Architecture
Digital Wizards
About Architecture
Explore the cutting-edge technologies behind High-Performance Computing (HPC)! This domain focuses on the principles of CPU and GPU architectures, parallel processing, and system scalability. Learn how modern processors are designed, how GPUs accelerate computations, and how parallel algorithms transform the way we solve large-scale problems. Whether you're passionate about optimizing performance, exploring distributed computing, or advancing HPC applications, this domain provides the knowledge and skills to push the boundaries of computational efficiency.
Events & Sessions
Introduction to CUDA Programming
2024-11-13
This session introduced the basics of CUDA programming, focusing on GPU computing and parallel processing concepts. Participants learned foundational skills to write and execute simple CUDA programs, gaining insight into how GPUs accelerate computation. The session provided a starting point for exploring high-performance programming and solving real-world computational challenges.
Ongoing Projects
Parallel Computing for GANs
Implementing a Generative Adversarial Network (GAN) using parallel computing techniques to accelerate training, manage larger datasets, and enhance output quality. The project utilizes CUDA for GPU programming and PyTorch for GAN development, requiring access to the Institute's GPU cluster for testing and validation.
