Event Details

Please note: All times listed below are in Central Time Zone


Provide TurnKey container clusters on OpenStack

A very popular way to run workloads on a cloud infrastructure is containers and when a workload has certain size you need a container orchestrator. The solution that stands out the most is kubernetes, but Docker swarm Mesos/DCOS are popular too. End-Users can use the OpenStack APIs to deploy kubernetes or another solution and manage the whole infrastructure. Magnum is a TurnKey solution for container clusters, including Kubernetes, Docker Swarm and DC/OS.

 

We will share how we run a container service and lessons learned from running Magnum in production. We will present how Cloud operators can provide a managed container service allowing users to manage only their applications and not worry about managing the infrastructure. Magnum’s focus is cluster lifecycle operations both from the user’s and operator’s perspective, so common user requests are scaling on demand, upgrading the infrastructure, monitoring and repairing clusters.


What can I expect to learn?

From this talk, you can expect to learn how to provide a turnkey container infrastructure service on OpenStack.

  • Network architecture for container cluster
  • Security best-practices for kubernetes
  • Optimize OpenStack resource usage
  • Container cluster upgrades
  • Monitoring clusters and applications
  • Repairing clusters
Thursday, November 15, 4:20pm-5:00pm (3:20pm - 4:00pm UTC)
Difficulty Level: Intermediate
Software Engineer, CERN
Spyros Trigazis is a computing engineer and a member of the CERN Cloud infrastructure team which provides computing resources to the High Energy Physics community. Spyros is the PTL of the Magnum project since the Pike release. FULL PROFILE
Catalyst IT Limited
Feilong is an active contributor for OpenStack since 2012 and was a Glance core member and Zaqar PTL. Now He is mainly working on Magnum and Kubernetes. FULL PROFILE