ZTE shares network management intelligence and AI application in SDN/NFV Industry Alliance seminar
22 September 2017, Shenzhen, China – ZTE Corporation (0763.HK / 000063.SZ), a major international provider of telecommunications, enterprise and consumer technology solutions for the Mobile Internet, today announced that it attended the software defined networking/network function virtualisation (SDN/NFV) Industry Alliance Network Intelligence seminar, themed on network management intelligence and AI application, in Beijing on September 22nd 2017. At the event, which was attended by mainstream operators and vendors, ZTE’s service and management and orchestration (MANO) product department operations support systems (OSS) chief engineer, Wang Rui, shared how to apply AI technology to network management scenarios, improve network management efficiency and support business innovation.
As 5G approaches and the transformation of NFV/SDN networks accelerates, many new virtualised network features, such as multilayer networks, dynamic change, virtual and reality coexistence, bring many new challenges for network management. AI is needed to help customers refine the models and rules suitable for network management in a big data environment, to help make recommendations and assist decision-making.
“The introduction of AI provides the SDN/NFV network with vast potential for use in operation and maintenance, resource utilisation and efficiency,” Wang Rui said. “In our practice, the AI correlation algorithm and data cleaning improve the extraction efficiency of root cause analysis (RCA) rules by 70 percent, which greatly reduces the dependence on staff skills.”
RCA rules generation changes from the traditional accumulation of expert personal experience to the real-time generation of machine learning and optimisation. At the same time, the ZTE TensorFlow intelligent platform, with an NVIDIA graphics processing unit (GPU) and 100G remote direct memory access (RDMA) network construction of a high-performance computing (HPC) container cluster, realises the automation of the evaluation model, the hyper parameter selection and model optimisation, providing a wealth of machine learning, deep learning algorithms and greatly enhancing the utilisation rate of resources.
The ZTE cloud-oriented network management platform, CloudMaster, has a built-in AI engine, realised multi-layer fault correlation, real-time network fault positioning and real-time prediction in the virtual network, to provide a strategy of automatic decision-making. At the same time, work order intelligent scheduling gets significant improvement through the AI intelligent vehicle scheduling algorithm. A vehicles’ average daily mileage decreases by 20 percent and resource availability increases by more than 30 percent. CloudMaster provides end-to-end, fully automatic fault closed-loop and support for the intelligent transformation of network operation and maintenance.
ZTE will continue to work with customers and partners in AI, creating greater value for customers through industry digital transformation.