Abstract
Cloud architectures have become increasing common in the IT industry and academic circle. However, most cloud architectures only focus on availability but ignore economic effectiveness. Based on Eucalyptus, this paper proposes an effective way to balance high availability and economic profits. We designed a forecast mechanism using three new modules: the forecast module, adjustment module, and collection module. The forecast module uses a set of machine learning techniques to improve forecast accuracy. We carried out a comparative experiment and the experimental results prove the efficiency of the proposed forecast mechanism. Copyright © 2016 IEEE.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2016 IEEE International Conference on Progress in Informatics and Computing |
| Place of Publication | Beijing |
| Publisher | Institute of Electrical and Electronics Engineers, Inc |
| Pages | 508-512 |
| ISBN (Print) | 9781509034840, 9781509034833 |
| Publication status | Published - 2016 |
Citation
Lyu, H., Li. P., Yan, R., Masood, A., Sheng, B., & Luo, Y. (2016). Load forecast of resource scheduler in cloud architecture. In Y. Wang, & Y. Sun (Eds.), Proceedings of the 2016 IEEE International Conference on Progress in Informatics and Computing (pp. 508-512). Beijing: Institute of Electrical and Electronics Engineers, Inc.Keywords
- Forecast mechanism
- Virtual machine
- Cloud architecture
- Machine learning