ID	原文	译文
1853	进而应用 M/M/n/m 和 M/M/1 /m 排队模型分别刻画控制器集群的 Packet-in 消息处理过程和 OpenFlow 交换机的分组处理过程。	Then the M /M /n/m and M /M /1 /m queueing models are respectively applied to depict Packet-inmessage processing process of its controller clusters and packet processing process of its OpenFlow switches.
1854	在此基础上，建立 OpenFlow 分组转发优先制排队模型，进而推导出不同优先级的分组转发时延及其累积分布函数 CDF。	On this basis, we establish a prioritized queueing model of OpenFlow packet forwarding, and derive packet forwarding delays of different priorities and its cumulative distribution function.
1855	最后，借助控制器性能测量工具 OFsuite_Performance 进行实验评估，结果表明:与现有模型相比，本文所提的 M/M/n/m 模型更能准确估计控制器集群的实际性能。	Finally, experimental evaluation in virtue of the controller performance measurement tool OFsuite_Performance shows that our proposed M /M /n/m model can accurately estimate actual perform-ance of controller clusters compared with existing models.
1856	同时，采用数值分析的方法对比了多种情况下不同优先级的分组转发时延及 CDF 曲线，为软件定义核心网的实际应用部署提供有效参考。	Meanwhile, we contrast packet forwarding delays of different priorities in various cases and their CDF curves by numerical analysis, which provides effective references for practical deploy-ments of software-defined core networks.
1857	针对大数据背景下数据分类问题，已有的在线学习算法通常引入 L1范数正则化增强预测模型的稀疏性，但单一的正则化约束不能高效的获取稀疏模型。	To deal with data classification problems under the background of big data, many existing online learning algorithms usually take advantage of L1norm regularization to enhance the sparsity of the prediction model. However, asparse prediction model cannot be obtained efficiently by a single regularization constraint.
1858	基于此，提出了一种具有双重稀疏机制的在线学习算法(an onlinelearning algorithm with dual sparse mechanisms，DSOL)。	In this paper, an online learningalgorithm with dual sparse mechanisms (DSOL)is proposed.
1859	在 DSOL 算法中，一方面利用 L1 /2正则化项约束目标函数以增强预测模型的稀疏性，提高算法的泛化性能。	In DSOL algorithm, the objective function is constrained byL1 /2regularization in order to enhance the prediction model's sparsity, and then improve the generalization ability of DSOL.
1860	另一方面用改进的梯度截取法对数据特征进行选择，有效稀疏化预测模型。	Furthermore, an improved truncated gradient method is applied to enhance the sparsity of the prediction model through prop-erly selecting the features of data.
1861	通过 L1 /2正则化与改进的梯度截取策略的有机融合，有效利用了历史数据信息，提高了算法分类数据的性能。	By the organic integration of the above two sparse mechanisms, including the L1 /2regulari-zation and the improved truncated gradient method, some historical data information can be effectively utilized, and then theperformance of the algorithm in data classification can be greatly improved.
1862	通过与另 4 种代表性稀疏在线学习算法在 9 个公开数据集的实验对比表明 DSOL 算法对数据分类的准确性更高。	Extensive experiments between DSOL and other4 popular sparse online learning algorithms on 9 open data sets manifest that DSOL algorithm yields more favorable perform-ance on data classification.