This high-performance computing platform supports the work of a large consortium of UK scientific institutions leading the work in the development of new informatics tools and implementation of advanced analysis of crop genetics diversity data.
The development of improved crop varieties is vital to providing a sustainable and secure food supply to address some of the grand challenges of our time. Breeding more nutritious and climate resilient crop varieties is required to meet the demands of a growing population, but lack of available and accessible genetic diversity has historically limited progress. Developments in sequencing and genotyping technologies together with advances in environmental monitoring and characterisation are leading to rapid changes in the opportunities that are available to evaluate and utilise genetic diversity in crop plants and their wild relatives to adapt to the changing landscape of agriculture. Advanced bioinformatics and data analysis tools enable us to manage, model and mine vast amounts of genomic data to identify genetic markers and traits that provide the key to better varieties and new crops.
120 physical compute nodes offer access to a variety of Intel and AMD CPUs providing 5,224 compute cores (10,448 threads) in total. We use both AMD EPYC and Intel Xeon Scalable processors. A large complement of Nvidia-enabled GPU nodes provide a further 500,000 CUDA cores.
The cluster has 41 TB of available memory. Several large memory machines provide capacities from 1 to 4 TB, whereas the compute nodes have between 192 and 512 GB.
An multi-node BeeGFS parallel storage system (running over a 100-gigabit network backbone) provides access to 8 petabytes of storage. This is complemented by an addtional 5 PB of backup capacity.
This resource is available to researchers (and select collaborators) from any of the following UK research institutes:
The bulk of funding for this HPC resource was awarded as part of a BBSRC 18ALERT grant (BB/S019669/1) but all of the entities listed below have contributed in some way to either the purchase of hardware, software, or assistance with running and/or staff costs.
The HPC cluster was designed in collaboration with Dell Technologies and the input of their Public Sector Higher Education and Research Team. The cluster uses PowerEdge Servers, PowerVault Storage and Dell Networking.
Learn more at Dell Technologies - HPC