GeoHybrid: A hierarchical approach for accurate and scalable geographic localization

Geographic location and Grid computing are two areas that have taken off in recent years, both receiving a lot of attention from research community. The Grid Resource Brokers, which tries to find the best match between the job requirements and the resources available on the Grid, can take benefits by knowing the geographic location of clients, for a considerable improvement of their decision-taking functions. A measurement-based geolocation service estimates host locations from delay measurements taken from landmarks, which are hosts with a known geographic location, toward the host to be located. Nevertheless, active measurement can burden the network. Relying on database-driven geolocation and active measurements, we propose GeoHybrid. GeoHybrid estimates the geographic location of Internet hosts with low overhead as well better accuracy with respect to geolocation databases. Afterwards, we propose a geolocation middleware for grid computing. By defining the architecture and the methods of this service, we show that a promising symbiosis may be envisaged by the use of the proposed middleware service for grid computing.