At this year’s OpenStack Summit in Austin, Texas, the message was clear. OpenStack needs to pivot from a science experiment to a production system. Even though this is happening, it has been happening slowly. Some would argue that it has been achieved for the likes of PayPal and other extremely large institutions, such as AT&T. However, running, configuring, and installing OpenStack still takes more knowledge than the average enterprise system administrator has available to them. The new Certified OpenStack Administrator certification is a way to exhibit a level of competence for the age of the new OpenStack: the production-ready OpenStack.
However, that may not be enough. The OpenStack Foundation has to focus on ease of installation, ease of use, and what is actually necessary and sufficient to meet specific use cases. These use cases are growing from typical IaaS implementations to what lies on top of IaaS: the applications. This is where OpenStack integration with Kubernetes, Apache Mesos, OpenShift, Docker Swarm, Cloudify, and the like really pays off, as these tools are all about the application—any application. These tools bring a whole new level to OpenStack and are very disruptive technologies. Infrastructure as a Service is old hat these days, and OpenStack as well as other stacks do this extremely well.
So much so, in fact, that there is a stream of new announcements that make placing OpenStack into your environment much easier. The show floor was full of such items. Here are the big ones that I could see, but there are also smaller ones that deal with consulting services around OpenStack.
- VMware has improved VMware Integrated OpenStack with VIO 2.5, which back ports OpenStack Liberty with VIO’s Kilo implementation. My thought is that by VMworld, there will be a Mitaka release of VIO. In keeping with the collaborative nature of the OpenStack community, the VMware integrations are available through GitHub.
- Rackspace has announced OpenStack Anywhere. This is a unique approach for OpenStack delivery. Rackspace does not deploy OpenStack to just its data center, but to yours if that is desired. It deploys a Rackspace-built OpenStack within its Rackspace rack and ties that rack directly into Rackspace’s network and security operations centers. In effect, it fully manages the rack and OpenStack for the organization. It does this by providing the hardware and deploying raw OpenStack using a serious of Ansible recipes. The recipes and scripts are included within OpenStack’s upstream repositories, so any organization can easily deploy OpenStack.
- EMC had no announcements, but showed a covered rack. The consensus among attendees was that this was a VxRack deployment of OpenStack.
OpenStack is growing up fast; there are new technologies and companies joining the ecosystem. Collaboration is key to OpenStack. Collaboration is leading to easy deployments that include multi-OpenStack cloud instances, networking, and storage. Most importantly, knowledge is being shared with new products to ease OpenStack deployments.
The discussion on the show floor, in hallways, and in keynotes does not just cover infrastructure. Infrastructure seems more and more of a sideshow. Applications are central to OpenStack, and the second day’s keynote covered leveraging OpenStack to deploy applications rapidly and efficiently. This includes object storage for very large datasets, but also Kubernetes, OpenShift, Apache Mesos, Docker Swarm, and other application orchestration tools.
Infrastructure made easy using infrastructure as code mechanisms (such as Rackspace’s and VMware’s contributions) is an important but small part of the application deployment process, including spinning up whole cloud-based applications of thousands of containers within OpenStack. The application is King, Queen, and Jack. Applications are also the Joker, as modern applications seem hard to reach, requiring new knowledge and techniques.
This is the new OpenStack: a firm base for deploying next-generation applications using modern techniques for use in production today, not the future. While new skills are needed, more importantly, we need to break the siloed mindset. That has started within the OpenStack community as well as within major vendors and technologies.
We are heading toward a more collaborative environment. We need that as we grow our application footprints. Collaboration is the only way we will break down the barriers, close the gaps, and reduce security breaches as well as grow applications to the massive scale we’ll need in the future. OpenStack is leading the pack in terms of collaborative scale of applications.
Share this Article:
Latest posts by Edward Haletky (see all)
- Finding your Sensitive Data to Protect - March 27, 2017
- Scale and Engineering - March 23, 2017
- SDS and Docker: The Beginnings of a Beautiful Friendship - March 21, 2017