Are you curious to learn all about CernVM-FS for Containers? Watch the video recording of the past webinar here!
Objective of the webinar
Learn all about the latest developments with regard to CernVM-FS container integration.
Target audience
Site administrators and cluster administrators; CVMFS power users.
Webinar programme (1h)
-
Birds-eye view of CernVM-FS
-
Container distribution: some key figures
-
Using CernVM-FS repositories with container runtimes (docker, singularity, k8s)
-
Distributing operating system containers from CernVM-FS
-
Publishing in containers
-
Q&A (15 min)
Description of the webinar
Delivering complex software stacks across a worldwide distributed system is a challenge in high-throughput scientific computing. The global-scale virtual file system CernVM-FS distributes more than a billion software binaries to hundreds of thousands of machines around the world.
In this webinar, we will present the latest developments with regard to CernVM-FS container integration. Containers and CernVM-FS team up nicely: containers provide the isolation capabilities that decouple the application stack from the underlying platform and CernVM-FS provides efficient distribution means for the containerized software binaries. Containers are an enabling technology to harness opportunistic resources and HPC facilities. CernVM-FS enables the use of such resources at scale. In this webinar, we will show how existing repositories can be used with several popular container runtimes, such as docker, podman, singularity, and kubernetes. We will also show how operating system containers themselves can be efficiently distributed through CernVM-FS. Lastly, we will highlight an upcoming new way of publishing content from within a container. This makes it easy to set up, build and test and deploy-to-cvmfs pipelines on kubernetes.
About the speaker
Jakob Blomer is a staff member in the scientific software group at CERN. He is the original author and lead developer of the CernVM File System, which has been developed since 2009. Jakob received a PhD in computer science from the Technical University of Munich.
