It is often necessary to maintain a data structure, for example a search tree, for a set of computing nodes, which are interconnected dynamically and spontaneously. Instead of relying on a centralized "official" server to maintain the data structure, a grass-root and spontaneous approach is to distribute data structure to the participating nodes by taking advantages of their distributed resources. This study shows the feasibility of such an approach by designing a distributed search tree structure, called Pyramid, to operate atop end systems without relying on a centralized server. Pyramid allows the participating nodes to manipulate the data structure transparently through high-level operations such as search, insert, delete, update and query. Pyramid is designed based on the peer-to-peer model. Its self-configuration and -healing features enable the manipulation of information in an unreliable and unstable environment. A wireless application based on a prototype of Pyramid is presented.
|頁（從 - 到）||1248-1256|
|期刊||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|出版狀態||Published - 2004|
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)