Every day, a new product launch promises speed, efficiency or savings for businesses eager to stay a step ahead. But if you’re an IT pro, and it’s your job to onboard the new system, the allure gives way to practical considerations: How are you going to move all that critical data—including dependencies, settings and custom configurations—from one system to another so it all talks to each other the way it’s supposed to? Whether you’re tackling a one off project or looking to improve your in-house skills, these eight tips for mastering migration will help you form a strategy, determine costs and find technical solutions to ensure successful migration no matter what platform you’re on or which one you’re moving to.
Tip #1 – Factor downtime into your cost analysis
Today’s businesses have no tolerance for downtime on critical systems and data. Anytime a user can’t access that data, there’s a cost. But most businesses neglect to calculate the cost of downtime, even after experiencing it. Generic estimates only provide a partial picture of the true cost of downtime for your organization. In pricing out the total cost of ownership (TCO) for protection, a critical factor will be the downtime cost for each application in your ecosystem. This includes:
- Decreased productivity – How many employees are affected and for how
- Lost revenue – What are the direct losses from downtime, such as a missed
sale or an inability to bill for services?
- Financial performance – How does downtime affect key metrics like cash
flow, revenue recognition and stock price?
- Surprise expenses – Did you have to hire additional help, pay overtime, or
cover extra shipping or legal fees?
- Damaged reputation – Will customers, suppliers, partners and investors
look at you differently?
Before you consider how much the fix is going to cost, first understand the cost of the problem.
Tip #2 – Use the right tool for the job
The growth of public cloud services from major players like Amazon, Google and Microsoft has led to a proliferation of tools for migrating to a specific cloud platform. But these tools use inferior technology that opens the door for data loss during the migration process. They’re often platform-dependent, engineered for one type of migration scenario only.
Many businesses have service level agreements (SLAs) that dictate exactly how much downtime is allowed for a specific server or platform. A good migration tool gives you the flexibility to ensure these predetermined outcomes. Durability is especially urgent for mission-critical systems and data subject to regulatory compliance. A platform-dependent tool based on technology that introduces the risk of data loss into the migration process is simply not an option for these types of businesses.
Tip #3 – Fully plan out the migration
You can carefully plot your pre-migration steps but still end up with an unsuccessful migration if you don’t have a post-cutover plan. It’s not unusual for migrations to go smoothly at the beginning, only to fall apart later. For example, you may run into issues with the new platform that necessitate a return to the old one—but by that point, you may already have users on the new platform. You need a plan for returning to the original source without losing new data.
Tip #4 – Chunk workloads
Depending on the size of the migration project, organize server workloads into manageable groups based on criticality, difficulty, performance, etc. Figure out the organizational scheme and break projects down into logical units. Then attack those server groups accordingly.
Tip #5 – Perform fully functional tests
There’s no substitute for having a fully functional replica to test, with the entire server workload on the target platform. Fully functional tests are critical for consistently successful migrations.
Tip #6 – Know when to raise your hand
Cross-dependencies can be so complex that not even the system architect can discover all of them. If there aren’t enough resources to dedicate to migration, don’t bite off more than you can chew. The best route may be to hire professional services. A service provider with migration expertise can reduce the risk of a failed migration and may help control costs versus attempting to do it with insufficient resources.