Last week, I was in San Diego for the oVirt Workshop and LinuxCon, where I presented on “Integrating oVirt with Enterprise Storage”. I’d like to go over some of the major points that I covered in my discussion, but I’d also like to explain why NetApp is involved in oVirt in the first place. In order to keep the posts brief, I’ve broken things up and I’m starting with how NetApp got involved with oVirt in the first place. Here goes…
Around the same time that I was planning and launching my Captain KVM blog last year, Red Hat was almost done with porting Red Hat Enterprise Virtualization from its Windows/.Net roots to a fully open source project. Red Hat had long recognized that KVM was bigger than just RHEL or Fedora, and that RHEV needed to be bigger than just Red Hat. Major kudos to Red Hat for this.
Ultimately, this meant that RHEV would need an upstream project, so Red Hat and IBM put oVirt together (re-launched it, really) to give it to the open source community. But, it also needed doers and believers from the industry to push things along. Enter Cisco, Intel, SuSE, and NetApp. We were all approached to join oVirt at the board level and invited to meet with Red Hat and IBM.
Our first meeting as a group occurred during KVM Forum/LinuxCon in Vancouver, BC, Canada in 2011. That first oVirt meeting was really cool. All of the initial board level companies were all sitting together in the same room, all with the same level of excitement. Egos and politics were checked at the door.
NetApp was definitely interested in joining the community, which may come as a surprise to some folks. NetApp is hardly the first company that jumps to mind when someone brings up the topic of Open Source. However, in my specific organization at NetApp, one of our prime directives is to do more with Open Source, integrate with Open Source, create solutions around Open Source, and contribute to Open Source.
Granted, much of the Data ONTAP (NetApp operating system) code is proprietary, but NetApp has long been involved with Open Source. I think the single biggest ongoing contribution is the Linux NFS Client stack. Other Open Source tools and projects include pNFS, NFSometer, and the Snap Creator backup framework that I’ve covered previously. (BTW, I’m likely to have more posts about Snap Creator.. hint, hint.)
Anyway, back to oVirt (my ADD even creeps into my writing…). Red Hat approached NetApp about joining oVirt as the one partner that is solely focused on enterprise storage. I was then approached by the NetApp CTO to be NetApp’s representation on the board and I jumped at the opportunity. At that point, I had been pushing RHEL, KVM, and RHEV internally at NetApp to anyone that would listen for about 2 years. The partnership between NetApp & Red Hat was taking shape, and I had completed a series of projects around Red Hat & NetApp technologies. More and more customers started asking us about KVM; it was the perfect time and opportunity.
It really made sense for NetApp to be involved with oVirt from my standpoint. NetApp has a long track record for solid partnerships with virtualization companies and I really wanted RHEV to be part of that. I really want to see RHEV and KVM succeed in the data center and that means working upstream. I honestly believe that a strong partnership with a storage provider is a key to that success. But the partnership alone won’t bring success to oVirt, RHEV, or our joint customers.
The other big key is integration, and I’ll cover that in the next post..