Four Things We Learned about DevOps in London
by Angela Chapman
October 22, 2015
| 3 min read
Catching up with London DevOps:
PagerDuty recently hopped across the pond to host one of our first customer events in London. Thirty customers joined us for a discussion on the latest trends in Operations, what it means to be Operationally Mature, and how companies are adopting DevOps to drive business value. We had two phenomenal customer speakers in attendance to discuss their experiences with scaling IT teams, DevOps and PagerDuty best practices, followed by cheeky cocktails!
“Fireside” chat
Customer Speakers:
Rich Archbold, Director of Operations at Intercom, had a ton of insightful information about transitioning his team into an Operationally Mature Model. Before, Intercom’s Ops team was not measuring availability, latency, cost-metrics, and “pager pain.” Rich’s team established core values together and implemented a roadmap to specifically deal with on-call burnout and keeping the team happy.
Kit Reynolds, Project Manager with Trainline, is a long-time PagerDuty customer advocate. He’s scaled a team at two different companies and used PagerDuty at both. Kit transitioned Trainline from a helpdesk model to a DevOps model. Before, Ops at Trainline was dealing with a lot of noise and chaos. After implementing PagerDuty, teams were able to manage alert configuration themselves, which allowed them to spend more time on improving the products.
Co-Founder and CTO Andrew Miklas and VP of Products Jonathan Wilkinson address the crowd
Take-Aways:
Today’s customer expects everything to be fast and always on, so uptime is crucial. This creates an entirely new set of business challenges for organisations with complex IT departments and a need for more agile IT Ops. We reviewed the top trends we are learning from customers on the road and what you need to consider when transitioning from a more traditional IT organisation.
- DevOps is growing. People need end-to-end ownership to drive speed to market. Collaboration and tools that provide that autonomy are critical.
- Service-Oriented Architecture. The idea is that large, sophisticated systems should be broken down into small components that communicate with each other exclusively through defined APIs, and can go through the software lifecycle independently of one another. Adopting SOA enables DevOps to scale and supports graceful degradation.
- Metrics that Matter. These days, there is no shortage of data. Businesses should focus on the numbers that matter, those that the CIO actually cares about, and avoid the extra noise. It’s important to look at operational efficiency and team health by measuring metrics such as MTTR, MTTA, incident trends, hotspots and alert fatigue.
- Tooling. There are many specialised tools on the market today, so understanding the ops toolkit a business has constructed gives a lot of insight into how the business thinks about IT Ops. While there’s a desire to find “the right tool for the job,” there is also a need for all those tools to work well together.
Mingling
At PagerDuty, we pride ourselves on providing our customers with a reliable solution to improve uptime and engineers’ quality of life everywhere. Hitting the road this fall has allowed us to talk to customers from different size companies around the globe so we can continue to build the product that always has your back.
Have feedback you would like to share, or want to meet us on the road? Email events@pagerduty.com.
Cheers from the PagerDuty team!
