Achieve the best performance for your HPC applications with our Single Node Performance Optimisation course

12–13 November, 2024 

Join us for an intensive two-day course on Single Node Performance Optimisation at The University of Edinburgh, designed specifically for developers like you. This course is delivered as part of the ARCHER2 training programme, which is free for UK academics. 

Modern HPC systems, such as ARCHER2, are built with increasingly powerful nodes featuring multiple cores, vector floating point units, deep cache hierarchies, and NUMA memory systems. To harness the full power of these systems, it’s crucial to understand and overcome on-node performance bottlenecks. 

Course content

• Efficient Programming Techniques: Learn to optimise code for multiple cores and vector units. 
• Performance Analysis Tools: Get hands-on experience with Cray performance analysis tools. 
• Compiler and Batch Processing Options: Discover how to use compiler options and batch processing to enhance performance. 
• Hand Tuning of Code: Master the art of hand-tuning your code for maximum efficiency. 

Our Single Node Performance Optimisation course is perfect for software developers who: 

• Are familiar with software development on ARCHER2 or other HPC facilities. 
• Use programming languages such as C, C++, or Fortran. 
• Are interested in optimising the performance of their applications. 

This course is free for academics based in UK institutions. Participants must bring a laptop and abide by the ARCHER2 Code of Conduct. 

Location 

The course will be held at the Bayes Centre, the University of Edinburgh's Innovation Hub for data science and artificial intelligence. The venue is well-connected and easily accessible from several cities. 

Meet the Instructors 

• Mark Bull is a Senior Research Fellow at EPCC and has been with the centre since 1998. He specialises in benchmarking, parallel algorithms, and parallel programming APIs. Mark is also a representative on the OpenMP Architecture Review Board and has extensive experience in teaching and training for HPC systems. 
• Adrian Jackson is also a Senior Research Fellow at EPCC. Adrian is the Director of Research and Co-Director of the MSc in Imaging, Vision, and High Performance Computing. His expertise spans high performance computing, data and storage, memory technologies, and application optimisation. Adrian has developed and delivered numerous courses for both academic and industrial audiences. 

How to Register 

Register now to secure your spot. 

Join us and take the first step towards becoming an expert in single node performance optimisation. We look forward to seeing you in Edinburgh!