Abstract
Request-based Object Consistency Algorithm for Content Distribution Network (CDN)-ROCA was proposed to download modified object from web server by polling web server in Time-to-Refresh (TTR) interval. In order to make refresh interval more accurate, Object Fuzzy Clustering Algorithm (OFCA) was introduced to find commonness among objects and to optimize ROCA. OFCA adopted concept hierarchy to describe object name property, as well as heuristic method to find initial class mode. Trace-driven experiments show that the TTR value obtained by ROCA is according to object modification, and fuzzy clustering algorithm can improve ROCA remarkably.
| Original language | English |
|---|---|
| Pages (from-to) | 1480-1483 |
| Number of pages | 4 |
| Journal | Tien Tzu Hsueh Pao/Acta Electronica Sinica |
| Volume | 33 |
| Issue number | 8 |
| State | Published - Aug 2005 |
Keywords
- Concept hierarchy
- Content distribution network (CDN)
- Fuzzy clustering algorithm
- Object consistency