London OpenInfra May 2023 Meetup

It's been a minute, London OpenInfra fam! We're back in May for the first in-person meetup in what feels like forever, prompted by the perfect excuse of Kendall Nelson from the OpenInfra Foundation being in town.

We'll be graciously hosted and sponsored by G-Research at their new event space (Whittington House 19-30 Alfred Place, London WC1E 7EA).

The agenda is still being finalised with additional talks - keep an eye out for updates.

6:30pm - Deploying and managing baremetal Kubernetes with Ironic - Scott Solkhon, Cloud Engineer at G-Research
7:30pm - Self-service Kubernetes Platforms with RDMA on OpenStack - John Garbutt, Principal Engineer at StackHPC

· Deploying and managing baremetal Kubernetes with Ironic (Scott Solkhon)

G-Research uses Armada to distribute millions of batch jobs per day, across many 1000's of nodes, across many baremetal and virtual Kubernetes clusters. But how do we build and provision all of the nodes that make up our HPC farms within our private OpenStack cloud?
Ironic, of course!

Whether its an initial power on of a node to check that we got what we paid for, running workloads, moving a node from one network to a another, checking for cabling errors, or ensuring nodes are secure, compliant, and have firmware that is up to date, Ironic underpins the tooling and automation that drives the enrolment, provisioning and recycling of baremetal hardware across our datacenters. Come to this talk if you want to hear some of our successes, failures, and a few lessons learnt during G-Research's journey in moving Armada clusters from virtual machines to baremetal.

· Self-service Kubernetes Platforms with RDMA on OpenStack - John Garbutt, Principal Engineer at StackHPC

Azimuth helps users self-service create Science Platforms, such as JuyterHub and Slurm. Sometimes this requires self-service creation of RDMA enabled Kubernetes clusters. OpenStack can use SR-IOV using VF-LAG to provide RoCE RDMA within VMs. We make use of K8s Cluster API to provision K8s using OpenStack servers. We then use multus and macvlan CNIs to give k8s pods RDMA networking. Testing the performance is automated using a Volcano based K8s operator. We are working on also bringing this power to OpenStack Magnum.

NOTE: You must RSVP 48 hours prior to the event in order to be able to attend!