Key Points of Network Connectivity Deployment
Network connectivity deployment aims to enable Layer 2 and Layer 3 communication between devices at the core, aggregation, and access layers of a campus network, so that wired and wireless users can access the campus network and communicate with each other. Network connectivity deployment is the basis of campus network construction.
Based on the services and scale of the campus network, the network connectivity deployment on the campus network has the following key points:
Deployment Key Point |
Description |
Recommended Scenario |
|---|---|---|
Determine whether to use a two-layer or three-layer network architecture. |
A two-layer network architecture consists of a core layer and an access layer. A three-layer network architecture consists of a core layer, an aggregation layer, and an access layer. For details, see Deployment Differences Between Two-Layer and Three-Layer Network Architectures. |
For a small-scale network, use a two-layer network architecture that features simple networking, a small number of NEs, and fewer potential failure points. For a large-scale network, use a three-layer network architecture that features complex networking, a large number of NEs, and more potential failure points. In actual applications, a two-layer network architecture is also used when the transmission distance is short and the core layer has enough interfaces to directly connect to the access layer. This reduces the total cost and maintenance workload, and facilitates network status monitoring. |
Determine whether devices on the network need to set up a Cluster Switch System (CSS) or stack. |
It is recommended that a CSS or stack be deployed according to Typical CSS and Stack Deployment as this improves network reliability. |
To improve network reliability, it is recommended that a CSS be deployed at the core layer and a stack be deployed at the aggregation layer. If a large number of users need to access the network, deploy a stack at the access layer to increase the number of interfaces. |
Determine the locations of gateways on the network (that is, the boundary between Layer 2 and Layer 3). |
In most cases, Layer 2 switching services are deployed on downstream devices connected to gateways, and Layer 3 routing services are deployed on upstream devices connected to gateways. |
You are advised to determine the locations of gateways based on the network scale. That is, use core switches as gateways on a small-scale network, and use aggregation switches as gateways on a large-scale network. |
Determine whether to deploy Super Virtual Fabric (SVF). |
SVF virtualizes network devices at multiple layers into one logical device, and allows a control device to centrally manage and configure access devices, simplifying management and configuration. |
On a new network, deploy SVF to simplify management and configuration. |
Determine whether to use the native AC, standalone AC, or ACU2 solution. |
The native AC solution uses X series cards to implement wired and wireless convergence. In the ACU2 solution, ACU2 cards are typically installed on core switches. In the standalone AC solution, standalone ACs can connect to core or aggregation switches in off-path mode. For details about the differences between the standalone AC and ACU2 solutions, see Deployment Differences Between a Standalone AC and an ACU2. |
For new small- and medium-scale networks, the native AC solution is typically used to implement unified wired and wireless management. For medium- and large-scale networks, wired networks that require wireless capacity expansion, and networks with a large number of APs, the standalone AC or ACU2 solution is typically used. |
Determine whether to deploy ACs in off-path or in-path mode. |
ACs can connect to core or aggregation switches in off-path mode to manage APs. ACs can also directly connect to APs or access switches in in-path mode. The ACs also function as aggregation switches to forward and process APs' data and management services. |
It is recommended that ACs be deployed in off-path mode for a live network that is partially reconstructed. For small- and medium-scale new networks, deploying ACs in in-path mode is recommended to simplify the network architecture. |
Determine whether standalone ACs forward wireless data in direct or tunnel mode. |
In direct forwarding mode, ACs forward service packets of APs to the upper-layer network without performing CAPWAP encapsulation. |
Tunnel forwarding applies to scenarios where service data needs to be managed and controlled in a centralized manner. Direct forwarding applies to scenarios where high packet forwarding efficiency is required.
|
Deployment Description
- There are two-layer and three-layer network architectures. Compared with the three-layer architecture, the two-layer architecture does not have the aggregation layer. This chapter uses the three-layer architecture as an example. For differences between the two architectures, see Deployment Differences Between Two-Layer and Three-Layer Network Architectures.
- Multiple switches configured with the CSS or stacking function are virtualized into one logical switch, simplifying the configuration and networking. For a deployment example, see Typical CSS and Stack Deployment.
- An ACU2 is an AC card that can be deployed in the same way as a standalone AC. The only difference is that internal interconnection interfaces need to be configured between an ACU2 and a modular switch. For details about the deployment differences, see Deployment Differences Between a Standalone AC and an ACU2.
- The wireless network deployment examples described in this chapter apply to common and high-density WLAN scenarios. For details about wireless network deployment examples in agile distributed Wi-Fi and WDS backhaul scenarios, see Wireless Coverage Deployment.