Distributed P2P Computing within Triana: A Galaxy Visualization Test Case

  • Authors:
  • Ian Taylor;Matthew Shields;Ian Wang;Roger Philp

  • Affiliations:
  • -;-;-;-

  • Venue:
  • IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We discuss here a parallel implementation of the visualisation of data from a galaxy formation simulation within the Triana problem-solving environment. The visualisation is a test case for our prototype Triana Grid software using a decentralized Peer-to-Peer approach. The Triana distributed environment is middleware independent and designed in such a way as to make it easily extensible by users of the system. This novel approach is usable both within a Peer-to-Peer framework, using JXTA, and within service oriented architectures, such as OGSA, by the use of a Grid Application Toolkit (GAT) interface. Furthermore, Triana has a pluggable architecture that allows the middleware independent representations of distributed task-graphs to be written in several popular formats. This facilitates the use of the mechanism within other disparate systems. Galaxy-formation simulation codes generate data that represent a series of particles in three dimensions and their associate properties as snap shots in time. An important aspect of these types of calculation is the visualization of time sequence data.The user of the code would like to visualize this data as an animation in two dimensions with the ability to vary the point of view, project that particular two dimensional slice and re-run the animation. The recalculation of the projection can be computed in parallel resulting in a considerable speedup.