OpenStack is growing. Just looking around at the OpenStack Summit Paris, it’s exciting to see how far it’s come in just four years. It’s being used in production at many large enterprises and was recently named the most important open source project in the cloud. It was great to see Jim Zemlin from the Linux Foundation on the Summit keynote stage this morning. The size of the Summit and excitement around OpenStack are reminiscent of Linux conferences in the early 2000s – it’s a project that’s still developing, but on track to dominate.
But there’s a classic critique of open source projects – it’s that they have so many bells, whistles, knobs and switches that they are difficult to use. OpenStack is starting to fall into that trap.
As a community, we need to simplify OpenStack.
What do I mean by that?
First, we need to make it easier for users to adopt and use OpenStack. It’s something we’ve heard more frequently over the past several months. Whether they leverage the public cloud, a private cloud or various behind-the-scene services that run OpenStack, many users feel that OpenStack is getting too big and has gotten too complex.
OpenStack is growing to the point where it’s running the risk of losing users due to its complexity. To ensure OpenStack is simple to use, we need to simplify how we work as a community. We need to focus more on the OpenStack core than on the many tangents and splinters that grow off of it. We have to work together to make OpenStack simple, scalable and consistent.
The ultimate goal is manageability and scalability – those should be inherent.
Second, as a community, we need to make a concerted effort to look at OpenStack’s foundation, the core projects upon which everything else is built, and make sure they’re in proper order. That’s where we need to focus.
We should focus on easy integration, smooth upgrades, improved scalability and manageability and try to alleviate a lot of the cross-project dependencies that have arisen over the past year.
While it’s not our position to tell the community where it should and should not contribute, Rackspace plans to double down our efforts on some projects that we see as foundational to OpenStack’s overall simplicity and adoption, such as Cinder, Swift, Neutron, Nova, Glance and Keystone.
We are also working to simplify the user experience in our own products. We’ve simplified the use of OpenStack through our Rackspace Private Cloud and the support and management options we offer users to help them run it. This week’s launch of our production-ready Rackspace Private Cloud application stacks is another step toward simplifying OpenStack for users – it reduces the amount of time it takes to build and deploy these applications from weeks or months to just minutes. We are also constantly seeking new ways for users to get up and running on OpenStack quickly and easily in our public cloud and through the Managed Cloud services and support options we offer. Additionally, we are investing heavily in the OpenStack SDK project to provide end users with a single, consistent SDK.
For the past four years, OpenStack has had great momentum. It’s incredible how quickly it’s gotten to where it is today. But we want to echo what Dr. Stefan Lenz from BMW said in his Summit keynote this morning: to continue this success, we as a community have to get back to the basics and focus on the core platform what makes OpenStack valuable to businesses, universities and other organizations.
Our goal as a community should be to get OpenStack in the hands of the users. That’s what matters most. By simplifying OpenStack we’ll make it easier for people to build on the platform we provide.