Case Study: CERN achieves faster large-scale data analysis with Google Cloud Platform

Helping researchers at CERN to analyze powerful data and uncover the secrets of our universe

CERN, the European Organization for Nuclear Research in Switzerland, analyzes petabytes of data each year from the Large Hadron Collider and other experiments. As data volumes and compute needs surged, CERN needed a way to rapidly boost storage and processing capacity for highly variable, resource-intensive workloads such as particle physics analysis and machine learning. Google Cloud Platform was used through a CERN openlab project to explore cloud-based infrastructure.

Google Cloud Platform implemented Google Kubernetes Engine, Cloud Storage, and access to GPUs and other accelerator hardware to scale workloads quickly and flexibly. CERN reran a Higgs boson analysis that had taken more than 24 hours in 2012, and on Google Cloud it completed in just over 5 minutes using 25,000 cores. Google Cloud Platform also helped CERN achieve up to 5x cost savings on short jobs with preemptible VMs and sped up data access to 200 GB per second, supporting faster research and future discovery.

CERN

Ricardo Rocha

Computing Engineer