At TriNimbus we love DevOps so much, we even put it on a t-shirt! Fun fact: if you look up DevOps in Wikipedia, you can see what we mean (yes, that’s our Chief Cloud Officer wearing our special edition t-shirt!). TriNimbus was a Gold sponsor at DevOps Days 2017 and we wanted to share our experiences at the event through this post. While we spent most of our time at our booth, meeting attendees interested in understanding what TriNimbus is about, we also managed to sneak in to catch a session here and there.
After the initial onslaught of traffic at our booth from the opening hour, we had time to attend a session. While we missed Bridget Kromhout’s talk ‘Computers are easy; people are hard’, we made it to Augusto Rosa’s ‘Shomi Stories: A Cloud Migration, Team Building and Having Fun in between.
It was an inspirational talk about the virtue of empowering your DevOps team to do great things. (Full disclosure: our colleague Nadji worked with Augusto on this project, so we are probably biased on this one). Augusto walked the audience through the transformation that happened at Shomi with its DevOps team as their platform launched, matured and, unfortunately, eventually shut down.
Some significant points:
- Choose tools that enable collaboration (Git, Terraform, etc…)
- Use peer-review as a means to ensure visibility within a team but also across teams
- Use the methods and processes that best suit your organization (e.g. “Scrumban” a term he coined to explain how DevOps used a combination of Scrum and Kanban)
Augusto also had a strong message to the management of any organization with a DevOps team: You need to empower your DevOps team(s) and give them the tools and mandate to innovate.
All while poking fun at himself, he delivered a passionate and heartfelt account of his experience managing (among many teams) the DevOps team at Shomi.
We were also at Mia Henderson’s talk, ‘A year of containers: the agony and the ecstasy’. Her delivery was focused, impactful and full of truth.
In essence, the message was that one cannot be told what containers are… one needs to experience it for oneself. Nevertheless, Mia, rather than picking sides on which container tech is better instead advises us to examine the underlying (network and other) assumptions that the container technology is built on (i.e. Kubernetes and Docker swarm). In essence, pick what is best for your environment: don’t be dogmatic!
Something else Mia points out is how troubleshooting changes with an additional layer. So DevOps needs to prove out containers and graduate their usage from dev, stage and then production, all while working out the kinks and getting comfortable and confident along the way.
The benefits are tremendous: streamlined and far more nimble application deployments.
By the time she concluded her talk we were giving virtual hi-fives to her, she ended her talk with something to the effect of “... spend less time theorizing about things: just get up and do it!" - Well put Mia. 🙂
In the afternoon, we had the pleasure of catching Jason Hand’s ‘Beyond the Mean Time to Repair’. The talk focused on said metric (MTTR), why it is the magic metric, and reasons for (even though it is the magic number) why it shouldn’t be the be-all and end-all.
Jason made the point of reminding the audience that you have to prepare for failure, rather than just thinking you can prevent or handle it in code. If you think you can prevent it, you’re not developing your ability to respond to it and a failure will occur, eventually. So better to be ready for it!
Other points made during the talk were that high performing organizations respond and recover from failure 168 times faster than their peers, and that the ones that deploy more, end up experiencing a lot less issues in production than the ones that deploy more sporadically.
Finally, during his presentation Jason brought up two questions that bring a frown to IT faces: “How much did that outage cost?” and “ROI of DevOps?” Turns out the Puppet DevOps report brings in an interesting view on these questions, well worth taking some time to read it, especially since they can help drive the importance of DevOps to management.
Overall, we found the presentations we attended to be accessible and fun: there were a lot of anecdotal experiences that were shared by the presenters and the topics seemed to be well received by the audience - the questions from attendees were candid and seemed to show that they were listening.
Day 1 was also great for connecting with former colleagues and making new friends!
By the 2nd day, we barely had the chance to leave our booth, the day was spent entirely in multiple conversations with attendees about TriNimbus and the ways we help our customers on a daily basis.
There were also some great questions about Enterprise migrations, what are some good approaches to take your organization to the cloud, the common pain points and how to avoid them.
Some of the most interesting questions were about how to develop and apply a multi cloud strategy, this topic apparently concerns many of the technology organizations leaders. If this is of interest to our readers, please get in touch and we’d be happy to talk to you in more detail.
There were also a few potential candidates that were enquiring about what it mean to be a Solutions Architect at TriNimbus and what are the requisites in terms of knowledge and experience in order to apply and succeed at our company. We’re actively hiring, so please have a look at our open positions and if you think you are a potential fit, please apply!
We’re looking forward to meeting more of the community at future DevOps Days and other industry events!
(Note: In the spirit of collaboration, this post was a team effort - special thanks to our staff contributors: Fernando, Nadji and Alex!)