DHI Group Picks Bright Cluster Manager for Supercomputer Capacity Planning 

Print Friendly, PDF & Email

Amsterdam, Netherlands – Wednesday, November 4, 2020 – Today, Bright Computing, maker of Linux cluster automation and management software for HPC and machine learning, announced that DHI Group has upgraded its supercomputer, which now includes Bright Cluster Manager, to support its HPC environment  to extend to the public cloud for additional resources when needed to support pioneering environmental research projects.  Primal

DHI Group (Danish Hydraulic Institute) addresses challenges in water environments, how to save it, share it fairly, improve its quality, quantify its impact, or manage its flow. With over 50 years of experience in dedicated research and real-life experience, DHI Group combines its knowledge and expertise with technology to provide the right solutions for its global client base.

With clients requiring increasingly complex computations in areas such as wave modeling, ocean acidification, and offshore  wind energy, and with the business growing, DHI Group’s existing supercomputer was in need of modernization, and the IT group sought tools to help them improve workload management and capacity planning. In 2020, DHI Group took the decision to upgrade its supercomputing environment, with the goal that an increase in compute power would enable the Group to manage more computations simultaneously, process them more quickly, and achieve results faster.

DHI Group selected Dell Technologies to build and install the new supercomputing environment, which includes 360 cores and 2 GPU processors. During the selection process, DHI Group made the decision to include Bright Cluster Manager in the architecture. Bright Cluster Manager will provide comprehensive management of the DHI Group’s new supercomputing infrastructure, enabling them to administer HPC resources as a single entity, provisioning the hardware, operating systems, and workload managers from a single unified web interface.

With the previous environment, when there was a need for additional computational power, DHI Group had been forced to lease external facilities. With Bright Cluster Manager, DHI Group has much better control over workload management and capacity planning, using Bright’s intelligent monitoring and automation capabilities to manage its resources more effectively and efficiently. In addition, Bright Cluster Manager has the power to enable DHI Group to seamlessly burst into the cloud for more compute power, as needed.

Dell Technologies carried out the implementation in August 2020, using Bright Cluster Manager to provision and deploy the supercomputing environment from bare metal. The platform has been performing well since then, servicing the evolving needs of DHI Group’s clients.

Ole Svenstrup Petersen, Marine and Maritime R&D Area Manager at DHI Group, commented, “DHI Group is involved in some really exciting projects such as supporting offshore wind exploration to lower costs and support the green transition. These globally important areas of research demand state of the art technology and that’s where the HPC solution from Dell and Bright Computing comes in.”

Dan Kuczkowski, SVP Sales at Bright Computing, added, “DHI Group has harnessed the true power of a Dell and Bright solution, providing the very best HPC platform and technology to underpin this critical environmental research. We are delighted that Bright Cluster Manager is playing a part in these ground-breaking research projects.”

Petersen added, “Bright Cluster Manager is simple to use but powerful. We are currently using it to manage our HPC environment, and we are excited to explore how Bright can support our deep learning and our edge environments in the future.”

Bright Cluster Manager provides infrastructure management technology for Dell’s HPC systems to enable enterprises such as DHI Group accelerate their science. Together, Dell and Bright speed deployment and simplify operations and management of solutions for virtualization, hybrid cloud, deep learning, data science, HPC, and more.