High Performance Computing

🚩 About This Course 

➡️ This course provides a comprehensive foundation in the principles and practices of "High-Performance Computing".

➡️ Students will transition from sequential programming paradigms to mastering the tools and techniques required to leverage modern parallel architectures. The curriculum covers fundamental models (OpenMP, MPI, CUDA), explores the architectural features of CPUs and GPUs, and investigates cutting-edge technologies like TPUs and quantum computing.

➡️  A strong emphasis is placed on practical, hands-on experience through programming assignments and a final project, enabling students to design, implement, and optimize efficient parallel solutions to computationally intensive problems.

🎯 Learning Outcomes

This course provides a solid foundation in High Performance Computing (HPC) and its role in computational science.

Upon successful completion of this course, students will be able to:

• −      Explain the motivation and theoretical foundations of parallel computing, including Amdahl's and Gustafson's laws.
• −      Compare and contrast shared-memory (multicore) and distributed-memory (cluster) architectures and their corresponding programming models.
• −      Design and implement parallel programs using standard APIs: OpenMP for shared-memory, MPI for distributed-memory, and CUDA for GPU acceleration.
• −      Analyze the performance of parallel applications, identify bottlenecks, and apply optimization techniques.
• −      Evaluate emerging HPC trends (e.g., TPUs, Quantum Computing) and their potential impact on scientific computing and AI.

🪛 Pre-requisites: 

−         Evolved Computer Architectures,

−         Operating System I & II.

−         Proficiency in C, Python, and Bash (highly recommended).