Personal Workspace for Large-Scale Data-Driven Computational Experiment

  • Authors:
  • Yiming Sun;Scott Jensen;Sangmi Pallickara;Beth Plale

  • Affiliations:
  • Computer Science Department, Indiana University, Lindley Hall 215, 150 S. Woodlawn Ave., Bloomington, IN 47405, USA. yimsun@cs.indiana.edu;Computer Science Department, Indiana University, Lindley Hall 215, 150 S. Woodlawn Ave., Bloomington, IN 47405, USA. scjensen@cs.indiana.edu;Computer Science Department, Indiana University, Lindley Hall 215, 150 S. Woodlawn Ave., Bloomington, IN 47405, USA. leesangm@cs.indiana.edu;Computer Science Department, Indiana University, Lindley Hall 215, 150 S. Woodlawn Ave., Bloomington, IN 47405, USA. plale@cs.indiana.edu

  • Venue:
  • GRID '06 Proceedings of the 7th IEEE/ACM International Conference on Grid Computing
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

As the scale and complexity of data-driven computational science grows, so grows the burden on the scientists and students in managing the data products used and generated during experiments. Products must be moved and directories created. Search support in traditional file systems is arcane. While storage management tools can store rich metadata, these tools do not satisfy the nuances of the individual computational science researcher working alone or cooperatively. We have developed a personal workspace tool, myLEAD, that actively manages metadata and data products for users. Inspired by the Globus MCS metadata catalog and layered on top of the UK e-Science OGSA-DAI tool, myLEAD provides capture, storage and search tools to the computational scientist. In this paper we experimentally evaluate the performance of the myLEAD metadata catalog.