Network teams have historically been slow to adopt automation tools, but enterprises are beginning to realize the value of extending their automation strategy to their network. Hear from Andy Schulte, senior network delivery consultant at Evolving Solutions, on the benefits of network automation—and how to get it right.
While automation is an essential part of modern operations, IT networks have, until now, been slower to adopt automation tools. However, enterprises are recognizing that the efficiencies automation brings to their server, DevOps, cloud, and other environments should also be extended to the network.
Network teams deal with an array of responsibilities. They implement new technologies; manage IT security; and install, configure, and support critical equipment such as routers, firewalls, and switches. The necessary focus on these tasks—on simply keeping things running day to day—may partially explain why network teams generally lag in their adoption of automation. In any event, by applying and implementing a carefully crafted and well-defined strategy, network operations can also benefit significantly from automation.
Network Automation Use Cases
What can be achieved through network automation? One potential benefit is improved workflow efficiencies across various IT teams. Say your current process for provisioning servers involves issuing individual service tickets. The server team gets a ticket for the server build, and then a request for IP address or a DNS host name is created. There may be additional requests for load balancing and firewall rules. Traditionally, this is a manual process where communication across multiple teams often leads to substantial delays. But imagine using a tool like Ansible Automation Platform to manage the workflow. Now the server team can work with the job templates that are related to the server build while the network team manages its templates, but these actions are effectively woven together. Basically, a single request inputs all the information and then kicks off this workflow that does the rest automatically.
Or consider the network change process. A maintenance window is scheduled to implement a change. Using automation, all the relevant information about the network and its topology can be collected prior to implementing the change. Once the change occurs, automation allows you to collect the same information so it can be analyzed to ensure that the results of the change meet expectations.
Finally, automation can be implemented to conduct recurring failover and disaster recovery tests. While most enterprises design their infrastructure for redundancy and fault tolerance, far fewer actually test failure scenarios. Automation greatly simplifies the testing process. Many common network failure scenarios can be triggered using automation, allowing you to monitor the results on an ongoing basis and verify that your network is capable of handling failures and is truly redundant.
Getting Network Automation Right
While each of these examples represents a viable and valuable use of automation, they are also ambitious projects. It is important to understand that, for network team members who may be getting started with automation, it is best to start with something small. Conducting reports or collecting data are low-risk entry points to automation. For example, a service desk team needs information about client devices: the switch hostname and switchport number that a client device is connected to, the VLAN assigned to the switchport, the link status, and the 802.1x authentication status. A form can be created to automatically retrieve this information, saving service desk personnel the effort of logging into the network devices themselves.
Some other network automation tips:
- Collaboration is key: Network teams should develop a common automation environment and make common tools (e.g., a code editor like Visual Studio Code and a code repository like GitHub or GitLab) available to all team members. Schedule recurring automation working sessions to ensure all team members are comfortable with the automation tools.
- Leverage a source of truth system: Implement a system that will act as the source of truth for network automation. This should include a complete inventory of network devices and relevant data about each device (management IP, device type, site/region, etc.). Ideally, it should include a way to add custom fields/parameters for each device. Automation tools can then query this source of truth system to retrieve data as needed.
- Do not reinvent the wheel: The network team should work with other internal IT teams that are already using automation and automation tools. There is no need for the network team to create a code repository if the server team already has one.
- Properly store sensitive information: A secrets manager is a repository for storing passwords, certificates, tokens, and other sensitive data. Numerous options are available, including HashiCorp Vault and the credential manager within Ansible Automation Platform. A final point about security and automation: Be sure to establish distinct credentials for lab and production devices to prevent accidental automation that crosses from the lab to production.
Ultimately, everyone on your network team should be comfortable with the integration of automation. If only one person is implementing and using automation, what happens if that person leaves the organization? That knowledge goes out the door with them. A successful automation initiative starts with ensuring that everyone is on board and working as a team.
Let’s talk. To learn how to implement an effective network automation strategy, connect with one of our technical experts.