Cisco DNA Center and Catalyst Center
Cisco’s campus controller is the concrete product behind the ideas in the last two lessons. Called Cisco DNA Center when it launched and Cisco Catalyst Center today, it turns “intent” into device configuration and gives you a single dashboard for design, policy, provisioning, and assurance.
Intent-based networking
Intent-based networking (IBN) is the design philosophy the controller implements. Instead of typing commands that describe how to configure each device, you state the outcome you want — the intent — and the system works out the device configuration needed to achieve it, then keeps the network aligned with that intent over time.
A useful example: instead of assigning a VLAN and QoS policy interface by interface, an administrator says “voice endpoints on this floor get priority treatment.” The controller translates that into the specific VLAN, access-port, and QoS commands on the relevant switches, and then verifies that voice traffic is actually getting the priority it should. The user states what; the controller decides how.
IBN has three recurring properties worth memorising:
- Translation — converting intent into device-level configuration.
- Activation — applying that configuration to the network automatically.
- Assurance — continuously verifying that the network is behaving as intended and flagging deviations.
The four pillars
Cisco organises the controller’s features into four areas. You do not need to operate them for the exam, but you should be able to match each name to its job.
| Pillar | What you do | What it gives you |
|---|---|---|
| Design | Model sites, buildings, floors, device types, and credentials once. | A logical map of the whole campus |
| Policy | Express security and QoS intent as reusable rules. | Consistent policy applied across the fabric |
| Provision | Push configuration and fabric (VXLAN) rollout to devices. | Fast, templated deployment |
| Assurance | Monitor health, collect telemetry, run analytics. | Early warning and root-cause insight |
Automation and configuration
The controller automates by combining device discovery, templates, and policy. When a new switch is racked, the administrator adds it to the controller’s inventory. From then on, the controller owns its configuration: it can push a baseline from a template, apply the site’s policies, and keep the device consistent with the rest of its class.
This is where the idempotency and drift ideas from Lesson 1 pay off. Because the controller holds a model of what each device should look like, it can detect when a device has been changed by hand and either report it or restore the intended state. A command-line example of the kind of configuration the controller generates looks exactly like what you would type by hand — the difference is that a machine writes it, once, for every device, identically:
! Baseline template the controller would push to an access switch Switch(config)# vlan 20 Switch(config-vlan)# name DATA Switch(config)# interface GigabitEthernet1/0/10 Switch(config-if)# switchport mode access Switch(config-if)# switchport access vlan 20 Switch(config-if)# no shutdown
Beyond templates, the controller’s northbound REST API lets external systems drive it. A ticketing system can open a change, a script can call the controller’s API, and the controller applies the configuration. That is the full automation chain from this week: intent → controller API → device configuration.
Assurance and analytics
Automation without visibility is risky. Assurance is the controller’s monitoring and analytics engine. It continuously collects data from devices — interface counters, client onboarding, application experience, and telemetry — and turns it into a health score for the network, a site, or an individual client.
- Health dashboards — at-a-glance scores for network, clients, and applications, with drill-down to the offending device.
- Client 360 / Device 360 — a full history of a single client or device, showing every step of its onboarding and connectivity.
- Issues and trends — automatic detection of anomalies, with probable cause and suggested action.
- Telemetry — continuous streaming of data, rather than slow periodic polling, so problems surface in near real time.
Assurance is what distinguishes a modern controller from an older management tool. Traditional tools showed you raw counters; assurance tells you that a specific client is having a bad experience and points at the likely cause, using machine learning on the collected data to separate normal variation from a genuine problem.
Fabric provisioning
For a campus fabric, the controller can take the spine-leaf underlay plus the VXLAN overlay and build the whole thing from intent. The administrator defines the fabric sites, the control-plane and border nodes, and the policy; the controller then generates the underlay routing, the VXLAN encapsulation, the VTEP configuration, and the overlay policies on every participating switch.
| Fabric role | Function |
|---|---|
| Control-plane node | Runs the fabric control protocol (LISP/VXLAN) that maps endpoints to locations. |
| Border node | Connects the fabric to the outside world (WAN, other routing domains). |
| Edge node | A leaf that connects wired or wireless endpoints into the fabric. |
| Intermediate node | A switch inside the fabric that only forwards fabric traffic. |
Because the topology is defined once in the controller, adding a new switch to the fabric is a matter of assigning it a role; the controller computes the rest. That is the scale benefit of controller-based design made concrete.
Dashboards and APIs
Operators interact with the controller in two ways, and the exam may ask you to tell them apart:
- The graphical dashboard — a web interface with the health scores, maps, inventory, topology views, and assurance widgets described above. This is where humans work.
- The northbound REST API — the same capabilities exposed as HTTP endpoints, exchanging JSON, so scripts and other systems can work without a human. This is where automation lives.
Both do the same underlying thing: they talk to the controller, which talks to the devices. The dashboard is the API with a face on it. Everything from Lesson 5 onward — REST, HTTP methods, status codes, and JSON — is the toolkit for driving the API directly.
Review
Check what you have learned with these review questions.
| Key term | Meaning |
|---|---|
| Intent-based networking | Stating the desired outcome; the system translates, applies, and verifies it. |
| Translation | Converting business intent into device configuration. |
| Activation | Applying the translated configuration to the network automatically. |
| Assurance | Continuous monitoring that verifies the network matches intent. |
| Catalyst Center | Cisco’s campus controller (formerly DNA Center). |
| Telemetry | Continuous streaming of device/client data for analytics. |
| Template | Reusable configuration pattern with variables the controller fills in. |
| Border node | A fabric node that connects the fabric to external networks. |
Sources & references
Target exams: CCNA 200-301 v1.1 · CyberOps Associate 200-201. Last content verification: 2026-10-11.