Ben is a founding member of Animal Labs and our resident DevOps Jedi Master.
We sat down to learn how his experience shaped his career path and
crafted the unique skillset he brings to the team.
Q: Over the course of your 10+ year career, what has been your favorite DevOps project and why?
A: The most complex task I ever took on required increasing my client’s processing time by 8,400% I know that sounds absurd but cloud technology was still very new at the time and what we were aiming to accomplish was revolutionary for the client.
The project was for a leading GPS corporation. They needed to speed their map processing time from 1+ month to just a few hours. Part of this involved an enterprise scale cloud migration. In this case we were migrating to AWS and a big part of the project involved up-skilling their internal team.
Overall, the project took 2 years and involved networking over 2,000 computers and multiple data centers around the world so that they could move up to 800 terabytes of map data at a time.
To put that in context: whereas before it took them 1+ week to process the data for the city of Berlin, now it only took 2 hours. It was an incredible result and I’m proud to have been part of it. This experience is a big part of why I decided to start my own DevOps consulting firm. I enjoyed being a change leader and working with cutting edge solutions to help my client achieve groundbreaking results.
Q: What’s involved in such a transformative project? What difficulties did you have to solve for?
A: Well, I had to turn a non-distributed system into a distributed one. They started out with an onsite database that stored global road network data that was on an annual push. They needed me to transform that into a cloud-based data center that was capable of continuous delivery. In order to do that, I needed to accommodate for distance between computers, manage database migration and relocation, manage the computer processing connections, train their internal team on these changes… these projects when done properly are usually a huge undertaking.
Imagine: how do you prevent thousands of computers from downing the network entirely? You have to build in failure resiliency and network durability, plan solutions for data sovereignty issues, ensure everything is modularly upgradable, solve throughput flow problems, build out database divisions… it’s like pulling all the pieces of a super computer apart and stretching the, around the world and then finding a way to glue them together so that all the data continues to flow properly.
Q: What is your recommended infrastructure platform? Do you always work with Amazon Web Services (AWS)?
A: AWS is great and I work with it a lot but it’s not the right choice for every company out there. Google Cloud Platform (GCP) and Microsoft Azure are obviously top contenders in the space but there are many other solutions out there as well and it really depends entirely on the company, their parameters and goals.
When my clients need help migrating to the cloud the most important part of my job is usually simply helping them pick the right vendor. Many consultants pick the ones they have partnerships with and know best – that’s not how I work. I work with my client to identify their unique business requirements and advise what’s best for them – not me. I am not a reseller and I make it a point to know the entire landscape well.
Q: What’s your ‘dream project’?
A: I’ve really been wanting to get involved in another cloud-based, greenfield project. I want to be able to leverage all the latest and greatest solutions out there. So, for example, a project involving enablement of IoT is likely to require this.
Q: And when you’re not pulling super computers apart? What are some of the ways you like unwind?
A: I’ve really gotten into welding lately! I guess my form of relaxation still involves building structures. One of my favorite bands is the Miles Davis Ensemble. I like to just listen to their music and start creating.