ID	原文	译文
3783	为了解决大数据流式计算平台中存在计算负载剧烈波动，但集群因资源不足而遇到性能瓶颈的问题，提出了 Flink 环境下基于负载预测的弹性资源调度（LPERS-Flink）策略。	In order to solve the problem that the load of big data stream computing platform fluctuates drastically whilethe cluster was suffering from the performance bottleneck due to the shortage of computing resources, the load predictionbased elastic resource scheduling strategy in Flink (LPERS-Flink) was proposed.
3784	首先，建立负载预测模型并在此基础上提出负载预测算法，预测集群负载的变化趋势；	Firstly, the load prediction model wasset up as the foundation to propose the load prediction algorithm and predict the variation tendency of the processing load.
3785	其次，建立资源判定模型，以判定集群出现资源瓶颈与资源过剩的问题，由此提出弹性资源调度算法，制定弹性资源调度计划；	Secondly, the resource judgment model was set up to identify the performance bottleneck and resource redundancy of thecluster while the resource scheduling algorithm was proposed to draw up the resource rescheduling plan.
3786	最后，通过在线负载迁移算法执行调度计划，实现高效的节点间负载迁移。	Finally, the online load migration algorithm was proposed to execute the resource rescheduling plan and migrate processing loadamong nodes efficiently.
3787	实验结果表明，该策略在负载剧烈波动的应用场景中有较好的优化效果，实现了集群规模和资源配置对负载变化的及时响应，降低了负载迁移的通信开销。	The experimental results show that the strategy provides better performance promotion in theapplication with drastically fluctuating processing load. The scale and resource configuration of the cluster responded to the variation of processing load in time and the communication overhead of the load migration was reduced effectively.
3788	随着全球数据量的激增，集中式云计算无法提供低时延、高效率的视频监控服务。	With the explosion of global data, centralized cloud computing cannot provide low-latency, high-efficiency video surveillance services.
3789	基于此，提出分布式边缘计算模型，在边缘端直接处理视频数据，减少网络的传输压力，缓解中央云服务器的计算负担，降低视频监控系统的处理时延。	A distributed edge computing model was proposed, which directly processed video data at theedge node to reduce the transmission pressure of the network, eased the computational burden of the central cloud server,and reduced the processing delay of the video surveillance system.
3790	结合联邦学习算法，采用轻量级神经网络，分场景训练模型，并将其部署于计算能力受限的边缘设备上。	Combined with the federated learning algorithm, alight weight neural network was used, which trained in different scenarios and deployed on edge devices with limitedcomputing power.
3791	实验结果表明，对比通用神经网络模型，所提方法检测准确度提高 18%，模型训练时间有效减少。	Experimental results show that, compared with the general neural network model, the detection accu-racy of the proposed method is improved by 18%, and the model training time is reduced.
3792	针对天地一体化网络时变多维资源与差异多样化业务导致网络端到端时延难保障、网络资源利用率难提升等问题，提出基于时变图的天地一体化网络时间确定性路由算法及协议。	In terms of the difficulties to construct the end-to-end path and improve the utilization of network resources,caused by the time-varying multi-dimensional resources and diverse services over the space-ground integrated networks,the time-varying graph-based time deterministic routing algorithm and protocol for the space-ground integrated networkwas proposed.