The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XXXIX-B4
https://doi.org/10.5194/isprsarchives-XXXIX-B4-261-2012
https://doi.org/10.5194/isprsarchives-XXXIX-B4-261-2012
31 Jul 2012
 | 31 Jul 2012

HIERARCHICAL OPTIMIZATION MODEL ON GEONETWORK

Z. Zha, J. Jiang, and X. Zhou

Keywords: GeoNetwork, HOM, Performance, Metadata; Web based; Performance

Abstract. In existing construction experience of Spatial Data Infrastructure (SDI), GeoNetwork, as the geographical information integrated solution, is an effective way of building SDI. During GeoNetwork serving as an internet application, several shortcomings are exposed. The first one is that the time consuming of data loading has been considerately increasing with the growth of metadata count. Consequently, the efficiency of query and search service becomes lower. Another problem is that stability and robustness are both ruined since huge amount of metadata. The final flaw is that the requirements of multi-user concurrent accessing based on massive data are not effectively satisfied on the internet. A novel approach, Hierarchical Optimization Model (HOM), is presented to solve the incapability of GeoNetwork working with massive data in this paper. HOM optimizes the GeoNetwork from these aspects: internal procedure, external deployment strategies, etc. This model builds an efficient index for accessing huge metadata and supporting concurrent processes. In this way, the services based on GeoNetwork can maintain stable while running massive metadata. As an experiment, we deployed more than 30 GeoNetwork nodes, and harvest nearly 1.1 million metadata. From the contrast between the HOM-improved software and the original one, the model makes indexing and retrieval processes more quickly and keeps the speed stable on metadata amount increasing. It also shows stable on multi-user concurrent accessing to system services, the experiment achieved good results and proved that our optimization model is efficient and reliable.