GPU Programming with CUDA
Direct GPU control for custom kernels: essential when PyTorch abstractions aren't enough. This comprehensive curriculum is designed to give you hands-on experience with GPU Programming with CUDA.
About this Course
Direct GPU control for custom kernels: essential when PyTorch abstractions aren't enough. This comprehensive curriculum is designed to give you hands-on experience with GPU Programming with CUDA. This HPC & ML Infrastructure Engineers curriculum is designed to give you hands-on experience and deep conceptual understanding.
Across 6 intensive modules, you'll tackle real-world challenges and build practical projects that reinforce your learning. By the end of this journey, you'll have the skills and proof of work to demonstrate your expertise.
What you'll learn
Master the core concepts of cuda architecture & execution model.
Gain hands-on experience with cuda memory hierarchy: shared, global & constant.
Understand the architecture behind warp divergence, parallel reduction & atomic operations.
Implement production-grade building custom pytorch c++/cuda extensions.
W1
CUDA Architecture & Execution Model
CUDA Architecture & Execution Model
Master the core concepts of cuda architecture & execution model.
3 videos•25m
2 readings
3 topics
1 homework
Topics
1.1
GPU Architecture Explained: How NVIDIA & AMD Power Modern AI
GPU Hardware Architecture & Streaming Multiprocessors
8 minutes
1.2
CUDA Explained for Beginners | Threads, Blocks, Grid (GPU Programming Made Easy)
Thread Hierarchy & Grid Launch Configuration
9 minutes
1.3
CUDA Memory Coalescing Explained: Access Pattern Optimization for GPUs | Uplatz
Memory Allocation & Data Transfers
8 minutes
References
Week 1: CUDA Architecture & Execution Model
Week 2: CUDA Memory Hierarchy: Shared, Global & Constant
01
Understanding CUDA Memory Hierarchy: A Practical Guide
alirezalm.github.io02
CUDA Refresher: The CUDA Programming Model - NVIDIA DeveloperCUDA Programming Guide - NVIDIA Documentation HubMemory Hierarchy — CUDA ProgrammingMemory Hierarchy in CUDA: Global, Shared, and Constant MemoryCUDA Memory Management: Global, Shared, and Constant MemoryThe Hidden Bottleneck: How GPU Memory Hierarchy Affects Your ...
developer.nvidia.com03
CUDA Programming Guide - NVIDIA Documentation Hub
docs.nvidia.comWeek 3: Warp Divergence, Parallel Reduction & Atomic Operations
Week 4: Building Custom PyTorch C++/CUDA Extensions
Week 5: Shared Memory and Synchronization
Rate this course
Community Insights
0Join the discussion
Sign in to share your thoughts and technical insights.
Loading insights...