The first OpenStack Summit was held in Austin, Texas, in 2011 with 75 participants. In celebration of five great years, organizers brought the event back to its inaugural site April 25-29 of this year. The event featured more than 7,000 visitors from 50 countries, including users, DevOps teams, vendors, analysts, and me. We generally discussed how we can make OpenStack more agile and responsive to user needs and got a great look at the latest Mitaka release.
OpenStack allows organizations to scale quickly and get applications and services into the hands of users faster. Gartner gave a great keynote about the bi-modal future of the cloud. They pointed out that traditional IT infrastructures with enterprise-level apps and databases will continue to play a critical role for many businesses and can’t be completely eliminated while they build out their cloud architecture. Future workloads will demand that enterprises respond to pressures from both increased innovation and market disruption while delivering apps and services seamlessly to meet ever-increasing demands from end-users. Successful organizations will engage hybrid infrastructure technologies and leverage the scalability of cloud environments to deliver agile services and apps to their users, wherever they may be.
I continue to be amazed by the diversity of riders on the OpenStack bus and their endless ingenuity. There was a great case study presented by AT&T at the conference, showing how they leverage OpenStack to enable network function virtualizations (NFV) on their private cloud. The volumes of data they deal with are truly humbling: more than 114 petabytes daily from their mobile network alone! The growth in data has been unimaginable, an increase of over 150,000% in just nine years. While that is already dizzying, demand doesn’t seem likely to ease up any time soon. The company projects volumes will grow by ten-fold in the next four years. AT&T anticipates growth of approximately 50 billion new connected devices by 2020, requiring over 400 million new servers.
Scalability is an issue of urgent concern for many companies, but we covered much more at the conference. Other OpenStack Summit sessions touched on emerging technologies like containers, SDN/NFV, bare-metal, and the much-anticipated Internet of Things (IoT). By leveraging the varied expertise of its open-source community, OpenStack will support teams working in all of these areas and more.
The Missing Middle
While we heard lots of great things from enterprise-scale businesses like AT&T, Volkswagen, Comcast, and Walmart during the summit, there was a conspicuous absence of smaller and mid-sized companies. At enterprise scale, OpenStack offers significant TCO benefits, but this is in part because of the IT resources they are able to throw at standing-up a hybrid cloud. For smaller operations with fewer personnel and narrower ranges of expertise, OpenStack can be overwhelming. To be effective, OpenStack deployments require a shift in IT culture. Getting VMware system admins and DevOps teams united in planning a datacenter’s future is not easily done. As several speakers pointed out, 95% of all problems with private-cloud deployments result from a failure to evolve IT’s operational model. To be effective, OpenStack requires a fundamental change in both mindset and operations, and change is never an easy thing.
While enterprise-scale telecom NFV cases continue to lead in OpenStack deployments, we are also seeing upticks in adoption by other sectors including enterprise private cloud, public cloud service providers, academia, and government. Overall, 65% of all OpenStack deployments are now in production, a 33% increase over 2013 levels.
If you are ready to explore what OpenStack can do for your business and what infrastructure it will take to support it, take a look at some of our solutions, including simple cloud storage deployment and management with software-defined storage for HPE Helion environments.
You can also refer to our website for more information on HPE OpenStack.