Tuning and Optimization for HPC

Compiler

• Most important Intel Compiler Options and Directives

Node Level Optimization

• see: Intel Software Documentation Library
  
• Quick-Reference Guide to Optimization with Intel® Compilers
  
  
• Tutorial: Using Auto Vectorization
• A Guide to Vectorization with Intel® C++ Compilers
  
• Guide to Auto-Vectorization
  
• Requirements for Vectorizable Loops
• Tutorial: Finding Hotspots - Fortran Sample Application, Linux*
• Get Started with Intel® VTune™ Amplifier
• Developing Multithreaded Applications: A Platform Consistent Approach
•  Measuring and Understanding Memory Bandwidth

MPI

• Optimization of MPI Applications - HLRS

• Intel recommendations for Autotuning

OpenMP

• Improving OpenMP Scaling

• OpenMP home page. The central source of information about OpenMP
• LRZ/RRZE Courses: OpenMP_2day_course.pdf

Tools

• Intel Performance Tools
• Information on Hardware and Topology
• Timing and Profiling
• Hardware Perfomance Counters
• MPI, OpenMP, Parallelization, Vectorization, SIMD Analysis
• Memory Leaks