IP-Geolocation Mapping for Involving Moderately-Connected Internet Regions

Current IP-geolocation mapping schemes [10], [12], [13], [14] primarily take delay-measurement approach, and most of them are based on the assumption of a strong correlation between networking delay and geographical distance between the targeted client and the landmarks. In this paper, however, we investigate a large region of the Internet and find the delaydistance correlation is weak. We call this kind of Internet regions moderately-connected Internet regions. But we discover a more probable rule — with high probability the shortest delay comes from the closest distance. Based on this closest-shortest rule, we propose a simple and novel IP-geolocation mapping scheme for involving moderatelyconnected Internet regions, called GeoGet. In GeoGet, we take a large number of Web servers as passive landmarks and map a targeted client to the geolocation of the landmark that has the shortest delay. We further use JavaScript at targeted clients to generate HTTP/Get probing for delay measurement. To control the measurement cost, we adopt a multi-step probing method to refine the geolocation of a targeted client, finally to city level. We have implemented GeoGet, and the evaluation results show that when probing about 100 landmarks, GeoGet correctly maps 35.4% clients to city level, which outperforms current schemes such as GeoLim [12] and GeoPing [10] by 270% and 239%, respectively; and the median error distance in GeoGet is around 120km, outperforming GeoLim and GeoPing by 37% and 70%, respectively.