Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document.
Note: Even the most advanced machine translation cannot match the quality of professional translators.
Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Hardware Centralized VXLAN Using the Gateway/Spine/Leaf Three-Layer Architecture
Figure 1 shows the hardware centralized VXLAN using the gateway/spine/leaf three-layer architecture. The border leaf node is used as the Layer 3 VXLAN gateway, the spine node aggregate east-west traffic, and the leaf node functions as the Layer 2 VXLAN gateway.
- Overall design:
- Extend the border leaf node group flexibly based on the service type and scale, as shown in Figure 2.
- Deploy ARP broadcast suppression globally and traffic suppression on an interface to prevent broadcast traffic from being flooded.
Figure 2-2 Networking of the flexibly extended border leaf node group
![]()
- Border leaf node:
- Border leaf nodes can work in active-active mode or constitute a stack. It is easier to deploy and maintain the stack, but the service interruption time is long during version upgrade. The centralized active-active gateways are therefore recommended.
- Routers, border leaf nodes, and spine nodes are fully connected at Layer 3, implementing ECMP-based forwarding. Routers and border leaf nodes are connected at Layer 3, and the paths between them are used as backup paths.
- VAS devices are connected to border leaf nodes in bypass mode. M-LAG is recommended when centralized active-active gateways are used, and M-LAG is recommended when the stack is used.
- Two paths are planned between border leaf nodes. One path is used as the peer-link of M-LAG, and the other path is used as the backup of the uplink.
- Spine node:
- The spine node aggregates east-west traffic, so it needs to provide large-capacity switching. You are advised to use the CE12800 as the spine node because the CE12800 can allow connected leaf nodes to be extended.
- Spine nodes are deployed in standalone mode. Spine nodes and leaf nodes are directly connected through Layer 3 routed interfaces, implementing ECMP-based forwarding. This design achieves load balancing of traffic, non-blocking forwarding, and fast convergence.
- Leaf node:
- When NICs of a server are connected in load balancing mode, leaf nodes support multiple networking such as the stack, M-LAG, and SVF composed of fixed devices. M-LAG is recommended because of its high reliability. When NICs of a server are connected in active/standby mode, leaf nodes use the standalone mode.
- When leaf nodes constitute an M-LAG, the Monitor Link group needs to be deployed. The uplink is associated with all downlinks, preventing traffic interruption when the uplink fails.