What is NetDevOps?

DevOps is a culture.

It is a mindset with a foundation in the Agile Manifesto.

Bridging the gap between Dev and Ops, while providing better software, more efficiently, through automation and replication.

There is also continuous delivery. DevOps is many things, but what is still on the horizon is bringing networking into the fold.

This is NetDevOps.

Bringing the DevOps principles to networking, and networking into the DevOps team.

Cumulus networks has provided a great start to the NetDevOps movement. The tools you are already using are incorporating it as well. Ansible, Chef, Puppet, and Red Hat are beginning to incorporate DevOps into networking.

This brings us to NetDevOps. The intersection of networking and DevOps.

What is NetDevOps?

NetDevOps is infrastructure as code. It is network automation. It is open communication. Similar to DevOps, the NetDevOps goals are as follows:

• Prevent fat fingers
• Repetitive tasks
• Reproduce and dispose things
• Deliver code

The result is network scalability and a replicable process.

Infrastructure as Code

If NetDevOps is infrastructure as code, it makes sense to provide a definition:

“Infrastructure as Code (IaC) is the process of managing and provisioning computing infrastructure (processes, bare-metal servers, virtual servers, etc.) and their configuration through machine-processable definition files, rather than physical hardware configuration or the use of interactive configuration tools.”

What this provides is the ability to apply and deploy changes at scale, with the removal of manual configurations.

Traditionally, network changes had to be done on physical servers; manually and individually. With the application of infrastructure as code, this is accomplished remotely with the use of code. The code is then replicable and can be scaled. A benefit to this process is the ability repeat the processes that work particularly well.

The result is network automation.

Network Automation

Network automation is the result of the NetDevOps approach. As DevOps places a large emphasis on automating processes, NetDevOps looks to share this goal. This is accomplished through a number of tools, Datapath.io being one of them.

Network automation is setting parameters around network requirements, and implementing a set of code to replicate the process. The tools in your current stack of IT automation already allow you to accomplish this for other IT related functions, so let’s bring this to networking.

How Do I Incorporate NetDevOps?

The first step in a NetDevOps approach is to pick an infrastructure which allows for the deployment of infrastructure as code for your network.

Two approaches are as follows:

White-box Switches and Linux

NetDevOps, in its current form, is done by setting up White-box Switches, installing Linux, and then utilizing your current DevOps IT stack. The White-box switches provide the ability to run code for your network.

NetDevOps Through a Cloud Provider

Making the transition to a cloud provider, such as AWS, provides infrastructure as code services through AWS CloudFormation. As the rest of the DevOps team may already be employing tools that allow for automation in the cloud environment, applying this to networking is the next logical step.

The goal of NetDevOps is to find tools that allow you to monitor and automate your network. Whether this is to gain improved network speeds and performance, or to diagnose issues, NetDevOps brings the DevOps and networking team together.

References

Article originally posted on the Datapath.io Blog.