Mesosphere’s Michael Hausenblas: Who Should Have the Last Word in Automation?

5 May 2016 9:08am, by

Michael Hausenblas
Michael Hausenblas is a developer and cloud advocate at Mesosphere, where he helps build and operate distributed applications. He shares his experience with distributed operating systems, large-scale data processing, and IoT applications, through demos, blog posts and public speaking engagements. He also heavily contributes to open source software, such as Apache Mesos, Apache Myriad, Kubernetes.

When an organization starts working with containers, developing a solid infrastructure can present a challenge. Without the proper tooling in place or an automated solution, having a host go down, or a sickly service falter can quickly decimate one’s system. Mesosphere offers management with its Datacenter Operating System (DC/OS), providing developers with a way to run their entire cluster in harmony.

In this episode of The New Stack Makers podcast embedded below, The New Stack founder Alex Williams spoke with Mesosphere’s Michael Hausenblas to hear more about Mesosphere’s take on orchestration, utilizing containers at scale, and meeting the needs of the open source community while also being realistic about the time involved with creating software for open source distribution. The interview was conducted for the new TNS eBook, “Automation & Orchestration with Docker & Containers,” out now.

“I think the main thing really is how can you realize this next step after Docker build, Docker run, is how to really put the containers into production and keep it running. For us, that’s the thing where Mesos is kind of unique and able to run all kinds of different frameworks,” Hausenblas said.

Listen to the podcast here:

The discussion can also be heard on YouTube.

Deciding what pieces of one’s workflow should be automated and when human decision-making is required has long been a topic of discussion in IT circles. Hausenblas noted even with an automated workflow and orchestration implemented; developers will still want to have established checkpoints where they can roll back to a previous version if a change isn’t working out as intended.

[cycloneslider id=”ebook-3-sponsors”]

“The question is just, with this manual checkpointing, will that go away when we fully automate everything or do we still at some point in rolling out a new version have a human giving the last [word]?” Hausenblas asked.

Hausenblas also discussed the finer points that come with bringing together different application vendors for a common goal of improving an open source project. In particular, the often-supported demand for enterprise-level 24/7 customer service without having to pay for anything which can be seen in some OSS communities. “Nothing comes for free there. Equally, on the other hand, you need to make sure that the open source world has access to new stuff. Otherwise, the adaptation is not as right.”

Feature image via Pixabay.

This post is part of a larger story we're telling about the state of the container ecosystem

Get the Full Story in the Ebook

Get the Full Story in the Ebook