Session Description
Platform Engineering promises “Golden Paths” and low cognitive load, but usually assumes you’re in the cloud. What if your “cloud” is a multi-site supercomputer? At CSCS, we’re using the vCluster framework to turn raw bare-metal into immutable, software-defined environments. Learn how our Platform Automation team configures the Alps supercomputer to deliver production-ready clusters for everything from weather forecasting to AI - proving that even the most unforgiving hardware can be tamed with a DevOps mindset.
Speaker
Software Engineer at CSCS, bridging the gap between Cloud-Native DevOps and Bare-Metal HPC.
Michele Brambilla is a Senior Software Engineer at the Swiss National Supercomputing Centre (CSCS), specializing in Platform Engineering for High-Performance Computing. Armed with a PhD in computational physics and battle scars from instrumenting neutron sources at the Paul Scherrer Institut, he now tackles the challenge of automating bare-metal infrastructure at supercomputer scale. On the Platform Automation team, Michele develops software-defined solutions for the “Alps” infrastructure, while also serving on the observability team to ensure these massive systems remain manageable, debuggable, and - ideally- boring.
