Bull Information Systems has signed a strategic alliance deal valued at around £800,000 with the University of Liverpool to install a 40 Teraflop High Performance Computing (HPC) system.

The deal incorporates a five-year partnership and support agreement under which Bull will implement an integrated 40 Teraflop HPC system for the University based around its own ‘bullx’ blade hardware and bullx supercomputer software suite.

The University of Liverpool has established a significant group of HPC users who increasingly rely on access to a centrally housed cluster resource but its current clusters have reached ‘end of life’.

The university also wants to meet a longer-term goal of ensuring that the quality and profile of the HPC-based research it is undertaking. 

Bull builds and designs all of its own HPC solutions and delivers these directly to the customer. As a primary manufacturer and supplier, it also always contracts directly with HPC customers, according to Bull. 

Cliff Addison, manager of Advanced Research Computing at the University of Liverpool said the university chose Bull because it could provide not only with high-quality HPC solutions but access to extensive expertise and new thinking in HPC and ongoing guidance in implementing and running the solutions and the ability to test out the latest technologies at its industry-leading benchmarking facility.

“We also liked Bull’s specific knowledge of and commitment to open source software and its willingness to contribute proactively to the engagement process,” Addison said.

Built on Intel’s new ‘Sandy Bridge’, Bull’s solution for the University has the capability to deliver stable and impressive performance over a range of research applications, Bull UK & Ireland CEO Andrew Carr said.

All of the main compute blades in the Bull implementation are concentrated into a pair of water cooled racks thereby minimizing space and cooling infrastructure demands.

Bull said the in-built Infiniband fabric allows for rapid data transfer and supports job runs ranging from throughput to highly scalable; utilization can therefore be expected to be high which given the inherent stability, resilience and comprehensive support offered will maximize ‘up time’ and therefore  deliver a superior return on investment (‘ROI’).