Jove
Visualize
Contact Us

Related Experiment Videos

Garuda: a scalable tiled display wall using commodity PCs.

Nirnimesh1, Pawan Harish, P J Narayanan

  • 1Center for Visual Information Technology, International Institute of Information Technology, Gachibowli, Hyderabad, India. nirnimesh@research.iiit.ac.in

IEEE Transactions on Visualization and Computer Graphics
|August 23, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

"Visualization matters" - stereoscopic visualization of 3D graphic neuroanatomic models through AnaVu enhances basic recall and radiologic anatomy learning when compared with monoscopy.

BMC medical education·2024
Same author

Designing perspectively correct multiplanar displays.

IEEE transactions on visualization and computer graphics·2012
Same author

Raytracing Dynamic Scenes on the GPU Using Grids.

IEEE transactions on visualization and computer graphics·2011
Same author

Real-time ray tracing of implicit surfaces on the GPU.

IEEE transactions on visualization and computer graphics·2010
Same author

Virtualized reality: perspectives on 4D digitization of dynamic events.

IEEE computer graphics and applications·2007
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Garuda is a new framework for scalable tiled display walls. It efficiently renders large datasets by transmitting only necessary scene data, reducing network load and enabling interactive visualization of massive models.

Area of Science:

  • Computer Graphics
  • High-Performance Computing
  • Visualization Systems

Background:

  • Tiled display walls offer scalable, high-resolution visualization but require efficient software to handle large datasets.
  • Existing solutions like Chromium face network bottlenecks due to transmitting massive geometric data every frame.

Purpose of the Study:

  • To develop a scalable client-server framework, Garuda, for rendering applications on tiled display walls without modification.
  • To address the network and rendering challenges associated with large-scale, high-resolution visualizations.

Main Methods:

  • Garuda employs an object-based scene graph on the server, using an adaptive culling algorithm to identify visible objects for each display tile.
  • It transmits only relevant scene graph portions to clients, which cache data to exploit inter-frame redundancy.

Related Experiment Videos

  • A multicast protocol and geometry push philosophy optimize data transmission and client synchronization.

    • Main Results:

    • Garuda achieves transparent rendering of Open Scene Graph (OSG) applications on tiled displays.
    • The system demonstrates scalability for large tile configurations and massive datasets.
    • Server and network loads exhibit sublinear growth with an increasing number of tiles, proving efficiency.

    Conclusions:

    • Garuda provides a scalable and efficient solution for driving large tiled display walls.
    • The framework enables interactive rendering of massive models by optimizing data transmission and reducing rendering overhead.
    • Its client-server architecture and novel culling/transmission methods make it suitable for next-generation large-scale visualization systems.