DevOps Stage speakers · 2019

Roman Chepurnyi & Oleksii Asiutin

Director of Infrastructure Engineering / Staff Software Engineer, Infrastructure, thredUP

United States / Ukraine
Talk language: English

Roman Chepurnyi and Oleksii Asiutin (thredUP) gave a joint talk “Kubernetes Navigation Stories” at DevOps Stage 2019 in Kyiv, 18–19 October 2019.

About the speakers

Roman is a Director of Infrastructure Engineering at thredUP – the largest online consignment store for women's and children's clothes. Before thredUP he managed DevOps teams at Hotwire – a travel website that offers airline tickets, hotel rooms, rental cars, and vacation packages. Roman has a vast experience building infrastructure and infrastructure teams and loves to share his knowledge. His main professional interests are Cloud Infrastructure, Security, Site Reliability, Engineering Tools, Performance and Test Engineering. With his strong leadership skills and open-minded creative cultural views Roman builds super-productive engineering teams. In his spare time he likes sailing in the San Francisco Bay

Oleksii is a Staff Software Engineer at thredUP. In the past he worked both as a Software Developer and a System Administrator and it helps him to populate DevOps culture among teams. He is interested in Kubernetes ecosystem and works with it's toolset a lot. He also writes automation tools which helps other engineers to be more productive. Apart from his Infrastructure day-to-day goals he enjoys to code, teaching and mentoring engineers to build their career path. His personal interests are sport, driving and art.

Talk

Talk · DevOps Stage 2019 · Kyiv · Stream C · Day 1, 18:15 - 19:00

Kubernetes Navigation Stories

thredUP is the largest online consignment store for women’s and children’s clothes. thredUP was launched in 2009 as a monolithic application running on Amazon Web Services. Since then it has evolved into a fully containerized distributed system powered by Kubernetes

We have started Kubernetes evaluation in 2017 and within a year migrated all production services on it. Now we are 100% in Kubernetes for 1.5 years. Migration has enabled faster releases with a cycle time under 20 minutes, decreased hardware cost by 60% while a number of services has doubled, unified infrastructure automation, decreased new service rollout time from weeks to minutes/hours.

Was our Kubernetes journey finished after the migration? Definitely not. thredUP team will share key learnings from after-migration processes about what technologies and solutions worked best for us and where we spent time troubleshooting and improving. In particular we have focused on development and staging experience, user authentication, cloud-native CI pipelines, applications telemetry and service mesh. We also share our experience with Kubernetes security hardening, autoscaling and tell you about a new service creation within our infrastructure.

All these Kubernetes Navigation Stories are from captains who sails regularly.

From the DevOps Stage conference archive (2019). This site is not connected to that conference or its organizer — see our disclaimer. Speakers can ask for changes or removal via hello@devopsstage.com.