Skip to main content

CERN selects software tools to enable its HPC cloud

CERN, the European Organisation for Nuclear Research, is using Platform Computing's LSF grid infrastructure to pilot its cloud computing environment for scientific collaboration. Using Platform’s private cloud management and HPC cloud-enabling software solutions, Platform ISF and Platform ISF Adaptive Cluster, CERN believes the cloud project will allow it to deliver increased computing performance and offer better infrastructure services to its 10,000 researchers from 85 countries.  The scientists are working at the new high energy frontier in particle physics to explain fundamental mysteries of the universe such as why particles have mass and the nature of all the 'missing mass' in the universe.

'For CERN's cloud computing initiative, we needed an infrastructure that would support our existing grid in a heterogeneous environment that could manage both the VMs and physical machines necessary for our researchers to run projects smoothly since their computing needs change constantly as the data is processed,' said Tony Cass, group leader, Fabric Infrastructure and Operations, CERN. 'Platform’s ISF and ISF Adaptive Cluster, combined with the Platform LSF grid workload management solution already in place, will provide our users the scalability and flexibility they need to manage their clusters and share data centre resources while adhering to our requirements for open standards.'

At CERN, massive amounts of scientific data are processed and must be distributed to researchers in near real time. As a result, CERN’s cloud infrastructure has to provide the capacity necessary to support production and analysis of more than 15 petabytes of data per year, processed by 60,000 CPU cores, allowing scientists to manage workloads themselves as opposed to a centralised IT management department at CERN’s laboratory near Geneva. Because CERN uses Platform’s LSF grid and workload management solution to enable the extensive scalability to analyse its vast research data, the laboratory chose to partner again with Platform to explore how to more effectively utilise their resources in a virtualised cloud environment. 

Media Partners