Proceedings of the ACM SIGSPATIAL Second International Workshop on High Performance and Distributed Geographic Information Systems

Abstract

High performance computing and distributed systems have become prominent elements in the landscape of computing and information technologies. High performance and distributed GIS (HPDGIS) have emerged as a growing area of theoretical and applied research. This growth is driven by geospatial problems in numerous fields that are increasingly computationally intensive and require collaboration support. Efficient handling of massive spatial databases, shared and role-based access to distributed data, and high end computing services are fundamental to the near-real-time response times required for many GIS and associated decision support applications. The initial feasibility and tremendous potential of HPDGIS have recently been demonstrated by exploiting rapidly developing cyberinfrastructure capabilities. It is therefore important to bring together researchers and practitioners to map out fundamental research areas centered on HPDGIS and its tight connections to advances in high performance computing, distributed systems, and associated GIS and spatial analysis applications and this inaugural ACM SIGSPATIAL HPDGIS 2010 is designed to do just that. This proceeding contains papers selected for publication and presentation, to this HPDGIS'10 workshop, held at San Jose, California, USA on November 2, 2010 in conjunction with the 18th ACM SIGSPATIAL International Conference on Advances in Geographic Information System. This year's program also features an outstanding keynote talk titled as "High Performance Computing with Spatial Datasets" by Dr. Shashi Shekhar from the University of Minnesota. The workshop attracted research papers on a number of HPDGIS themes including data intensive GIS, parallel processing algorithms for GIS problems, GIS based on cloud computing, service-oriented GIS, and spatial middleware. Research papers: "A MapReduce Approach to Gi*(d) Spatial Statistic" and "Spatial Scene Similarity Assessment on Hadoop" illustrate the use of the map-reduce framework to resolve two typical data-intensive problems in GIS and spatial analysis. In "Towards Personal High- Performance Geospatial Computing (HPC-G): Perspectives and a Case Study", the author advocates the use of a low cost personal HPDGIS environment developed by using parallel computing capability afforded by Graphic Processing Unit architecture. A theoretical framework for modeling the cost of a distributed service on cloud is discussed in "A Cost Model for Distributed Coverage Processing Services", while the paper titled as "Cloud Computing for Geosciences: Deployment of GEOSS Clearinghouse on Amazon's EC2" experimentally demonstrates the use of cloud computing for GIS and spatial analysis. "High Performance Computing: Fundamental Research Challenges in Service Oriented GIS" identifies a set of fundamental research challenges for the realization of service-oriented GIS. "A Distributed Resource Broker for Spatial Middleware Using Adaptive Space-Filling Curve" presents a spatial middleware component to enable HPDGIS applications by exploiting computational capabilities of cyberinfrastructure.