【摘要】 随着云计算和大数据技术的快速发展，负载预测作为可优化资源分配和提高系统性能的关键任务受到了广泛关注。为了更好地对负载进行长短期时序预测和提高负载预测精度，提出了一种卷积神经网络（CNN）、长短时记忆网络（LSTM）和一次指数平滑（SES）的组合模型，并采用麻雀搜索算法（SSA）优化模型参数。经过处理的数据集分别输入到SSA-CNN-LSTM模型和SSA-SES模型中，然后对这2个模型的输出值用权重系数法进行组合得出最终的输出值。实验结果表明，本文模型在阿里云公开数据集Cluster-trace-v2018上的云计算负载预测中的性能评估值，相较于CNN-LSTM、SES、CNN-LSTM-SES、SSA-CNN-LSTM与SSA-SES模型，预测精度更高。更多还原

【Abstract】 With the rapid development of cloud computing and big data technology,load prediction has received widespread attention as a key task that can optimize resource allocation and improve system performance. In order to better predict the long-and short-term time series of loads and improve the accuracy of load prediction,this paper proposes a hybrid model of Convolutional Neural Network(CNN),Long Short-Term Memory(LSTM)network,Single Exponential Smoothing(SES)and uses Sparrow Search Algorithm(SSA)to optimize the model parameters. The processed dataset is input into the SSA-CNN-LSTM model and SSA-SES model respectively and the output values of these two models are combined using the weight coefficient method to obtain the final output value. The experimental results show that the performance evaluation values of this paper model in cloud computing load prediction on the Alibaba Cloud public dataset Cluster-trace-v2018 are higher than those of CNN-LSTM,SES,CNN-LSTM-SES,SSA-CNN-LSTM and SSA-SES models.更多还原