GODIVA: Lightweight Data Management for Scientific Visualization Applications

  • Authors:
  • Xiaosong Ma;Marianne Winslett;John Norris;Xiangmin Jiao;Robert Fiedler

  • Affiliations:
  • -;-;-;-;-

  • Venue:
  • ICDE '04 Proceedings of the 20th International Conference on Data Engineering
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Scientific visualization applications are very data-intensive,with high demands for I/O and data management.Developers of many visualization tools hesitate to use traditionalDBMSs, due to the lack of support for these DBMSson parallel platforms and the risk of reducing the portabilityof their tools and the user data. In this paper, we proposethe GODIVA framework, which provides simple database-likeinterfaces to help visualization tool developers managetheir in-memory data, and I/O optimizations such asprefetching and caching to improve input performance atrun time. We implemented the GODIVA interfaces in astand-alone, portable user library, which can be used by alltypes of visualization codes: interactive and batch-mode,sequential and parallel. Performance results from runninga visualization tool using the GODIVA library on multipleplatforms show that the GODIVA framework is easy to use,alleviates developers' data management burden, and canbring substantial I/O performance improvement.