The current accuracy of cloud computing resource consumption forecasts can be improved. This study introduces an innovative resource forecasting framework to address the challenges of predicting resource usage within cloud data centers. The framework includes data preprocessing, a long short-term memory network based on an attention mechanism, and a search mechanism based on an improved genetic algorithm. The intrinsic properties of the data are preserved by the self-organizing mapping network, which helps remove noise and reveal the underlying patterns of the data. The combination of the attention mechanism and long short-term memory network enhances the model's attention to key time steps, thus enhancing the prediction ability. An improved genetic algorithm was used for weight optimization so that the model could adaptively adjust the attention weight allocation. In the experiment, the framework proposed in this study was validated based on Google cluster tracking data. Based on a comparison with the benchmark algorithm, the proposed framework has advantages in the three indices for both short-term and long-term predictions, which proves that the framework has higher accuracy. The modified Friedman test and Nemenyi test proved that the proposed method’s prediction performance is better than that of the two benchmark algorithms in terms of the root-mean-square error.
(13)式中,为权重,为偏置;(14)式中,为权重,为偏置;(15)式中,为权重,为偏置。这里,表示先前的隐藏层单元,这些隐藏层单元逐个添加三个门的权重。经过(15)式的处理,(C) t 变为当前的存储单元。(16)式显示了先前隐藏单元输出和先前存储单元单元的乘法,以tanh和sigmoid激活函数的形式在三个门添加非线性特征,其中和为权重,为偏置。(17)式为隐层单元的输出。假定和分别是之前和现在的时间步长,则:
WANGS, ZHUF, YAOY P, et al. A computing resources prediction approach based on ensemble learning for complex system simulation in cloud environment[J]. Simulation Modelling Practice and Theory, 2021, 107: 102202. DOI: 10.1016/j.simpat.2020.102202 .
[2]
VALARMATHIK, KANAGA SUBA RAJAS. Resource utilization prediction technique in cloud using knowledge based ensemble random forest with LSTM model[J]. Concurrent Engineering, 2021, 29(4): 396-404. DOI: 10.1177/1063293x211032622 .
[3]
NGODUYD. Noise-induced instability of a class of stochastic higher order continuum traffic models[J]. Transportation Research Part B: Methodological, 2021, 150: 260-278. DOI: 10.1016/j.trb.2021.06.013 .
[4]
XUM X, SONGC H, WUH M, et al. esDNN: Deep neural network based multivariate workload prediction in cloud computing environments[J]. ACM Transactions on Internet Technology,2022, 22(3): 75. DOI: 10.1145/3524114 .
[5]
ARIZAJ, JIMENOM, VILLANUEVA-POLANCOR, et al. Provisioning computational resources for cloud-based e-learning platforms using deep learning techniques[J]. IEEE Access, 2021, 9: 89798-89811. DOI: 10.1109/ACCESS.2021.3090366 .
[6]
KARIMM E, MASWOODM M S, DAS S, et al. BHyPreC: A novel Bi-LSTM based hybrid recurrent neural network model to predict the CPU workload of cloud virtual machine[J]. IEEE Access, 2021, 9: 131476-131495. DOI: 10.1109/ACCESS.2021.3113714 .
[7]
NAWROCKIP, GRZYWACZM, SNIEZYNSKIB. Adaptive resource planning for cloud-based services using machine learning[J]. Journal of Parallel and Distributed Computing, 2021, 152: 88-97. DOI: 10.1016/j.jpdc.2021.02.018 .
[8]
KUMARJ, SAXENAD, SINGHA K, et al. BiPhase adaptive learning-based neural network model for cloud datacenter workload forecasting[J]. Soft Computing, 2020, 24(19): 14593-14610. DOI: 10.1007/s00500-020-04808-9 .
[9]
NGUYENT, DO T, LEK, et al. An LSTM-based approach for predicting resource utilization in cloud computing[DB/OL].[2024-01-09]. DOI: 10.1145/3568562.3568647 .
[10]
ASHAWAM, DOUGLASO, OSAMORJ, et al. Retraction note: Improving cloud efficiency through optimized resource allocation technique for load balancing using LSTM machine learning algorithm[J]. Journal of Cloud Computing, 2023, 12(1): 171. DOI: 10.1186/s13677-023-00562-z .
[11]
DOGANIJ, KHUNJUSHF, MAHMOUDIM R, et al. Multivariate workload and resource prediction in cloud computing using CNN and GRU by attention mechanism[J]. The Journal of Supercomputing, 2023, 79(3): 3437-3470. DOI: 10.1007/s11227-022-04782-z .
[12]
KOHONENT. The self-organizing map[J]. Proceedings of the IEEE, 1990, 78(9): 1464-1480. DOI: 10.1109/5.58325 .
[13]
NIUZ Y, ZHONGG Q, YUH. A review on the attention mechanism of deep learning[J]. Neurocomputing, 2021, 452: 48-62. DOI: 10.1016/j.neucom.2021.03.091 .
[14]
DENGW, ZHANGX X, ZHOUY Q, et al. An enhanced fast non-dominated solution sorting genetic algorithm for multi-objective problems[J]. Information Sciences: An International Journal, 2022, 585(C): 441-453. DOI: 10.1016/j.ins.2021.11.052 .
[15]
REISSC, WILKESJ, HELLERSTEINJ L. Google cluster-usage traces: Format+ schema[DB/OL]. [2024-01-02].
[16]
AMVROSIADISG, PARKJ W, GANGERG R, et al. On the diversity of cluster workloads and its impact on research results[DB/OL]. [2024-01-02].
[17]
GUPTAS, DILEEPA D. Long range dependence in cloud servers: A statistical analysis based on Google workload trace[J]. Computing, 2020, 102(4): 1031-1049. DOI: 10.1007/s00607-019-00779-4 .
[18]
BIJ, LIS, YUANH T, et al. Integrated deep learning method for workload and resource prediction in cloud systems[J]. Neurocomputing, 2021, 424: 35-48. DOI: 10.1016/j.neucom.2020.11.011 .