arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2307.15941cs.LG
收藏

预测性自动缩放中的持续学习

Continual Learning in Predictive Autoscaling

Ant Group(蚂蚁集团)

Hongyan Hao, Zhixuan Chu, Shiyi Zhu, Gangwei Jiang, Yan Wang, Caigao Jiang, James Zhang, Wei Jiang, Siqiao Xue, Jun Zhou

AI总结:

针对预测性自动缩放在异常流量下性能下降的问题,提出基于密度记忆选择与提示网络学习的持续学习方法DMSHM,仅用少量历史日志实现高精度预测,并在公共与工业数据集上超越现有方法。

摘要:

预测性自动缩放用于预测服务器的工作负载并提前准备资源,以确保动态云环境中的服务级别目标(SLOs)。然而,在实践中,其预测任务常常在由外部事件(如促销活动和应用程序重新配置)引起的异常流量下遭受性能下降,对此常见的解决方案是使用长时间历史数据重新训练模型,但这以高昂的计算和存储成本为代价。为了更好地解决这一问题,我们提出了一种基于回放的持续学习方法,即基于密度的记忆选择与基于提示的网络学习模型(DMSHM),仅使用一小部分历史日志即可实现准确预测。首先,我们发现在预测任务中应用基于回放的持续学习时存在样本重叠现象。为了克服这一挑战并有效整合新样本分布,我们提出了一种基于密度的样本选择策略,该策略利用核密度估计计算样本密度作为参考来计算样本权重,并采用权重采样来构建新的记忆集。然后,我们基于提示表示实现基于提示的网络学习以优化参数。最后,我们在公共和工业数据集上进行了实验,以证明我们提出的方法在记忆容量和预测准确性方面优于最先进的持续学习方法。此外,我们展示了DMSHM在实际工业应用中的显著实用性。

英文摘要:

Predictive Autoscaling is used to forecast the workloads of servers and prepare the resources in advance to ensure service level objectives (SLOs) in dynamic cloud environments. However, in practice, its prediction task often suffers from performance degradation under abnormal traffics caused by external events (such as sales promotional activities and applications re-configurations), for which a common solution is to re-train the model with data of a long historical period, but at the expense of high computational and storage costs. To better address this problem, we propose a replay-based continual learning method, i.e., Density-based Memory Selection and Hint-based Network Learning Model (DMSHM), using only a small part of the historical log to achieve accurate predictions. First, we discover the phenomenon of sample overlap when applying replay-based continual learning in prediction tasks. In order to surmount this challenge and effectively integrate new sample distribution, we propose a density-based sample selection strategy that utilizes kernel density estimation to calculate sample density as a reference to compute sample weight, and employs weight sampling to construct a new memory set. Then we implement hint-based network learning based on hint representation to optimize the parameters. Finally, we conduct experiments on public and industrial datasets to demonstrate that our proposed method outperforms state-of-the-art continual learning methods in terms of memory capacity and prediction accuracy. Furthermore, we demonstrate remarkable practicability of DMSHM in real industrial applications.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑