Enabling AVX-512 Vectorization and Using Intel Performance tools

NewSpecial

• This event is being organised jointly between UIS and the Centre for Theoretical Cosmology at DAMTP.

• In this workshop, we will start with an overview of the latest micro-processor architectures and how the intrinsic parallelism has been implemented in hardware, mainly the SIMD instructions and multi-threading. Then we focus on how to define and measure processor and memory performance and how this is related to the application level. In particular we describe the roofline model approach, which gives an estimation and a visual model useful to estimate the application performance and the limitation of the underlying hardware.

• With the help of examples and use cases, we pinpoint you to possible inefficiencies both on threading and vectorization and we explain remedies, hints and strategies to be considered to ensure an application delivers great performance on today’s scalable hardware and upcoming future generations.

• Furthermore we will show how performance analysis tools like Intel® Advisor and Intel® VTune™ Amplifier, together with examples and use cases, pinpoint you to inefficiencies both on threading and vectorization and also give hints to remedies.

• Attendees should be comfortable with either C/C++ or Fortran programming language and basic Linux command, like make and ssh. No previous experience in vectorization and parallelization is required and profiling tools, as well.

• Please note that if you are not eligible for a University of Cambridge Raven account you will need to book by completing this form.

• All current University members (departments and colleges) who are software developers looking to improve performance of scientific applications on recent intel hardware.

• Knowledge of either C, C++ or Fortran.
• A laptop to bring to the workshop (which will be used to connect to VMs provided).

Number of sessions: 2

• Introduction of Parallel Studio, including all the Cluster tools
• Modern Computer Architecture
  • Moore's law
  • SIMD instructions
  • Simultaneous multithreading
  • Cache and Memory System
• Roofline model
  • How to measure performance
  • Arithmetic intensity
  • Roofline chart
• Code optimization process using Intel C++ Compiler and AVX-512 with nbody sample
  • Introduction to code modernization approach
  • Scalar and serial optimization
  • Vectorization
  • Parallelization
• Profiling tools and libraries
  • Intel Advisor
  • Intel VTune
  • Intel MKL