• Subscribe

iSGTW Link of the week - Kung Fu Panda digs grids

Link of the week - Kung Fu Panda digs grids

Using grid computing, digital artists working on Kung Fu Panda are performing 75,000 to 100,000 batch-processing jobs each night, each ranging from a few minutes per frame to 15 hours per frame, depending on the rendering complexity of the digital frame sequence.
Image courtesy of Dreamworks

Grid computing isn't always about saving the world. It can be about creating a new world, in this case, the world of the Kung Fu Panda, a new animation feature by Dream Works to be released in June.

Creating animation films like Kung Fu Panda typically requires around 25 million render hours from start to finish. The Dream Works computer farm comprises a 4000-processor Linux cluster, but when their digital artists are all simultaneously pinging the compute cluster to retrieve digital frame data, the system's input-output comes under a lot of strain.

Enter grid computing, in the form of Ibrix parallel file server software. DreamWorks added Ibrix to its animation system about six months ago, aiming to accelerate their high-resolution lighting and rendering applications by up to five times.

Ed Leonard, DreamWorks chief technology officer, said that he expects the grid technology to become a fixture for all future DreamWorks Animation films. The studio releases at least two films every year and is generally developing 10 different animation features at a time.

Scenes that once took two hours to create are now completed in seconds, Leonard said. "That's pretty remarkable, and we're scratching the surface on where we are going with this," he said.

Join the conversation

Do you have story ideas or something to contribute? Let us know!

Copyright © 2021 Science Node ™  |  Privacy Notice  |  Sitemap

Disclaimer: While Science Node ™ does its best to provide complete and up-to-date information, it does not warrant that the information is error-free and disclaims all liability with respect to results from the use of the information.


We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit ScienceNode.org — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on ScienceNode.org” containing a link back to the original article.
  4. The easiest way to get the article on your site is to embed the code below.