Large Scale Computing

In this joint SURF-DTL Special Interest Group we bring together the expertise in building and providing a computational infrastructure for large-scale life science research data processing. We discuss requirements, use cases, infrastructure solutions and models to scale out compute resources between medical centers, research institutions, SURFsara, and internationally if needed. Models to scale-out, e.g. temporary expand your capacity to remote locations, include federation of compute resources and virtualisation.

Why?

Simulations and data volumes that need to be processed are steadily increasing in size. As a consequence, infrastructure for high capacity and high throughput computing becomes increasingly important, also in relatively new areas such as Big Data and AI/ML. Local compute resources are often needed for the analysis of on-site data and low-threshold access for development, education and regional collaboration. As data volumes and simulations grow, local resources may become too small and data and software must migrate to larger resources, without interrupting the research significantly. The use of cloud computing is also becoming increasingly important.

The purpose of this SIG is to inform, discuss and exchange information, experiences and best practices on topics that fall within the scope of the SIG.

What?

The SIG addresses a wide range of topics related to Large Scale Computing. The scope is primarily at the operational level.  Important issues that need to be addressed on a tactical/strategic level will be sent to the appropriate bodies. Topics that are relevant to include:

• High Performance Computing (capability), High Throughput Computing (capacity) and Cloud Computing
• Infrastructure and technologies (hardware and software) for analysis, storage and transport
• System management, including expertise, optimization of usage, data management, security and privacy
• User support and user needs
• Migration of users, applications and data between Tier-3, Tier-2, Tier-1 and cloud resources
• HPC applications in science and education, including chemistry, physics, astronomy, climate, life sciences, engineering, Big Data and AI/ML
• Scientific software (development tool and methods, scalability, sustainability, engineering/expertise)
• Sustainable solutions, green computing, data center technology

The SIG organizes regular seminars. Presentations can be given by members of the SIG, but also by third parties (e.g. scientists, suppliers, other institutions)

Who?

The SIG is open to all experts that manage and support HPC systems (Tier-1, Tier-2, Tier-3) and cloud-based solutions at universities, medical centers and research institutes, but is also open to those who apply HPC in a broad spectrum of science and education, researchers and scientific programmers.