Interactive GPU volume raycasting in a clustered graphics environment

Christian Noon; Eric Foo; Eliot Winer

doi:10.1117/12.911646

17 February 2012 Interactive GPU volume raycasting in a clustered graphics environment

Christian Noon, Eric Foo, Eliot Winer

Proceedings Volume 8316, Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling; 83161L (2012) https://doi.org/10.1117/12.911646
Event: SPIE Medical Imaging, 2012, San Diego, California, United States

Abstract

This research focuses on performing interactive, real-time volume raycasting in a large clustered graphics environment using custom GPU shaders for composite volume raycasting with trilinear interpolation. Working in this type of environment presents unique challenges due to the distributed nature, and inherently required synchronization of data and operations across the cluster. Invoking custom vertex and fragment shaders in a non-thread-safe manner becomes increasingly complex in a large clustered graphics environment. Through use of an abstraction layer, all rendering contexts are split-up with no changes to the volume raycasting core. Therefore, the volume raycasting core is completely transparent from the computing platform. The application was tested on a 6-wall immersive VR system with 96 graphics contexts coming from 48 cluster nodes. Interactive framerates of 60 frames per second were produced on 512x512x100 volumes, and an average of 30 frames per second for a 512x512x1000 volume. The use of custom configuration files allows the same code to be highly scalable from a single screen VR system to a fully immersive 6-sided wall VR system. Through the code abstraction, the same volume raycasting core can be implemented on any type of computing platform including desktop and mobile.

Citation Download Citation

Christian Noon, Eric Foo, and Eliot Winer "Interactive GPU volume raycasting in a clustered graphics environment", Proc. SPIE 8316, Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling, 83161L (17 February 2012); https://doi.org/10.1117/12.911646

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available