General-purpose computing on graphics processing units

Did you know your computer can do more than just compute numbers?

Imagine you're trying to edit a massive photo collection quickly. You want to apply filters to thousands of photos simultaneously.

By using a GPU, which is great at handling lots of pictures at once, you can apply filters to many photos at the same time instead of one by one, making the process much faster.

Example

Instead of taking 10 minutes to apply a filter to one photo, you can do it for 100 photos in just 1 minute using a GPU.

Remember this

The key insight is that GPUs can process many tasks in parallel, significantly speeding up tasks that involve lots of repetitive operations.

Text adapted from Wikipedia, licensed under CC BY-SA 4.0.

Related concepts

Loop nest optimization

Can speeding up your computer make tasks quicker?

Matrix multiplication algorithm

Ever wondered how computers speed up multiplying huge numbers?

tensor cores are

Why can computers crunch numbers faster than humans?

gradient accumulation simulates larger batch sizes without more memory

Can you train a machine like you do with a computer?

batch size affects generalization: larger batches find sharper minima

Larger batch sizes lead to sharper minima, enhancing generalization by providing more accurate gradient estimates

Overlapping subproblems

Ever calculated a huge Fibonacci sequence by hand?

Swipe through 100 ML concepts daily

Open Pocket Polymath