Paul Graham

Accelerating Python on GPUs

Is your Python code hitting a performance wall? Learn how to leverage the massive parallelism of GPUs with minimal code changes.

Accelerating Python on GPUs
#1about 2 minutes

The rise of general-purpose GPU computing

NVIDIA's evolution from a graphics hardware company to a leader in general-purpose computing was accelerated by the use of GPUs for AI with models like AlexNet.

#2about 4 minutes

Why GPUs outperform CPUs for parallel tasks

As single-threaded CPU performance plateaued, GPUs offered a path forward with their massively parallel architecture designed for simultaneous computation.

#3about 6 minutes

Understanding modern GPU architecture and operation

GPUs work with CPUs by offloading compute-intensive code and use thousands of threads to hide memory latency, leveraging streaming multiprocessors and high-bandwidth memory.

#4about 7 minutes

Introducing the CUDA parallel computing platform

The CUDA platform is a complete ecosystem with compilers, libraries, and frameworks that enables developers to program GPUs using various languages and abstraction levels.

#5about 3 minutes

Leveraging specialized hardware like Tensor Cores

Specialized hardware like Tensor Cores can be used transparently through high-level libraries like cuDNN or programmed directly with low-level APIs for maximum performance.

#6about 6 minutes

High-level frameworks for domain-specific acceleration

Frameworks like Rapids provide GPU-accelerated, drop-in replacements for popular data science libraries such as Pandas (cuDF) and NetworkX (cuGraph) with minimal code changes.

#7about 10 minutes

A progressive approach to programming GPUs in Python

Developers can choose from a spectrum of Python libraries, from simple drop-in replacements like CuNumeric and CuPy to JIT compilers like Numba and direct kernel programming with PyCUDA.

#8about 6 minutes

Developer tools and learning resources for GPUs

NVIDIA offers a comprehensive suite of developer tools for profiling and debugging, along with learning resources like the NGC repository, DLI courses, and community events.

Related jobs
Jobs that call for the skills explored in this talk.

Featured Partners

Related Articles

View all articles
DN
Dr. Andy R. Terrel - NVIDIA
What’s the latest in NVIDIA CUDA Python
Python and NVIDIA CUDA have long been friends. Over the last year, NVIDIA teams are working to improve the Pythonista’s experience. This means a top-to-bottom update to the CUDA Platform is fueling the GenAI movement, e.g. llama3, gpt and nemo. These...
What’s the latest in NVIDIA CUDA Python
DC
Daniel Cranney
The State of WebDev AI 2025 Results: What Can We Learn?
Introduction The 2025 edition of The State of WebDev AI offers a detailed snapshot of how developers are using AI today, which tools have gained the most traction over the past year, and what these trends suggest about the future of the industry. In...
The State of WebDev AI 2025 Results: What Can We Learn?
DC
Daniel Cranney
Dev Digest 157: CUDA in Python, Gemini Code Assist and Back-dooring LLMs
Inside last week’s Dev Digest 157 . 🕹️ Pong in 240 browser tabs 👩‍💻 Gemini Code Assist free for 180k code completions a month 📰 AI is bad at coding and summarising the news 🕵️ Private GitHub repos show up in AI chats 🐍 CUDA for Python developers 🖥️ ...
Dev Digest 157: CUDA in Python, Gemini Code Assist and Back-dooring LLMs

From learning to earning

Jobs that call for the skills explored in this talk.