Domain Adversarial Adaptive Short-Term Workload Forecasting Model

doi:10.3778/j.issn.1002-8331.2211-0036

Abstract

Abstract: The accuracy of workload prediction is one of the main factors affecting the elastic resource management of cloud platforms. And there are a large number of short task workload sequences with insufficient historical information and unsmooth characteristics in the cloud, which makes it difficult to select appropriate models for accurate prediction. In this paper, a domain adversarial workload prediction model is proposed. The model uses SSA（singular spectrum analysis） to smooth the workload and solve the problem of irregularity. Similarity calculations are performed by combining MASS_V4（the fourth version of Mueen’s algorithm for similarity search） with temporal features to obtain suitable source-domain data-assisted migration prediction. The GRU（gated recurrent unit） is used as the reference to construct the network, a new loss function defined is with Y-discrepancy, and a prediction model is constructed with strong short-workload feature representation ability in the adversarial process. The proposed method is compared with other commonly used cloud workload prediction algorithms on two real cloud platform datasets and both show higher prediction accuracy.

Key words: cloud computing, workload prediction, domain antagonism transfer learning, the fourth version of Mueen’s algorithm for similarity search（MASS_V4）

摘要： 负载预测的精度是影响云平台弹性资源管理的主要因素之一。而云平台中存在着大量的短任务负载序列，其历史信息不足和不平滑的特性导致难以选择合适的模型进行精准预测。对此提出了一种领域对抗自适应的短任务负载预测模型。该模型采用奇异谱分析（singular spectrum analysis，SSA）对样本进行平滑处理；联合第四版本的Mueen相似度搜索算法（the fourth version of Mueen’s algorithm for similarity search，MASS_V4）与时间特征进行域间相似性计算，获得合适的源域数据来辅助迁移预测；将门控循环单元（gated recurrent unit，GRU）作为基准器构建网络，并利用Y差异定义新的损失函数，通过对抗过程建立出表征能力强的短任务负载预测模型。将所提方法在两个真实的云平台数据集上与其他常用的云负载预测算法对比，均表现出较高的预测精度。

关键词: 云计算, 负载预测, 域对抗迁移学习, MASS_V4

LIU Chunhong, JIAO Jie, WANG Jingxiong, LI Weili, ZHANG Junna. Domain Adversarial Adaptive Short-Term Workload Forecasting Model[J]. Computer Engineering and Applications, 2023, 59(24): 289-297.

刘春红, 焦洁, 王敬雄, 李为丽, 张俊娜. 领域对抗自适应的短任务负载预测模型[J]. 计算机工程与应用, 2023, 59(24): 289-297.

References

[1] NALINI S，ANDREWS J.Recent security challenges incloud computing[J].Computers & Electrical Engineering，2018，71：28-42.
[2] AL Y，PARAISO F，DJARALLAH N，et al.Elasticity in cloud computing：state of the art and research challenges[J].IEEE Transactions on Services Computing，2017，11（2）：430-447.
[3] MASDARI M，KHOSHNEVIS A.A survey and classification of the workload forecasting methods in cloud computing[J].Cluster Computing，2020，23（4）：2399-2424.
[4] GAO J，WANG H，SHEN H.Machine learning based work load prediction in cloud computing[C]//International Conference on Computer Communications and Networks，August 3-6，2020.Honolulu，HI：IEEE，2020：1-9.
[5] REISS C，TUMANOV A，GANGER G R，et al.Heterogeneity and dynamicity of clouds at scale：Google trace analysis[C]//Proceedings of the Third ACM Symposium on Cloud Computing，New York，USA，2012：1-13.
[6] KUMAR J，SINGH A K.Workload prediction in cloud using artificial neural network and adaptive differential evolution[J].Future Generation Computer Systems，2018，81：41-52.
[7] HU H L，TANG M J，BAI C C.DATSING：data augmented time series forecasting with adversarial domain adaptation[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management，New York，NY，USA，2020：2061-2064.
[8] LAPTEV M，YU J F，RAJAGOPAL R.Reconstruction and regression loss for time-series transfer learning[C]//Proceedings of the Special Interest Group on Knowledge Discovery and Data Mining（SIGKDD） and the 4th Workshop on the Mining and Learning from Time Series（MiLeTS），London，UK，2018：1-8.
[9] MATHELIN A，RICHARD G，DEHEEGER F，et al.Adversarial weighting for domain adaptation in regression[C]//2021 IEEE 33rd International Conference on Tools with Artificial Intelligence（ICTAI），Washington，DC，USA，2021：49-56.
[10] RAGAB M，CHEN Z H，WU M，et al.Contrastive adversarial domain adaptation for machine remaining useful life prediction[J].IEEE Transactions on Industrial Informatics，2020，17（8）：5239-5249.
[11] HE Q Q，PANG C I，SI Y W.Multi-source transfer learning with ensemble for financial time series forecasting[C]//2020 IEEE/WIC/ACM International Joint Conference on WI-IAT，Australia，2020：227-233.
[12] ABDULLAH M，SHENG Z，YAN Z，et al.The fastest similarity search algorithm for time series subsequences under Euclidean distance[EB/OL].[2022-08].http：//www.cs.unm.edu/~mueen/FastestSimilaitySearch.html.
[13] 李国芳，王力，龙飞.基于奇异谱分析的高斯噪声降噪改进算法[J].计算机工程与设计，2016，37（8）：2143-2150.
LI G F，WANG L，LONG F.Gaussian noise de-noising optimal algorithm based on singular spectrum analysis[J].Computer Engineering and design，2016，37（8）：2143-2150.
[14] GRUNERBEL L，HENIRICH F，DIEBOLDER D，et al.Wearable decubitus prophylaxis tool based on machine learning methods[C]//2022 IEEE International Conference on PerCom Workshops.Pisa，Italy：IEEE，2022：730-734.
[15] MOHRI M，MEDINA M A.New analysis and algorithm for learning with drifting distributions[C]//23rd International Conference on Algorithmic Learning Theory，Lyon，France，October 29-31，2012：124-138.
[16] REISS C，WILKES J，HELLERSTEIN J L.Google cluster-usage traces：format+ schema[J].White Paper，2011：1-14.
[17] GUO J，CHANG Z H，WANG S，et al.Who limits the resource efficiency of my datacenter：an analysis of alibaba datacenter traces[C]//2019 IEEE/ACM 27th International Symposium on Quality of Service（IWQoS），Phoenix，AZ，United States，2019：1-10.
[18] XU X，MENG Z.A hybrid transfer learning model for short-term electric load forecasting[J].Electrical Engineering，2020，102（3）：1371-1381.
[19] 方义秋，卢壮，葛君伟.联合RMSE损失LSTM-CNN模型的股价预测[J].计算机工程与应用，2022，58（9）：294-302.
FANG Y Q，LU Z，GE J W.Forecasting stock prices with combined RMSE loss LSTM-CNN model[J].Computer Engineering and Applications，2022，58（9）：294-302.
[20] SHI Q Q，YIN J M，CAI J J，et al.Block hankel tensor ARIMA for multiple short time series forecasting[C]//Proceedings of the AAAI Conference on Artificial Intelligence，New York City，United States，2020：5758-5766.
[21] KHURANA A S，SHARMA G，MIGLANI N，et al.An intelligent fine-tuned forecasting technique for Covid-19 prediction using neuralprophet model[J].Computers Materials & Continua，2022，71（1）：629-649.
[22] 张忠林，张艳.改进FA优化LSTM的时序预测模型[J].计算机工程与应用，2022，58（11）：125-132.
ZHANG Z L，ZHANG Y.Improved FA optimizing LSTM time series prediction model[J].Computer Engineering and Applications，2022，58（11）：125-132.