Empower Your Programming Potential: Mastering Hybrid Algorithms, Libraries, and GPU Programming for Optimal Performance and Scalability.
Who is the school designed for?
Our face to face school is aimed at both technical and research staff, as well as individuals who want to enhance their technical skills. The skills you will develop can be applied across a variety of industries from finance to transportation. In finance, professionals accelerate risk analysis and portfolio optimisation with GPU programming and numerical libraries. Engineers benefit from optimised code in automotive and aerospace industries. Machine learning practitioners optimise model training and large-scale computations using GPU programming and numerical libraries. High-performance computing applications demand GPU acceleration and optimised numerical routines for efficient tackling of computational challenges.
This School has been funded by the ExaTEPP Project.
Why enrol?
During the three-day school, our experts will help you enhance your understanding and build your technical skills and increase your understanding of the benefits of using library routines over custom code, enhancing efficiency and scalability. For example, you will master developing application software using OpenACC directives, tailored for processors with massively parallel computing resources, enabling effective handling of complex computational challenges.
With our specialist support you will:
- Learn why using library routines is usually better than writing your own code, even for relatively simple operations.
- Develop an appreciation of the how numerical libraries such as ScaLAPACK have evolved as the underlying hardware has changed.
- Understand how the design of current numerical library routines extracts optimal performance from modern architectures.
- Learn the importance of uncertainty quantification in solving real life problems.
- Gain knowledge and hands-on experience in developing applications software via OpenACC directives for processors with massively parallel computing resources (GPUs accelerators)
Pre-requisites
- Knowledge of Linear Algebra, basic probability, and stochastic processes
- Basic C/C++ or Fortran competency, including familiarity with variable types, loops, conditional statements, functions, and array manipulations. No previous knowledge of GPU programming is assumed.
Create a free account on our Training Portal to register for the summer school.