If you ask a handful of network engineers what keeps them up at night, it’s safe to say many would mention data center migrations. Not only can it take a full networking team several grueling months to complete – putting off other important IT initiatives – most organizations are bound to face multiple network outages during this timeframe.
The threat of impending downtime puts enormous pressure on IT teams and that’s no surprise when you factor in the cost of IT downtime, which according to Gartner is an average of $5,600 per minute. But migration day no longer needs to be the network equivalent of doomsday.
By implementing automation and following three critical phases of a data center migration project, network engineers can sleep easier at night knowing the tedious and risky parts of the project can be handled with little to no risk at all.
Phase 1: Assess what you’re working with
The first step to any migration project is to have complete understanding of every aspect of the network. However, that’s easier said than done with the complexity that comes with today’s modern hybrid environments - consisting of hundreds of thousands of components that need to be discovered, verified, and documented during the migration process. Network engineers must inventory every piece of network equipment, develop topology and design diagrams of the data center, list every application affected by the network and develop detailed maps of application dependencies.
This is the phase in which most organizations fail during a migration project. While the majority of organizations realize the importance of understanding how each and every device and application interconnect throughout the network, many organizations simply don’t have the resources to document their networks properly. These teams still rely on tedious, manual documentation methods, which become obsolete quicker than the time it takes to create them. In today’s landscape, manual documentation doesn’t cut it and essentially leaves network teams flying blind as they embark on the data center migration process.
Automating network diagramming not only takes the painstaking tasks off an engineer’s plate, allowing them to focus on more strategic initiatives, it also gives teams a leg up on special projects like data center migration. Automated systems can create a “digital twin” of the entire network, including information like device data, topology, design configuration, routing, etc., which update themselves automatically whenever a change takes place. Such systems must also adapt to any data center network environment in order to support both traditional and software-defined infrastructure, like Cisco ACI or VMWare NSX.
Phase 2: Plan, test, adjust
Most organizations take a shot-in-the-dark approach to data center migration, migrating the network now, then troubleshooting any issues that arise later. However, any sophisticated migration strategy will include a lot of planning, testing and adjusting beforehand. With the help of a centralized automation system, network teams can easily develop detailed connectivity mapping of the new data center layout and routing protocols, identify virtualization, consolidation, and enhanced storage opportunities, as well as identify space, circuit, hardware and software requirements, all before the migration takes place.
With the ability to visualize and validate every configuration and check key application flows and routing table changes in advance by leveraging their network’s digital twin and using automation that adapts to the knowledge of what must be verified, teams can proactively guard against misconfiguration and reduce any outages after the migration. Once the teams feel confident about their planned changes, they can automatically execute them to save time and avoid human errors.
Phase 3: The final check + troubleshooting
When all is said and done, and the network is migrated, the final step for network teams is to do one last check to ensure it is running smoothly. While you may have planned your migration down to every last painstaking detail, it’s always possible something didn’t go according to plan. At the same time, migrations inevitably lead to small changes in the network, making troubleshooting much slower and less reliable. If and when network teams realize the network isn’t operating as usual, manual troubleshooting techniques could result in hours or even days spent trying to identify what exactly the issue is.
Automation that adapts to the knowledge about what needs to be validated and tested, on the other hand, greatly reduces the steps it takes a human to diagnose and mitigate an issue by hand. Dynamic maps along with executable runbooks quickly allow network teams to identify the root cause of an issue and automatically remediate that issue, minimizing any downtime as a result of migration changes.
When it comes to data center migrations, network engineers typically don’t have a lot to look forward to – besides months of anxiety about unplanned downtime due to a migration mishap. But, it doesn’t have to be that way. With the right tools in place, network teams can rest easy knowing the process is being managed properly and they are doing everything in their power to ensure a successful migration project.