HP Labs cloud-computing test bed: VideoToon demo

The Blue Marble Image courtesy of NASA Goddard Space Flight Center

 

The HP, Intel and Yahoo! Cloud Computing Research Test Bed will provide efficient and powerful hardware platforms managed by flexible and scalable system services that support a variety of application domains. Its main objective is to support researchers who are developing new ways of managing data centers and experimenting with new cloud services.

To make this concrete, we showcase on this page an example of one such experiment: testing out a new combination of some unique HP, Yahoo!, and Intel® technologies working together to build a cloud-computing service.

In this example, Thomas Sandholm wants to make the HPL VideoToon processing technology available as a service to cartoon-ize videos. To do so he combined VideoToon with Intel® VT Virtualization Technology to provide efficient performance isolation, HP Labs Tycoon to do agile market-based allocation, and Yahoo! Apache Hadoop to simplify parallelization. The entire test was pulled together in less than 2 weeks. It's now ready for trying out at scale, to explore the effects of multiple users competing for resources.

The goal of the testbed is to make this kind of experimentation equally easy for many other users.

 

Demo Video



VideoToon Cloud Computing Test Bed Demo @ Yahoo! Video
VideoToon Cloud Computing Test Bed Demo @ YouTube

 

Technologies

The technologies used on top of the testbed in the video include:

  • Intel® Virtualization Technology (VT) provides greater flexibility and maximum system utilization by consolidating multiple environments into a single server, workstation, or PC.

  • HP Labs Tycoon is a technique to dynamically, securely, and reliably allocate virtual and physical computing resources on an on-demand basis.

  • HP Labs VideoToon is a video processing technique based on a non-photorealistic rendering algorithm developed for use in the HP R-series digital cameras.

  • Yahoo! Apache Hadoop is an open-source framework for running applications on large clusters that implements a computational paradigm named Map/Reduce.

Video transcript

»View the transcript of the VideoToon demo

 

Credits

Thomas Sandholm: himself

Directed by: Brett Bausk

Written by: Thomas Sandholm

Edited by: Brett Bausk

Produced by: Kevin Lai and Thomas Sandholm

Demo Programming: Thomas Sandholm

VideoToon Programming: Nathan Moroney and Pere Obrador

Tycoon Programming: Kevin Lai and Thomas Sandholm

 

 

The Blue Marble Image courtesy of:
NASA Goddard Space Flight Center Image by Reto Stöckli (land surface, shallow water, clouds). Enhancements by Robert Simmon (ocean color, compositing, 3D globes, animation). Data and technical support: MODIS Land Group; MODIS Science Data Support Team; MODIS Atmosphere Group; MODIS Ocean Group Additional data: USGS EROS Data Center (topography); USGS Terrestrial Remote Sensing Flagstaff Field Center (Antarctica); Defense Meteorological Satellite Program (city lights).