Case Study

Queen Mary University of London

Univa Grid Engine Maximizes University HPC Cluster Performance Several Orders of Magnitude

Globally recognized for pushing the boundaries of research and innovation, Queen Mary University of London (QMUL) runs a high-performance computing environment that supports over 2,000 students and researchers across all disciplines. The HPC cluster comprises 5,000 InfiniBand-interconnected cores (including nodes for genome analysis and GPU nodes for Machine Learning) and 2PB high-performance storage running hundreds of commercial and open-source applications of various types.

The University's previous open-source workload scheduler presented performance-impacting bugs that seriously impaired the usability of the cluster. While restrictions and workarounds were implemented, ultimately students and researchers could no longer run their preferred software on the cluster.

QMUL needed a future-ready enterprise-grade HPC workload orchestration solution backed by solid support and expertise. They selected Univa Grid Engine for its rich features, high performance, large installation base, expert support, and easiest upgrade path. The University's migration to Univa was "painless" with consistently high resource usage realized immediately. Most importantly, with the substantial performance improvements, researchers and students are now utilizing their preferred software regardless of application type.

"Upon deploying Univa Grid Engine, the impact was immediate," Simon Butcher, Head of Research Applications, Queen Mary University of London said. "We now experience consistently high usage from our HPC cluster and have gained new flexibility that will take us into the future."

To access the case study .pdf, please complete the form below. Fields marked with (*) are mandatory.