Product, Company

Why the Open Container Project matters, and why we joined it

For more than five years, DC/OS has enabled some of the largest, most sophisticated enterprises in the world to achieve unparalleled levels of efficiency, reliability, and scalability from their IT infrastructure. But now it is time to pass the torch to a new generation of technology: the D2iQ Kubernetes Platform (DKP). Why? Kubernetes has now achieved a level of capability that only DC/OS could formerly provide and is now evolving and improving far faster (as is true of its supporting ecosystem). That’s why we have chosen to sunset DC/OS, with an end-of-life date of October 31, 2021. With DKP, our customers get the same benefits provided by DC/OS and more, as well as access to the most impressive pace of innovation the technology world has ever seen. This was not an easy decision to make, but we are dedicated to enabling our customers to accelerate their digital transformations, so they can increase the velocity and responsiveness of their organizations to an ever-more challenging future. And the best way to do that right now is with DKP.

Jun 22, 2015

D2iQ

D2iQ

On Monday, Docker announced the Open Container Project (OCP), a non-profit organization that will define common standards for software containers. Mesosphere is proud to be a founding member of the OCP, joining other founding members including Google, Amazon, Microsoft, Red Hat, and Intel in this important effort.

 

Containers are a fundamental software technology of our time. They are part of everything we do here at Mesosphere and we believe they will help define the next decade of computing. The OCP is an important step toward ensuring that application containers remain on their current path of revolutionizing the way applications are developed and deployed. Consolidation around the OCP container standard (the GitHub repository is here) will ensure interoperability, benefiting customers with competing implementations (of the same standard) that don't create vendor lock-in.

 

Fueling the container revolution

 

At Mesosphere, we're all about containers and we've been supporting them for our customers since the company was founded. They're a major reason why our Datacenter Operating System (DCOS) is able to schedule workloads across thousands of servers so flexibly, quickly and efficiently.

 

Our Apache Mesos-based Datacenter Operating System has always been container-centric and we made containerization a pluggable module so that we can support every major container format as it emerges and is embraced by the community. We currently support Docker containers and we also launch non-Docker tasks inside resource-isolated Linux cgroups. We worked with CoreOS on Rocket (now rkt) and its App Container specification, partnered with VMware on its Project Photon, and we are intimately familiar with Microsoft's Windows container technologies.

 

As interest has grown in container-based PaaS and orchestration platforms, we've been leaders in that area as well. We built and made available as open source our own Marathon service, but we also support third-party projects such as the Google-led Kubernetes and Docker's Swarm. In fact, Kubernetes is already available today as an installable service for users of our DCOS who want to experiment with Kubernetes alongside their big data jobs, legacy web applications and other services on the same cluster. We will also be demoing Docker Swarm on the DCOS at DockerCon.

 

Standards accelerate

 

However, while containers have become a foundational component for building simple, flexible, efficient and portable applications, the container itself is not a particularly new or distinctive piece of technology. Linux containers, which are the core of today's popular container formats, have been part of the Linux operating system since 2006. Mesos, the distributed systems kernel that undergirds the DCOS, has used Linux containers since 2009.

 

While Docker is the most well-known container format today, it's not the only one. At last count there were a half-dozen credible alternatives to Docker emerging, some from fear of one company exerting too much control and others for technical reasons. Whatever the reasons, they have all been creating fractures in the landscape while doing little to advance the state of the art or bring fundamentally new capabilities to users.

 

Normally, competition is good, but innovation slows when competition creates balkanization around a core technology. This is why open standards are so important, and it's precisely why the Open Container Project is such an essential piece of industry cooperation. By collaborating on the precise definition of a container, we can allow for multiple competing implementations of the runtime that interprets it while maintaining broad compatibility across vendors and implementations.

 

Orchestrating value

 

The container format is the right place to standardize because, it turns out that the real value of containers actually is not found primarily in the container itself. The real value is in how they are combined to orchestrate the work of many developers and run at scale, in highly-automated ways, across hundreds or thousands of servers.

 

Think of a single container as a 2x4 board or a sheet of plywood. They can be useful on their own, but they achieve their ultimate potential when lots of them are combined with a blueprint, shingles, nails and so on to create a house.

 

At Mesosphere, we focus on providing the entire package -- from infrastructure to orchestration. This is why we've built our DCOS on Mesos and its why we've made technologies such as Marathon, Kubernetes and Docker Swarm available to our customers. We want to make it as easy as possible to use software that schedules containerized workloads and spreads them across host servers, and that turns dozens, hundreds, or thousands of containers into a distributed computing system or microservice-based application.

 

We are proud to be founding members of the Open Container Project. An open, standard and stable container format will let customers spend less time worrying about what type of building blocks they're using and instead focus their energies on building the best house. It will let the companies competing to build those houses invest in the right capabilities, too.

Ready to get started?