The primary difference is that we provide phone support -- direct human access -- at our basic [support] level. If there is a problem, you don't have to look at a Twitter feed to see if the network down. Anywhere else, if you want to talk to a human being, you have to use a different pricing tier.
Also, there is a philosophical disagreement about how you design an application. In some cases, [cloud providers] will say the resiliency is the application's problem and if there is a problem, it is because the application wasn't designed to make due with our resiliency capabilities.
While we feel it is very important to allow customers to build their application in a resilient fashion, we believe that you shouldn't just leave everything up to the application because that puts too much burden on the developers.
So, in our world, we believe we have some responsibility for business continuity. That is the enterprise mindset. Enterprises expect that. As a vendor, you have some responsibility in ensuring that your customer can get the resiliency out of your system.
IDGNS: Rackspace recently launched a program to bring in partners to offer OpenStack deployments, which the company plans to turn into a federated network in which customers can easily move around their virtual workloads from provider to provider. What is HP doing to foster interoperability with its own OpenStack services?
Gillai: HP has always been a supporter of open solutions and interoperability. We have not used lock-in mechanisms to differentiate ourselves from the competition, and we don't plan to do that in the cloud. The differentiation that is going to come between my OpenStack and someone else's OpenStack will not be based on lock-in.
Specifically with Rackspace, we talk with them all the time. We consider them to be more of a partner than anything else. We're collaborating on a lot of things together. We're supporting the concept of ensuring of the ecosystem and maintaining interoperability.
IDGNS: What is the state of OpenStack? We've heard it still may be a bit rough around the edges. But it must be usable if HP is offering it in its enterprise cloud offerings. What work still needs to be done?
Gillai: I think one of the things that certainly needs a lot of work is the whole installation and upgrade experience, in terms of how automated it is. The promise of OpenStack is that it should be like your phone -- it upgrades automatically. We've added a lot of our own intellectual property to ensure that. While a large service provider can afford people [to install and upgrade OpenStack], an enterprise doesn't want to deal with that -- It just wants to pick up the latest releases and have it work.
The key in the next 24 months is to really make it to be plug and play as much as can be expected for the average enterprise to use. But I have no doubt it will get there -- there are a lot of people working on it.
I think there is also work coming out now around performance information -- getting more data about what is going on.
The first thing you need to understand about OpenStack is that OpenStack isn't the solution to everything. OpenStack is a kernel you build on top of and that is where companies like HP can differentiate by adding capabilities. Some of these things may be part of the OpenStack kernel and some may be a value-add that is provided by vendors like HP. That's OK, as long as it doesn't break the base OpenStack.