As your business evolves, the data center can expand unevenly, allowing legacy and new technologies to coexist. It might also grow in very dissimilar ways through mergers and acquisitions. As a result, your IT staff supports an heterogeneous mix of hardware architectures, operating systems, and applications. The workload diversity can increase the stress on your IT staff as well as the likelihood of human error. You want to move to more homogeneous workloads that can better meet your future business needs, simplify daily operations, reduce operational costs, and reduce risk in your IT environment.
The purpose of any workload transformation or migration is to change workloads from their current modes of operation to appropriate future modes of operation. How you achieve the change depends on the types of workloads you manage and your business needs. Typical project objectives include the following:
Migrate workloads between physical, virtual, and cloud infrastructures.
Upgrade workloads to newer hardware, different hardware vendors, or hosted provider hardware.
Consolidate workloads on virtualization host servers or to cloud infrastructures.
Lift and shift equipment from location A to location B.
Move virtual files to newer virtualization host servers, running the same or different virtualization hypervisors.
Decommission old workloads as you retire software and services.
Your transformation or migration project might be a combination of any of these goals, or thousands of instances of the same one.
Business takes priority. Transformations require minimal downtime for mission critical applications and services. Each workload transformation has different priorities and windows of opportunity based on business demands. Schedules must consider the availability of target facilities, network resources, equipment, and the IT staff needed to plan and execute the transformation. Management and organization stakeholders want to track the progress and status of your projects.
Defining the original state of a workload can be tedious. You create a profile of the workload that includes information about its compute infrastructure, operating system, applications, data, and configuration. Because workloads might be upgraded or repurposed over time, the profile might need to be augmented or updated before you execute the transformation. The related proposed workload profile might also need to change as appropriate to the revisions to the original workload.
Large-scale IT transformation projects typically occur over an extended period in a production environment that might span multiple locations. Complex projects with massive numbers of workloads might take months or even years to complete. It might be possible to plan details only a few months in advance. Projects require multiple phases, not a one-time effort.