Source: Kubernetes London

Kubeflare

Kubernauts,

2019 marked Kubernetes as one of the most successful OSS [collaboration] projects of the previous decade and what better way to kick off the new decade than by meeting up for a night full of Kubernerding!

Please join us on Tuesday, 11th February for the night of great talks we've lined up for you! We will mostly focus on the operators and how to make Kubernetes success at large organizations as uSwitch.

This event would not be possible without our marvellous sponsors: OpenShift and Cloudflare. Let's give them a round of virtual applause (**silent roar**).

Agenda:

6:30 - 7:00 Socialise, make friends, eat, drink and be merry.
7:00 - 7:20 Databricks and CronJobs operators, Stuart Leeks
7:25 - 7:45 Ukube: U get better abstractions. Kubernetes at uSwitch, Joseph Irving
7:50 - 8:00 break
8:00 - 8:20 ETCD Operator by Jetstack, James Laverack
8:30 - 9:00 PANEL hosted by Miles Bryant, where the audience will be able to ask everything you wanted to know but were scared to ask... to our panellists: James Munnelly, Stuart Leeks, Joseph Irving and Steve Wade.

Name: Stuart Leeks
Title: Databricks and CronJobs operators
Bio: By day, Stuart Leeks is a Software Engineer working for Microsoft in the UK. He has worked with a wide range of customers from small ISVs to large enterprises to help them be successful building with the Microsoft technology stack. Whilst Stuart has experience of a diverse set of technologies, he is most passionate about the web and cloud.
Stuart is a web geek, lover of containers, cloud nut, feminist, performance & scalability enthusiast, father of three, husband, salsa dancer & teacher, and loves bad puns. He has been writing code since the days of the BBC Micro and still gets a kick out of it.
@stuartleeks

Abstract: The Microsoft Commercial Software Engineering team works jointly on projects with customers and as part of a current engagement we have been working on a couple of operators. The first is an operator for Azure Databricks allowing clusters and runs to be created and managed via Kubernetes CRDs. The second is an operator to ensure that time-sensitive cron jobs run at the correct time in a cluster with node autoscaling enabled (i.e. where waiting for the cluster to scale up to have capacity might delay the job too long). This session will take a look at the lessons we learned on this journey.

Name: Joseph Irving
Title: Ukube: U get better abstractions
Bio: Joseph is a platform engineer at RVU (Uswitch), he’s currently working on simplifying multi-cluster setups and taking two years to get PRs merged into Kubernetes. He previously worked at the UK Home Office and has been running Kubernetes Clusters since the alpha days.

Abstract: When we bet on Kubernetes, we bet on abstractions. It allowed us to unify our infrastructure and gave us a common language to solve problems. In this talk we will go over a few Open Source tools we’ve created that best demonstrate the power of this language:

- Mutating webhooks and sidecars to give applications short-lived database credentials;
- Envoy control planes to give us HA over multiple Kubernetes Clusters;
- Controllers to generate Dashboards and Alerts for teams based on SLOs.

Name: James Laverack
Title: ETCD Operator by Jetstack
Bio:

Abstract:

Newsletter
  • Get the latest DevOps jobs, events and curated articles straight to your inbox, once a week

  • Community Partners