Step 3 in Planning your Cloud Migration: Execute without Disruption
As we’ve been discussing in this 3-part blog series, an enterprise-scale cloud migration offers transformative benefits—but getting there can be tricky.
In Step 1, we talked about making sure inaccurate data doesn’t get in your way. Step 2 advised that it’s critical to know your apps and workloads (from your elbows). In this post, we’re ready to dive into execution.
Once you have a clear, accurate and actionable view of your environment, it’s time to execute your migration plan.
The main challenge here, of course, is how to carry out your migration with minimal disruption to normal operations, at the lowest cost, and over the shortest period of time.
- 44% of government organizations have experienced some form of IT failure
- Gartner estimates that the average cost of downtime can top $5,600 PER MINUTE, which adds up to $7.2 million PER HOUR
Carefully planned migration events can minimize or eliminate human errors while significantly shortening outage windows.
Use Runbooks
We advise using step-by-step “runbooks” to ensure that proper shutdown, migration and startup procedures are properly sequenced. Traditionally, runbooks were manually created, using spreadsheets and MS Project. But this process could mean that essential steps get missed, tasks are performed out of order, which could lead to corrupting databases, falling out of regulatory compliance, or worse.
If you have the in-depth understanding of your environment, including the interdependencies between assets, applications and storage along with rules for sequencing, the next step is to generate runbooks.
Our migration experts believe this is so critical that we made sure our TransitionManager software was able to automatically generate even the most complex runbooks in just seconds. And runbook templates, which we called “recipes” can be developed and archived in the system so that you have a record and can repeat them on every project in the future.
Find a Better Way to Communicate
When you’re ready to execute the plan, you’ll gain great efficiencies if you can establish a centralized virtual platform as your command center. There you should keep the multitude of tasks, assignments, and sequencing. We find that having this central execution hub is much more productive and effective than a serial chain of e-mailed tasks or countless “war room” meetings. Real-time communications and collaboration are critical to your success.
Don’t Rely on Automation Alone
Automation is powerful but tools must integrate together and work as an orchestrated toolchain if you want to ensure the correct movement and sequence of applications, infrastructure and workloads.
Potential pitfalls of Automation:
- Automated transport tools that operate without awareness of a workload’s dependencies.
- Reliance on automation without human analysis and planning to instruct tools how to execute
- Automation without orchestration may introduce error when moving complex applications if tasks are done out of sequence
- Neglecting compliance. Compliance never ceases and security is a large barrier for many organizations when it comes to migrations.
If you’ve gathered accurate data, have a clear and actionable view of your environment, and have a game plan that will orchestrate both human and automated tasks, you can execute with confidence that you won’t interrupt key workstreams, risk vital data, or deal with costly delays.
Read our recent case study how Happy State Bank launched a major initiative to consolidate different databases, refresh their infrastructure and build a resilient IT environment that would enable them to execute a switchover plan without disrupting services to customers.
Related posts: