We are now entering the intelligent digital era, in which data centers are no longer merely business support centers; rather, they are becoming value creation centers. As such, they help carriers further deepen digital transformation, enable data monetization, create greater business value, and achieve operational excellence. Huawei's Private Cloud/Public Cloud DCN Solution implements full automation of device configuration and network services based on the iMaster NCE-Fabric management and control unit, and gains insights into each application interaction data flow on the entire network, making it easy to detect faults. This solution adopts CloudEngine series switches to build an ultra-broadband fabric network with low latency and zero packet loss, helping customers build an intent-driven cloud data center network, reshape user experience, and enable service agility.
Designing a data center network is a complex task. Experts are responsible for designing the network architecture, scale, and parameters such as interconnection, external connection, interfaces, and protocols.?This process involves a heavy workload and is often inefficient.?Deliverables generated in the design phase, such as low level design (LLD), cannot be seamlessly adapted into the network construction phase, while a large number of repetitive manual operations are required. iMaster NCE-Fabric integrates design and network construction, and provides standard network architectures and network design templates. You can flexibly modify or select a template to implement automatic network deployment in one click. All you need to do is connect and power on devices, and then start the one-click deployment. The network deployment will then take about 40 minutes. After that, iMaster NCE-Fabric verifies the deployed network to check the network connectivity and whether the network is consistent with the design, ensuring successful network deployment.
Nearly 40% of faults on the network are caused by human error, such as logical vulnerabilities in network design, deviation from administrator intents, and misoperations. Risk evaluation of network changes and policy adjustment involves a heavy workload.
iMaster NCE-Fabric provides the simulation verification module. With live network configuration, topology, and resource information, iMaster NCE-Fabric analyzes whether network resources are sufficient. It displays detailed connection relationships and analyzes the impact of changing configurations on the corresponding services through network modeling and formal verification algorithms. This enables network engineers to evaluate change risks in advance and eliminate human error.
Data centers are central to supporting services and therefore have zero tolerance for network interruption. However, traditional network O&M involves a lot of manual work. Service continuity may be severely affected if a network fault occurs, due to difficulty in locating the fault.
iMaster NCE-Fabric uses Telemetry to collect traffic on the management, forwarding, and data planes of the entire network and comprehensively evaluates network health based on service experience. It proactively forecasts faults and detects existing faults within 1 minute. And with Huawei's proprietary AI algorithm, iMaster NCE-Fabric performs in-depth feature mining and learning to locate the root causes of 75 types of typical faults (grouped into 7 categories) within 3 minutes. In addition, an intelligent decision-making system enables iMaster NCE-Fabric to analyze the fault impact and recommend preferred troubleshooting methods for quick rectification of typical faults within 5 minutes.