Prediction Model of Wavelet Neural Network for Hybrid Storage System
The Hybrid storage system needs to distinguish the data state to manage data migration. The frequently data may be placed solid-state hard disk to improve the accessing performance. Here a novel prediction model of the frequently accessing data that is called hot access data is proposed. This model extracts the workload features and is built based on the wavelet neural network to identify the data state. The prediction model is trained by the sampling data from historical workloads and can be applied in the hybrid storage system. The experimental results show that the proposed model has better accuracy and faster learning speed than BP neural network model. Additionally, it has better independent on training data and generalization ability to adapt to various storage workloads.
KeywordsHybrid storage system Wavelet Neural Network Prediction model Data migration
This project is supported by Shandong Provincial Natural Science Foundation, China (No. ZR2017MF050), Project of Shandong Province Higher Educational Science and technology program (No. J17KA049), Shandong Province Key Research and Development Program of China (Nos. 2018GGX101005, 2017CXGC0701, 2016GGX109001) and Shandong Province Independent Innovation and Achievement Transformation, China (No. 2014ZZCX02702).
- 2.Huang, D.M., Du, Y.L., He, Q.: Research on ocean large data migration algorithm in hybrid cloud storage. J. Comput. Res. Dev. 51(1), 199–205 (2014). (In Chinese)Google Scholar
- 4.Kgil, T., Roberts, D., Mudge, T.: Improving NAND flash based disk caches. In: Proceedings of the 35th International Symposium on Computer Architecture, pp, 327–338, Beijing, China (2008)Google Scholar
- 5.Koltsidas, I., Viglas, S. D.: Designing a flash-aware two-level cache. In: Proceedings of the 15th International Conference on Advances in Databases and Information Systems, pp, 153–169, Berlin, Heidelberg (2011)Google Scholar
- 7.Guo, Y.C., Wang, L.H.: Hybrid wavelet neural network blind equalization algorithm controlled by fuzzy neural network. ACTA Electronica Sin. 39(4), 975–980 (2011). (In Chinese)Google Scholar
- 9.Do, H.T., Zhang, X., Nguyen, N.V., Li, S.S., Chu, T.T.: Passive-Islanding detection method using the wavelet packet transform in grid-connected photovoltaic system. IEEE Trans. Power Electron. 31(10), 6955–6967 (2016)Google Scholar