create a virtual machine environment that maps to the equivalent physical resources that you are planning to sell.

If you are looking for software solutions for this take a look at things like Eucalyptus, CloudStack, OpenNebula, OpenStack, etc.

What you really want to make sure you handle well is the monitoring aspect, which it sounds like you have some handle on in terms of your current QoS monitoring already, but you may want to look into monitoring tools such as Cacti/RRDTool, Graphite, Nagios, Zabbix, and Zenoss.

This blog talks about benchmarking and the physical to virtual mapping that they did: http://www.zmanda.com/blogs/?p=665

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.