Logical Architecture
iMaster NCE-CampusInsight uses Huawei's big data analytics platform, receives device data through Telemetry technology, and analyzes and presents network data using intelligent algorithms. Figure 1-20 shows the logical architecture of iMaster NCE-CampusInsight.
- Data collection: iMaster NCE-CampusInsight interconnects with network devices through southbound interfaces for device management. The following types of southbound interfaces are supported: Telemetry-based HTTP/2 + ProtoBuf, SNMP, and Syslog. The following describe these interfaces in detail:
HTTP/2 + ProtoBuf: iMaster NCE-CampusInsight uses Telemetry-based HTTP/2 + ProtoBuf interfaces to collect device performance data. ProtoBuf is a mechanism for serializing structured data and is widely applied in data storage and communication protocols. HTTP/2, on the other hand, uses the Secure Sockets Layer (SSL) and Transport Layer Security (TLS) protocols for authentication and encryption of communication channels.
SNMP: iMaster NCE-CampusInsight supports standard SNMPv2c and SNMPv3 interfaces, through which it can interconnect with network devices. SNMP is an application-layer network management protocol based on the TCP/IP architecture. SNMP uses the User Datagram Protocol (UDP) as its transport-layer protocol, and can be used to manage network devices that support proxy processes.
Syslog: Syslog is a protocol for forwarding system log information on IP networks, and has become a standard industrial protocol for recording device logs. iMaster NCE-CampusInsight receives log data reported by devices through the Syslog protocol.
- Data analysis: The big data analytics platform can collect and analyze millions of data flows per minute based on the distributed database, high-performance message distribution mechanism, and distributed file system. Of these, the distributed database provides distributed computing, aggregation, and storage of large amounts of real-time data, as well as supporting multi-dimensional data retrieval and statistics query in seconds. The machine learning algorithm library currently contains multiple network O&M and analysis algorithms, providing intelligent services for upper-layer O&M applications. It can be constantly expanded.
- Data service: iMaster NCE-CampusInsight provides a large number of application services for data analysis based on typical O&M and troubleshooting scenarios of campus networks. For example, it can intelligently detect connection, air interface performance, roaming, and device issues, analyze connection and performance issues, play back user journeys, analyze AP details, and detect the quality of audio and video services.