The distributed hash table (DHT) mechanisms have been proposed to manage data in very large, structured peer-to-peer (P2P) systems. DHT algorithms provide efficient exact-match object search capabilities without requiring global indexing and are hence extremely scalable. However, range queries - which are very common with spatial data - cannot be executed efficiently in these systems because the adoption of uniform hash functions to ensure excellent load balancing unfortunately destroys any existing data locality. In this report we propose a novel technique to preserve spatial locality information while also keeping some of the load balancing properties of DHT based systems. We describe our design as an extension of content-addressable networks (CAN) and illustrate the feasibility of supporting spatial range queries.
[1]
Mark Handley,et al.
A scalable content-addressable network
,
2001,
SIGCOMM '01.
[2]
Jon Louis Bentley,et al.
Quad trees a data structure for retrieval on composite keys
,
1974,
Acta Informatica.
[3]
Aaron Harwood,et al.
Hashing Spatial Content over Peer-to-Peer Networks
,
2003
.
[4]
Peter Steenkiste,et al.
Location-based node IDs : enabling explicit locality in DHTs
,
2003
.
[5]
David R. Karger,et al.
Chord: A scalable peer-to-peer lookup service for internet applications
,
2001,
SIGCOMM '01.
[6]
Antony I. T. Rowstron,et al.
Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems
,
2001,
Middleware.