GTC 2012 Keynote by Jen-Hsun Huang (Part 03): The NVIDIA Kepler GPU Architecture

NVidia CEO Jen-Hsun Huang speaks on GTC (GPU Technology Conference) 2012 on key differences between Fermi and Kepler. Key differences are

  1. SMX: Streaming Multiprocessor neXt generation
  2. Hyper-Q: keep GPU busy with queuing
  3. Dynamic Parallelism: GPU can now assign work to itself, e.g., run Quicksort without help from CPU

The video shows heatmaps of some GPUs, i.e., one can see graphically how much work is done on the GPU. The video is here:

Also have a look at the white-paper at NVIDIA‘s site: NVIDIA-Kepler-GK110-Architecture-Whitepaper

A similar presentation conducted by Stephen Jones, from NVIDIA: Inside the Kepler Architecture,


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.