HPC Clusters

High performance computing accelerates scientific discovery. HPC is one of the most powerful tools you can use to gain more insight, grow your business, get better results, and maximise your investments.

ClusterVision has been engineering innovative and complex HPC clusters for sixteen years. Through our work in HPC, we have always been aiming to advance how HPC can help our customers excel even more. We don’t just offer simple hardware solutions. We are a full-service solutions provider, meaning that we take care of everything from start to finish, from A to Z. Whether it’s software, benchmarking, remote system administration, optimisation, or moving your cluster to a new location, ClusterVision has the right people and expertise to handle it.

We are vendor and supplier independent, meaning that we look for the right partners to put together your HPC solution. This allows us to fully customise and develop an individual, unique solution that perfectly matches your demands. It also enables us to be truly impartial when it comes to performance, compatibility, and cost. We are experts in providing turn-key cluster designs that fit your every need.

Services

Turn-key clusters

We can relieve you of any stress when it comes to sourcing a new cluster. From the initial cluster design to installation and instructing your staff. You don’t have to worry about a thing.

Our turn-key solutions are scalable and flexible to meet your exact demands.

Consultancy

Whether it is optimising your applications, providing you insight into cooling solutions, or checking whether your current installation still fits your needs, we have the right people you can talk to. Our staff has extensive experience in HPC, cloud, and storage.

We can even move your cluster to a new location, do a recabling of your installation, or train your staff on certain applications.

Support

Supporting and maintaining your cluster has never been easier. Our support team is highly skilled and has a wide experience in solving technical issues. We can tailor our support packages to your specific requirements and usages of your cluster.

Assembly & Quality Control

ClusterVision has a dedicated cluster production and implementation infrastructure that allows for a rapid, trouble free deployment and support of cluster solutions at your premises. Our production infrastructure includes professional project management, quality control procedures, a production line specifically designed for high-end servers, and extensive hardware burn-in procedures.

Clusters are complex products which only work to full satisfaction if all hardware and software components work together as a whole. Therefore, we subject our production process to the highest quality standards. Our dedicated production facility is specifically designed for high-end servers and server racks. Our production facility is subject to the highest standards of quality control and health and safety regulations. Our emphasis on the importance of a high-quality production line is based on our long experience in working with high-end server hardware.

Our production facility has the space and infrastructure to build complete clusters in server racks. This allows us to fully test the final solution and eliminate production errors, faulty hardware components and hardware/software incompatibilities before shipment.

Performance

In our cluster designs we always aim for maximum performance, while taking into account budget, stability, and compatibility. We try to make sure that you get the most HPC processing power out of your investment. We achieve this by carefully selecting hardware and software components from a wide variety of technology partners. Selecting the right combination of components is crucial to achieve an optimally performing cluster. Choosing wrong or ineffective combinations of components can seriously limit the performance of your cluster.

Our HPC systems architects and our project engineers work closely together so that the chosen design is installed as efficiently as possible so that you can start using your cluster within no time.

Benchmarking is a vital process in designing the ideal cluster to meet your needs. This service ensures that you get the best performance out of your cluster, while taking into account overall cost, energy requirements, compatibility, and the applications you need to run.

Our cluster eco-system TrinityX is optimised for maximum performance running typical scientific and engineering applications.

Hardware

Performance enabled by Intel Xeon, AMD EPYC, or IBM POWER9 processors

System acceleration possible with NVIDIA V100 or AMD Radeon Instinct GPUs, or Intel Xeon Phi co-processor

Several interconnect solutions available, like InfiniBand, Omni-Path, and Ethernet

Software

Supported software stacks and parallel file systems, including BeeGFS, IBM Spectrum Scale, Lustre, Panasas PanFS, Red Hat, CentOS, Scientific Linux, and our very own cluster eco-system TrinityX

Cooling

Besides traditional air cooling, we can also provide advice on active rear door rack cooling, direct-to-chip cooling, water cooling, mineral oil cooling, immersed cooling (Asperitas), and liquid cooling

CASE STUDY

Exploring the growth of galaxy clusters using HPC at the University of Hull

High performance computing enables faster and more robust scientific discovery by significantly enhancing data processing capability. The University of Hull has appropriately named their cluster Viper to reflect this speed and power. For Alex Sheardown, a PhD student at the university’s E.A. Milne Centre, HPC is vital to his research in astrophysics.