The modern business organization revolves around the capabilities of their IT infrastructure. More than anything else – they rely on fast application delivery and performance. Here’s the reality – high-performance computing has allowed companies to create powerful density models around their most critical applications. Now, as more users and data points require even more resources – organizations must look at next-generation supercomputers for even greater optimization. But how do you deploy these machines for true agility? How can you ensure configuration control as well as application performance?
In this case study from Cray – we take a look at a powerful new model behind high-performance computing. Cluster supercomputers, like the CS300 system, are modular and energy efficient clusters with flexible configurations designed for medium- to large-scale deployments. Being able to scale is one thing – but what about efficiency? New supercomputer clusters now take all of these environmental and performance aspects into consideration. Platforms like the CS300 are based on industry-standard building block platforms and feature the latest processor, coprocessor and accelerator technologies. They address massively parallel applications and large-scale data analytics challenges. Here’s the cool part – they’re designed to have either air-cooling or liquid-cooling capabilities.
As the case study outlines – With funding from the National Science Foundation (NSF), the National Institute for Computational Sciences (NICS) at the University of Tennessee built built “Beacon,” a Cray CS300-AC cluster supercomputer with 48 compute nodes and six I/O nodes joined with InfiniBand connectivity. In all, the cluster includes 768 conventional cores and 11,520 accelerator cores, delivering the hybrid environment NICS needed to explore a variety of programming and processing scenarios. Additionally, NICS configured Beacon with Intel SSDs and stepped up to 256 GB RAM per node (a large amount at the time), enabling them to explore data movement and new data paradigms as well as other paradigms of computational use.
Download this case study today to learn how NICS at the University of Tennessee has seen up to 4.5 times better performance using a single Intel Xeon Phi coprocessor compared with a processor from the Intel Xeon processor E5 family. Remember, application density and high-performance computing are all critical aspects in designing a system capable of powerful data analysis. However, you also need a platform which can scale and create direct cooling and power efficiencies as well. Ultimately, this helps you crunch numbers faster – and improve overall HPC efficiency.
Get this case study from the insideHPC White Paper Library.