Applying Automation to a Network Upgrade

Jason Baudreau
By Jason Baudreau April 20, 2017 4 minute read
Jason Baudreau has a passion for enterprise networking, writing and geeking out with Smart Home technology. With 8 years' experience in Systems and Network engineering, Jason now specializes in network automation technologies at NetBrain.

Whether an organization outgrows their existing network or hardware just becomes end-of-life, a network upgrade is virtually inevitable for any business. While it’s a common issue, it’s one that comes with many hurdles for IT teams. It is a difficult task to  navigate the requirements of both the current and new network and determine what new equipment and configurations are necessary. This process can take several weeks or even several months, putting other critical networking and IT projects on hold.

For most organizations, a hardware upgrade like this is broken down into five phases, each of which come with a different set of challenges.

Data Center Network Upgrade

Phase 1: Analysis of Existing Network Infrastructure

It always comes back to documentation and network mapping. Many organizations realize that they are missing or have limited documentation of their existing network infrastructure, limited understanding of application requirements and traffic flows and limited understanding of existing design such as security, routing, or layer-2.

Phase 2: Requirements Gathering & Design

This is a time-consuming process to create documentation for an entirely new network design. With manual mapping processes this can take many days.

Phase 3:  Design Review & Evaluation

Network teams need to ensure that the newly configured devices meet compliance standards – whether for SOX, HIPAA, or general security. It’s a significant challenge to validate the new network design before it goes live, and have limited means to test this process.

Phase 4: Implementation

Implementing configuration changes into a new network manually is a painstaking process and often results in many errors.

Phase 5: Operation

Once the new network is up and running, it takes time for network engineers to get up to speed on the newly deployed technologies. This results in an inefficient troubleshooting process should a  network incident occur.

It’s easy to see that a network upgrade like this results in a bevy of challenges for network teams. The good news is that NetBrain can drastically reduce the manual work required for this process and help ensure a smoother network transition. Here’s how NetBrain helps:

Eliminate Manual Network Documentation

NetBrain can automate documentation, both for the pre-existing and post-upgrade networks. Dynamic Network Maps have nearly infinite detail to document any aspect of design (e.g. firewall policies, dynamic routing, multicasting, layer-2 design). Even asset reports and design documents are fully automated. This saves hours of painstaking documentation updates and checks for the networking teams.

Network teams also need to understand critical application flows so they can minimize interruptions during the network upgrade and accommodate these applications with the new design. Manually mapping these flows can take weeks – NetBrain can map each application path instantly.

Validate Compliance and Security Policies

By modeling network changes virtually (e.g. leveraging a simulation tool like VIRL) network teams can create a digital model of the proposed design. NetBrain can discover and map the simulated network to visualize and validate the new design. For example, engineers can implement a routing change on the simulated network and visualize the resulting traffic flows. This allows for critical validation before actually converting to the new network – an incredibly valuable testing mechanism.

A further safeguard that network teams should take is to proactively guard against misconfiguration. To ensure that every network upgrade meets pre-defined design and security standards, implementation engineers should execute design and security validation Runbooks. These Runbooks will scan each configuration to ensure they meet the predefined ‘golden’ requirements.

Accelerate Troubleshooting

To ensure that network engineers can troubleshoot quickly and efficiently once the network is upgraded, the design team can create Executable Runbooks for the operations team to use once the upgrade is complete. These Runbooks can include design guides and troubleshooting best practices which help improve hand-off and minimize the impact of network outages after the upgrade.

Network diagnoses can also be automated through Executable Runbooks. Network monitoring tools are great for identifying issues, but provide little insight into the cause of the problem. NetBrain can cut troubleshooting time in half by automating hundreds of diagnoses.

Simplifying the network upgrade process even further is that with NetBrain, a Dynamic Network Map becomes the single pane of glass for operations, integrating with existing systems and workflows. This cuts down on relying on the command line interface and helps provide a clearer picture of the network.

While automation still can’t physically pickup and install new network hardware, it can be extremely powerful for almost every other aspect of a network upgrade. Since the IT team is always under constant pressure to deliver services on time, it can be a real life-saver.