Ubuntu Advantage for Infrastructure: an Interview with Canonical

canonical logo

On April 29, 2019, Canonical made headlines by officially announcing the availability of Ubuntu Advantage for Infrastructure If you are unfamiliar with Canonical and the work that they do:

Canonical is the publisher of Ubuntu, the OS for most public cloud workloads as well as the emerging categories of smart gateways, self-driving cars and advanced robots. Canonical provides enterprise security, support and services to commercial users of Ubuntu.

Ubuntu Advantage for Infrastructure changes the entire landscape of service offerings for open-source software. Instead of itemizing and charging for each and every component or add-on, Canonical promises its customers a per-node service package, regardless of the technologies running on it. I was able to sit down and chat with Stephan Fabel, who was generous enough to provide a bit more detail around this exciting announcement.

Petros Koutoupis: Tell us about yourself.

Stephan Fabel: My name is Stephan Fabel, and I am Director of Product over at Canonical. So, I am running a team as the Product Manager, and I am responsible for the portfolio of products that go out to our customers.

Petros: For our readers who are unfamiliar, what is Ubuntu Advantage?

Stephan: As you might know, Ubuntu always has been freely available as an open-source Linux distribution for everybody to consume. And, for those users who wish to enter that commercial relationship with Canonical, either because they are interested in our additional bit-streams that we offer like kernel patches, extended security maintenance, FIPS compliance crypto libraries, or because they would like to get support for each of those open infrastructure components that we are covering, Ubuntu Advantage is the program that they would subscribe to.

Petros: What makes this recent announcement of Ubuntu Advantage for Infrastructure so exciting?

Stephan: Ubuntu Advantage is exciting because for the first time, it consolidates all of the popular open-source infrastructure frameworks, components and packages into one single per-node program. The idea is that as enterprises move forward and transition into the new reality of cloud native workloads with increased containerization, they roll out private or multi-cloud solutions to host these services. There is a necessity to introduce the portability of workloads across those multiple substrates. It is exciting to build for those use cases with a single program that allows them to change their minds or roll out another infrastructure under the same contractual terms they originally entered with. In other words, they don't need to go through another procurement process to buy Kubernetes or look at the legal terms again and renegotiate a services agreement if they intend to roll out OpenStack. They can simply consume Ceph and Swift and any other pieces of infrastructure included in UA for Infrastructure coverage, knowing and counting on the fact that Canonical stands behinds those packages as a trusted support partner and supplier of security-hardened packages in those technologies.

Petros: And that leads me to my next question: what is the extent of technologies covered?

Stephan: I think I mentioned the high-level ones that people tend to focus on. But there are other components in there as well. Docker support falls underneath that. We provide a Logging, Monitoring and Alerting (LMA) stack based on Prometheus and Grafana alongside the other components that are a part of this. There are just too many to list, right? Our standard support offering covers more than 6,000 packages, and with UA for infrastructure, it expands that quite a bit. There isn't a full list per se, but it's designed to cover the frameworks and technologies that go with it. For instance, in Kubernetes, you would get Jaeger support, Containerd support and the rest of the technologies that are part of a common roll out of Kubernetes. In OpenStack, you will be able to count on Swift, Ceph and Ironic support under the UA for infrastructure.

Petros: How does this compete with the other enterprise offerings?

Stephan: As we thought about the challenges our customers were facing, we noticed that there is an incredible amount of friction associated with the consumption of open source in the enterprise, and that friction came from a multitude of sources. Can I operate this? Do I have the skillset to run OpenStack or Kubernetes? We found a number of service offerings to address those challenges for our customers as they were trying to roll these technologies out. But there are also other sources of friction—that is, contractional friction and commercial or economical friction. The type of friction that forces roadblocks into the decision-making process of operators in enterprises and telcos alike that just simply increase all the frustration associated with that. Our responsibility as an open-source partner to many of the financial services, customers that we have in the vertical space (the telcos and service providers) and ultimately, in the enterprise, is to reduce that amount of frustration and to accelerate the adoption of open source. If we followed the traditional model and said that we can sell you one product to support the operating system and another to sell you support for Kubernetes and yet another that would cover OpenStack, that would produce unnecessary friction in the decision-making process and force unnecessary choices on the decision-maker. So we solved that by putting it all together, and instead of modeling our competitors, who charge extra for every add-on at every layer of the stack seeking to increase the amount of money they charge per core, we decided to disrupt that by saying "look, this is open-source infrastructure; your value differentiation comes at the application and not the infrastructure layer." Infrastructure should be easy to consume and easy to forecast and predict in terms of the economical impact it has on the operator. As a consequence, we believe that we designed UA for Infrastructure in such a way that it is in stark contrast to any other offering on the market today.

Petros: Is there anything else you'd like to share with our readers?

Stephan: The one thing that I would like to add to what I've already said is that we have designed UA for Infrastructure in such a way that it will continue to grow and cover more technologies as that field of open infrastructure evolves. It will not stay fixed, and it basically represents our commitment to our users and customers—that is, to always provide the best value and cost structure for that infrastructure that we can possibly extend. And so we are super proud of this consolidation that we have been able to do and for the volume of adoption we have seen in the market. We will also continue to drive the cost down.

Petros Koutoupis, LJ Editor at Large, is currently a senior performance software engineer at Cray for its Lustre High Performance File System division. He is also the creator and maintainer of the RapidDisk Project. Petros has worked in the data storage industry for well over a decade and has helped pioneer the many technologies unleashed in the wild today.

Load Disqus comments