OpenStack Part I: OpenStack Adoption Matures, Knitting Hybrid Clouds Together
As the OpenStack Foundation’s twice-annual conference returned to its roots in Austin, Texas, the focus of the OpenStack 2016 conference was integration into mainstream IT. The rapid increase in hybrid clouds is creating demand for OpenStack-based software that can be supported by a broader group of enterprise platform vendors – most of which are openly supporting OpenStack APIs for interoperability between clouds.
The presence of 7,500 attendees at the Austin conference underscores the broad interest in adopting OpenStack for multi-cloud deployments. Much of the discussion centered on the challenges of hybrid computing – with workloads spanning enterprise data centers and cloud service providers’ data centers.
However, OpenStack must become a more consumable technology for enterprise IT, requiring fewer OpenStack specialists for implementation. This will open up OpenStack development to a wider base of customers in mid-size companies. Today, OpenStack-skilled developers and system integrators are often needed to accelerate adoption within large deployments.
Expanding the Total Number of OpenStack Projects
As described at the Austin conference, most customer deployments are based on the major Open Stack building-blocks, such as Cinder (storage); Nova (compute), Swift (object service); Neutron (networking) and Glance (image service). These projects have been around longest – and are being deployed in production today.
While the “short list” of Open Stack projects is seen as foundational to cloud-native and cloud-enabled deployments, a large group of smaller, new projects are being developed under the moniker “Big Tent.” Discussions here showed that the fate of the Big Tent grouping is somewhat unsettled, or uncertain. Not yet decided: how to bring more of the Big Tent projects to completion without confusing the emerging marketplace for OpenStack-enabled software tools and applications.
OpenStack has great utility, especially in financial services, telecommunications and retail – all of which value real-time analytics and cloud computing to extend services to end-customers. Building with OpenStack tools speeds development of cloud-native workloads – and ties together many components of scale-out analytics, through workload management, monitoring and discovery of available resources.
Key trends affecting OpenStack and the OpenStack community developers include:
- OpenStack is being deeply adopted, especially in large enterprises with 10,000 or more employees. Main driver: better interoperability between clouds, and improved hybrid clouds.
- The OpenStack roadmap is being expanded to support scalability, availability manageability and security – all of which are key enterprise touch-points. Main driver: business managers are being called upon to fund new and emerging OpenStack technical projects.
- Wider acceptance of the key foundation building-blocks for OpenStack – including Nova (compute), Cinder(storage), and Neutron (networking) – are gaining acceptance as cornerstones for building out cloud infrastructure. Main driver: the technology works, and it is being knit into a broad array of vendor products and services.
- Customers are seeking more consistency across multiple APIs and OpenStack projects. Main driver: a proliferation of OpenStack projects has led to some degree of overlap in project scoping (as in the Big Tent grouping of projects), as customers seek easier ways to consume OpenStack functionality.
Why is This Happening Now?
The emergence of so many new OpenStack projects – and the emphasis on enterprise-readiness—is the inevitable result of the maturity of the Open Stack efforts, six years after the first OpenStack conference was held in Austin in 2010.
- OpenStack has become a priority for many companies, including Walmart, AT&T, Comcast, Wells Fargo Bank, and NTT, made the case for strong OpenStack adoption in enterprise cloud.
- The ecosystem growing up around OpenStack has expanded, as most major vendors of hardware and software want to be OpenStack-compatible, to connect their products with a broad, cloud-enabled technology layer.
- Working with widely installed enterprise software (e.g., VMware, SAP and Oracle), is high on the priority list of companies that have energetically built and deployed OpenStack software for their cloud-enabled applications and databases. That reality reflects OpenStack’s journey into the center of IT datacenter infrastructure.
Bottom Line for IT
We’ll see the two worlds of enterprise IT co-exist – enterprise applications and cloud workloads – for entirely practical reasons. Users cannot afford to rip-and-replace data center infrastructure. Rather, they will increasingly weave OpenStack into existing IT deployments – and they will link it to green-fields OpenStack projects addressing a new generation of scale-out, cloud-native workloads for business.