Intel HPC Advanced Training
This training is meant for advanced HPC developers and programmers to take advantage of latest Intel processors . We offer deep-dive into maximizing code performance and Develop high-performance parallel code and accelerate workloads across enterprise, cloud, high-performance computing (HPC), and AI applications using Intel Parallel Studio XE.
Product availability: AGAINST PO
FEATURES
- under this training will cover the the Advanced topics related to HPC like Open MP, MPI, vectorization and Code Modernization etc..
- Each topics has hands on section with sample programs
- Details explanation of Intel Tools and Patrice with Tools
SPECIFICATIONS
- Understanding Processor trends – Single/Multi/Many Cores, Advanced Processor Features -Instruction Pipeline,Branch Prediction, Hyper Threading, Vectorization, Concept of Cache. What is a cluster ?
- Introduction to Intel Parallel Studio. Details on Intel Compiler and its feature set. Specifics of important switches,
- Task and Data Parallel models. Shared and Distributed Memory systems.
Open MP:
- Introduction to threads, creation of threads, false-sharing, Work Sharing, Data Environment, Synchronisation.
Open MP Lab:
- There will be 3 exercises. Starting from a simple Hello world in OpenMp to Computing value if pi using area under the curve.
MPI:
- Introduction to MPI, Creating Processes, Groups/Ranks/ Communicators, One-sided/Cooperative communication, Introduction to advanced topics of topology, User Defined Data Types.
MPI Lab:
- There are 3 exercises, starting from a simple hello world program,
adding numbers, to passing token in a specific order.
Vectorization:
- Concepts and programming examples
Intel Tools:
- Introduction to these tools,capability, salient features. Vtune-demo
Code Modernisation:
- How to work with an existing program to make it more efficient. Use of Compiler optimization, More efficient Libraries, Adding threads, scaling with many cores.