It’s hard to believe OpenStack was founded just seven years ago.
I’ve been working with this open source platform since 2012, when I installed the Essex release from source code. Today, I sit on the OpenStack Foundation board and co-chair the Interop Working Group, which focuses on OpenStack interoperability guidelines.
While I won’t be attending the upcoming OpenStack Summit, to be held Nov. 6-8 in Sydney, Australia, it’s still the perfect time to assess OpenStack’s continuing evolution.
For the past five years, I have watched OpenStack improve exponentially, release to release, evolving from a virtualized compute and storage platform to a powerful infrastructure as a service layer that now enables other complimentary technologies, such as containers and NFV (network functions virtualization).
Long gone are the days of manual OpenStack deployment from source code; multiple deployment tools now make that simple. Automation tools such as Ansible and Puppet enable deployments of enterprise grade OpenStack solutions, consistently and with ease. In fact, OpenStack is now the standard for private cloud in the enterprise.
Simplifying the ecosystem
OpenStack technology has evolved, and so have the projects and project structure. From a handful of original projects, today there are more than 60. This growth was enabled by a switch from “integrated release” to a “big tent” model, which facilitated greater modularity and flexibility in deployment models.
With big tent, however, came additional complexity, confusing for a newcomer looking to deploy OpenStack, trying to figure out which of the 60+ projects were necessary. It also allowed anybody to spin up “something-as-a-service,” with new projects aiming to solve every potential issue in an OpenStack land grab. Those projects that failed to monetize lost developers and corporate support, stagnated and were retired by OpenStack Technical Committee. Other projects retired for a lack of maintainers. I think this process is a natural evolution for open source projects, and we will see more projects retired or put in maintenance mode.
This reduction in OpenStack projects, however, is driving towards simplification of the ecosystem.
The future of OpenStack
Going forward, we will see initiatives to simplify the project landscape with clear paths for deployments. We’ll continue to refine the answer to the question, “What is OpenStack, and what are its components?” Project Navigator has done a good job at starting this process. And while there are still a multitude of projects that can make OpenStack confusing to newcomers, its composable architecture drive the growth of OpenStack as a flexible and complex private cloud solution.
I would be remiss talking about OpenStack without talking about community and technical self-governance. While the OpenStack Foundation governs the legal and trademark aspects, the technical aspects are community driven. Community also has the power to influence the OpenStack Foundation’s work through its governing bodies: the board of directors, technical and user committees. Additionally, multiple working groups are focused on solving different issues.
The foundation, with the help of the community, has also been changing and evolving, just as OpenStack technology has changed and evolved. Summits are a great example of this evolution, and they reflect the changes in community well. The first summits were geared towards developers, with the design summit being the integral aspect of the OpenStack Summit.
The evolution of the OpenStack Summit
When I went to my first OpenStack Summit in San Diego in 2012, project Quantum was gaining momentum to replace nova-network, release cycles were tied to the summits and design sessions kept developers busy all day long. On the days that design sessions overlapped with general sessions, developers had to choose which sessions to attend.
We have come a long way since then: today, releases are no longer tied to summits, meaning developers are able to attend speaking sessions, and leave the design work for Project Teams Gatherings (strictly technical gatherings aimed at individual project work as well as cross-project collaboration). Nova-networking has been relegated to legacy status, and Quantum is now known as Neutron.
Summits now offer more training and hands-on sessions for all skill levels, sponsored by the OpenStack Academy, plus a forum where users, operators and developers can brainstorm requirements for the next release and gather feedback from previous ones.
And while this summit is expected to be on the small side due to its location, I expect it to be very interesting in terms of adjacent technologies and new use cases. Telcos have long loved combining OpenStack and NFV to deploy communications and cloud services, so I expect interesting use cases there. And of course, we’ll be hearing a great deal about one new technology in particular, containers. (And did you know that OpenStack is so popular with universities and researchers it now has its own track at the summit?)
OpenStack Sydney Summit sessions may be dominated by container talks, but I predict future summits will focus less on containers and more on edge computing, a new technology area people are starting to explore more. OpenStack is positioned at the forefront of edge computing thanks to its scalable and composable architecture. Learn more about Openstack and edge computing.
I truly believe OpenStack is one of the most successful and far reaching open source projects in the world, with even more successful community to back it. It’s been incredibly rewarding to be part of its success. I have not only watched OpenStack mature as technology that’s now used by countless companies and research organizations, I also have witnessed the OpenStack community grow, mature, and expand to every corner of the world.