Abstract
Web service composition is an important research of service computing. This paper aims to improve the web service dependency graph to raise the accuracy of service selection, and presents a parallel service composition algorithm based on MapReduce to enhance the efficiency of service composition, which addresses the current limitations of inaccuracy and inefficiency in massive service composition. Experimental results show that the strategies can acquire the feasible scheme of web service composition based on QoS and reduce the composition time by 68% on average in large-scale collection of web services to meet users' demands correctly and efficiently.
| Original language | English |
|---|---|
| Pages (from-to) | 667-676 |
| Number of pages | 10 |
| Journal | Journal of Internet Technology |
| Volume | 18 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - 2017 |
Keywords
- Parallel
- Web service composition
- Web service dependency graph