Parallel programming is a subject we have avoided for too long. But this is just one of the few optimisations that seem not to be in every programmer’s toolbox. During this one-day training you learn how to write software that makes full use of a modern X86-processor, PCI-bus and network. Together with several other performance programming techniques you will be able to make software fly.
After the introduction, where you are given a theoretical base in hardware and software concepts, we go through the following subjects:
- Cache-usage improvement.
- Memory-usage decrease.
- Scalability improvement.
- Programming language specific performance tuning.
- Auto-tuning.
- Function costs awareness.
- Picking better algorithms.
During the lab-sessions you will get hands-on with typical performance-problems.
The training is available for Java, .NET, C and C++.