基于分布式系统的资源运维管理可靠性提升方法

发布时间：2018-03-19 16:47

本文选题：资源运维管理　切入点：分布式网络系统　出处：《计算机应用》2017年S1期 　论文类型：期刊论文

【摘要】：提出一种基于分布式网络系统资源运维管理的可靠性提升方案,采用深度神经网络学习算法进行基于分布式网络系统的判断调度决策估计方法。通过这种资源分配方案,可以满足电网信息系统对一致性和监测评估处理方面的要求。在内网环境中的评估实验表明,基于本方案的资源运维管理方式,其使用率和成功执行率在任务个数较多时仍能维持较好的比率,且其任务管理策略也优于其他方式。在较难预测的复杂场景下,在维持计算复杂度的条件下对全网络系统的资源运维管理起到优化作用,同时在此过程中,对性能和相应速度等指标的波动具有快速处理能力。总之,这种方法不仅能提高分布式网络系统的使用率和成功执行率,还具有可靠性、高精度及可扩展性上的优势。
[Abstract]:This paper presents a reliability enhancement scheme based on distributed network system resource operation and maintenance management, and uses depth neural network learning algorithm to estimate the decision of judgment and scheduling based on distributed network system. It can meet the requirements of consistency and monitoring, evaluation and processing in power grid information system. The evaluation experiments in the inner network environment show that the resource operation and maintenance management method based on this scheme can be used. Its utilization rate and success rate can still maintain a good ratio when the number of tasks is large, and its task management strategy is superior to other methods. Under the condition of maintaining the computational complexity, the resource operation and maintenance management of the whole network system is optimized, and in the process, the fluctuation of the performance and the corresponding speed can be dealt with quickly. This method not only improves the utilization rate and successful execution rate of distributed network system, but also has the advantages of reliability, high precision and scalability.
【作者单位】：北京国电通网络技术有限公司;国网浙江省电力公司;清华大学电子工程系;
【基金】：国家电网公司科技项目(SGZJ0000BGJS1500433) 国网信息通信产业集团有限公司科技项目(SGITG-KJ-JSKF[2015]0003)
【分类号】：TP183;TP393.07

【参考文献】