Innovative IRF virtualization technology for distributed networks
Traditional distributed networks use 1+1 redundancy of channels and devices to ensure availability, which increases reliability network, but at the same time leads to low load of channels and devices and at the same time increases the complexity of management and maintenance. Focusing on the needs for virtualization of cloud computing advanced networks, H3C uses IRF technology in devices for distributed networks. IRF technology virtualizes multiple physical devices into a single IRF fabric, which significantly reduces network management and maintenance costs while increasing utilization (efficiency) of channel and device bandwidth. Thanks to IRF technology, the CR16000-F routers realize the following advantages:
Using link aggregation technology across different chassis for load balancing and redundancy of multiple backbone connections. This improves network reliability and improves link resource utilization.
Virtualizes multiple CR16000-F routers into a single IRF fabric, providing a unified control plane and transmission plane data. This simplifies the network topology, improves management efficiency and reduces maintenance costs.
Use proprietary H3C technology for fault-tolerant, stateful routing with real-time redundancy for control plane and data plane data across the entire virtualized architecture. This ensures that there are no service interruptions due to critical elements whose failure could lead to the failure of the entire system.
Powerful BRAS functionality
Traditional service routers provide telecommunications services such as mobile access, corporate communications and IP television. BRAS devices act as gateways to access the broadband network, providing authentication and user management functions. CR16000-F models combine the functions of a service router and BRAS, providing greater utilization of device resources and cost savings.
Intellectualaccountingby purpose(iTA) – ensures differentiation of service types depending on destination addresses and implements accounting, bandwidth management and QoS parameters for each service type.
Offers a unified user authentication mechanism for a large number of wired and wireless clients, supporting mobility requirements of wireless terminals and unified authentication for all services.
Switchingon reserveBRASinIRF withcontrolstate – eliminates service interruptions that can be caused by the failure of a critical element affecting the entire system and simplifies management and maintenance.
Industry-leading network operating system
The control plane of the CR16000-F router uses multi-core and symmetric multiprocessing (SMP) processing technologies enabled by the Comware V7 platform. Each software module operates in an independent address space with support for dynamic loading and isolated updates.
The Comware V7 platform supports distributed computing functions. Global services such as MPLS and BGP can be assigned to specific MPUs. By distributing global services across different MPUs, processor load is reduced and system performance is improved. Thanks to distributed computing in the system, it becomes possible to divide global services into subservices with the distribution of subservices among different processors.
Comprehensive set of services
The CR16000-F router uses an Open Application Architecture OAA) and offers Open Application Platform (OAP) modules to meet customization and upgrade needs. For example, firewall modules, intrusion prevention system (IPS) modules, and load balancing modules are offered to effectively combine core routers and service systems.
High performance and reliability
< p style="font-weight: 400;">The CR16000-F router offers traffic forwarding performance of 800 Gbps in each slot, as well as high port density of 10GE / 40GE / 100GE.
CR16000-F routers provide redundancy for all major components, including control boards, independent switching matrices, power supplies, fans, etc.
CR16000-F routers support a full range of reliability features to help maintain network availability even if an entire link or node fails, including BFD, LAG, NQA, ECMP, FRR and others.