Switched LAN Architecture

Access Layer

The access layer interfaces with end devices, such as PCs, printers, and IP phones, to provide access to the rest of the network. The access layer can include routers, switches, bridges, hubs, and wireless access points. The main purpose of the access layer is to provide a means of connecting devices to the network and controlling which devices are allowed to communicate on the network.

Distribution Layer

The distribution layer aggregates the data received from the access layer switches before it is transmitted to the core layer for routing to its final destination. The distribution layer controls the flow of network traffic using policies and delineates broadcast domains by performing routing functions between virtual LANs (VLANs) defined at the access layer. VLANs allow you to segment the traffic on a switch into separate subnetworks. For example, in a university you might separate traffic according to faculty, students, and guests. Distribution layer switches are typically high-performance devices that have high availability and redundancy to ensure reliability. You will learn more about VLANs, broadcast domains, and inter-VLAN routing later in this course.

Core Layer

The core layer of the hierarchical design is the high-speed backbone of the internetwork. The core layer is critical for interconnectivity between distribution layer devices, so it is important for the core to be highly available and redundant. The core area can also connect to Internet resources. The core aggregates the traffic from all the distribution layer devices, so it must be capable of forwarding large amounts of data quickly.

A Hierarchical Network in a Medium-Sized Business
Let us look at the hierarchical network model applied to a business. In the figure, the access, distribution, and core layers are separated into a well-defined hierarchy. This logical representation makes it easy to see which switches perform which function. It is much harder to see these hierarchical layers when the network is installed in a business.

The figure shows two floors of a building. The user computers and network devices that need network access are on one floor. The resources, such as e-mail servers and database servers, are located on another floor. To ensure that each floor has access to the network, access layer and distribution switches are installed in the wiring closets of each floor and connected to each of the devices needing network access. The figure shows a small rack of switches. The access layer switch and distribution layer switch are stacked one on top of each other in the wiring closet.

Although the core and other distribution layer switches are not shown, you can see how the physical layout of a network differs from the logical layout of a network.

Benefits of a Hierarchical Network

There are many benefits associated with hierarchical network designs.

Scalability

Hierarchical networks scale very well. The modularity of the design allows you to replicate design elements as the network grows. Because each instance of the module is consistent, expansion is easy to plan and implement. For example, if your design model consists of two distribution layer switches for every 10 access layer switches, you can continue to add access layer switches until you have 10 access layer switches cross-connected to the two distribution layer switches before you need to add additional distribution layer switches to the network topology. Also, as you add more distribution layer switches to accommodate the load from the access layer switches, you can add additional core layer switches to handle the additional load on the core.

Redundancy

As a network grows, availability becomes more important. You can dramatically increase availability through easy redundant implementations with hierarchical networks. Access layer switches are connected to two different distribution layer switches to ensure path redundancy. If one of the distribution layer switches fails, the access layer switch can switch to the other distribution layer switch. Additionally, If one of the distribution layer switches fails, the access layer switch can switch to the other distribution layer switch. Additionally, distribution layer switches are connected to two or more core layer switches to ensure path availability if a core switch fails. The only layer where redundancy is limited is at the access layer. Typically, end node devices, such as PCs, printers, and IP phones, do not have the ability to connect to multiple access layer switches for redundancy. If an access layer switch fails, just the devices connected to that one switch would be affected by the outage. The rest of the network would continue to function unaffected.

Performance

Communication performance is enhanced by avoiding the transmission of data through low-performing, intermediary switches. Data is sent through aggregated switch port links from the access layer to the distribution layer at near wire speed in most cases. The distribution layer then uses its high performance switching capabilities to forward the traffic up to the core, where it is routed to its final destination. Because the core and distribution layers perform their operations at very high speeds, there is no contention for network bandwidth. As a result, properly designed hierarchical networks can achieve near wire speed between all devices.

Security

Security is improved and easier to manage. Access layer switches can be configured with various port security options that provide control over which devices are allowed to connect to the network. You also have the flexibility to use more advanced security policies at the distribution layer. You may apply access control policies that define which communication protocols are deployed on your network and where they are permitted to go. For example, if you want to limit the use of HTTP to a specific user community connected at the access layer, you could apply a policy that blocks HTTP traffic at the distribution layer. Restricting traffic based on higher layer protocols, such as IP and HTTP, requires that your switches are able to process policies at that layer. Some access layer switches support Layer 3 functionality, but it is usually the job of the distribution layer switches to process Layer 3 data, because they can process it much more efficiently.

Manageability

Manageability is relatively simple on a hierarchical network. Each layer of the hierarchical design performs specific functions that are consistent throughout that layer. Therefore, if you need to change the functionality of an access layer switch, you could repeat that change across all access layer switches in the network because they presumably perform the same functions at their layer. Deployment of new switches is also simplified because switch configurations can be copied between devices with very few modifications. Consistency between the switches at each layer allows for rapid recovery and simplified troubleshooting. In some special situations, there could be configuration inconsistencies between devices, so you should ensure that configurations are well documented so that you can compare them before deployment.

Maintainability

Because hierarchical networks are modular in nature and scale very easily, they are easy to maintain. With other network topology designs, manageability becomes increasingly complicated as the network grows. Also, in some network design models, there is a finite limit to how large the network can grow before it becomes too complicated and expensive to maintain. In the hierarchical design model, switch functions are defined at each layer, making the selection of the correct switch easier. Adding switches to one layer does not necessarily mean there will not be a bottleneck or other limitation at another layer. For a full mesh network topology to achieve maximum performance, all switches need to be high-performance switches, because each switch needs to be capable of performing all the functions on the network. In the hierarchical model, switch functions are different at each layer. You can save money by using less expensive access layer switches at the lowest layer, and spend more on the distribution and core layer switches to achieve high performance on the network.