Name
Abstract | ||
|---|---|---|
Maximizing benefits from a cloud cluster with minimum computational costs is challenging. An accurate prediction to cloud workload is important to maximize resources usage in the cloud environment. In this paper, we propose an approach using recurrent neural networks (RNN) to realize workload prediction, where CPU and RAM metrics are used to evaluate the performance of the proposed approach. In order to obtain optimized parameter set, an orthogonal experimental design is conducted to find the most influential parameters in RNN. The experiments with Google Cloud Trace data set shows that the RNN based approach can achieve high accuracy of workload prediction, which lays a good foundation for optimizing the running of a cloud computing environment. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1109/IIKI.2016.39 | 2016 International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI) |
Keywords | Field | DocType |
Cloud cluster,Recurrent neural networks,Workload prediction | Workload prediction,Data mining,Task analysis,Workload,Computer science,Recurrent neural network,Computer network,Cloud computing | Conference |
ISBN | Citations | PageRank |
978-1-5090-5953-9 | 1 | 0.35 |
References | Authors | |
8 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Weishan Zhang | 1 | 396 | 52.57 |
| Bo Li | 2 | 971 | 111.71 |
| Dehai Zhao | 3 | 27 | 3.72 |
| Faming Gong | 4 | 22 | 5.90 |
| Qinghua Lu | 5 | 145 | 18.63 |