oVirt, in my opinion, is the biggest contender with VMware vSphere. oVirt has the weight and development resources of Red Hat behind it, which has undoubtedly slingshotted it ahead of the rest of the open source virtualization solutions out there. It has almost all of the “out of the box” features vSphere has, and it works extremely well.
There have been two major holdback concerning oVirt in the past. First, early on it only supported Fedora. This definitely scared many people away, myself included. That is no longer the case as it now supports Fedora, RHEL, and CentOS. The second major drawback is the complexity of installation. Overall the methodology is pretty simple. At a minimum, you need two machines. An oVirt Engine, which is the brains of the operation and powers the web interface, and you have the oVirt Node, which is the “hypervisor.” Although the overall methodology is simple enough, it can really be a pain to install and get working. But, that’s improving as well.
I wrote this guide to help you get your oVirt infrastructure built on CentOS 6.6 easily, and quickly. You will need two servers, at minimum. The good news is that one of them, the oVirt Engine, can be virtualized, running on your currently configured hypervisor of choice. As far as specs, you’ll want to try to be close to the following.
oVirt Engine Minimum Server Specs (can be virtual or physical machine)
- CentOS 6.6 x64
- 4 CPU Cores
- 4GB Memory
- 25 GB Hard Disk Space
oVirt Hypervisor Node (must be on physical machine)
- oVirt Node 3.4 ISO installed as OS
- 4 CPU Cores
- 6GB Memory
- 40GB Hard Disk (or iSCSI / NFS storage available)
This is enough to work with and get a good idea of what the oVirt platform is capable of. It’s also a solid foundation that can be grown and expanded on to form a production worthy infrastructure. So, lets get started.