How to Navigate Cloud Migration Risks to Ensure Success

Phonlamai Photo/

First, think about workload profiles.

The cloud may be one of the largest technology growth markets in recent history.  

Gartner predicts that public cloud services will grow from $260.2 billion in 2017 to $411.4 billion in 2020—a compound annual growth rate of 16.5 percent. The firm went on to say that, by 2021, 28 percent of all IT spending will be for services related to cloud-based infrastructure, middleware, applications and business processes. Gartner foresees double-digit growth in government use of public cloud services, with spending forecast to grow on average 17.1 percent per year through 2021. The firm says that almost half of government organizations today are actively using cloud services.  

But even with the traction that agencies have achieved in cloud migration, organizations have gone through starts, stops and reverses (repatriating workloads from the cloud back to the on-premise data center). That’s arguably because, once the workload was in production in the cloud, that decision to “migrate” a workload from on-premises to the cloud turned out to lack some combination of security, performance, reliability, manageability or cost effectiveness that was originally expected.

This can happen for several reasons:

  • Workloads were migrated to the cloud without a detailed understanding of the performance the agency could expect, and performance in production turned out to be unacceptable.
  • Over-provisioned cloud infrastructure reduced or eliminated the expected cost savings.
  • Workload dependencies were not well understood prior to migration, causing performance or security issues.

Unfortunately, the cost and lost time from a failure can significantly set back budgets and timelines for meeting the security, cost, performance, reliability and manageability requirements for an application or workload.

Evolution of Cloud Adoption  

In September, the White House released a draft of its cloud computing strategy for public comment entitled “Cloud Smart,” which will refocus the federal government’s adoption efforts. The new strategy updates the Obama administration’s “Cloud First” policy, established in 2010, to better reflect where agencies and the technology are today. While the 2010 policy asserted the potential benefits of cloud, Cloud Smart will stress mission outcomes.

Initiatives like Cloud Smart are sorely needed: in 2016, Nextgov cited a report that said 44 percent of organizations had experienced some form of IT migration failure. The report further stated that complex IT migrations could fail when a single weak link—for example, hardware, legacy software and applications, or a poor choice of cloud vendors—dooms the migration.

Government agencies today are beginning to realize that gaining an understanding of “workload profiles” and infrastructure performance requirements is key to determining how, or if, a workload is a good candidate for migration to the cloud or best kept on-premise. Here’s how:

  • They can identify and incorporate workload dependencies into their migration plan.
  • They can create and interpret “workload profiles” that represent their production applications
  • They can turn the workload profiles into highly realistic simulations.
  • They can run those simulated workloads in any of the public cloud providers or even on other on-premise infrastructure and use the resulting information to identify the suitability of any application or workload for the cloud.
  • Most importantly, they can identify the ideal workload cloud configuration and optimize costs while ensuring acceptable performance—all prior to migration.

What Agencies Must Do First

Today, regardless of the stage of cloud migration they are in, agencies must take steps to optimize the results of new workload migration efforts.  They must:

  • Understand which questions to ask to determine an application’s or workload’s cloud suitability.
  • Understand the best practices to adopt to de-risk cloud migrations.
  • Learn how available technology can help to de-risk the process through automation of application infrastructure discovery, dependency mapping, workload profiling and cloud performance testing—all supported by sophisticated analytics.

The Modernizing Government Technology Act, enacted last December, established an IT Working Capital Fund that would be devoted to achieving five objectives. Encouragingly, one of those objectives (Item B) specified transitioning “legacy information technology systems to commercial cloud computing and other innovative commercial platforms and technologies, including those serving more than one covered agency with common requirements.”

The MGT Act and the Cloud Smart initiative should help spur cloud migrations in the government sector, but agencies need innovative tools to put the projects in motion. With technology available for simulating and predicting the results of cloud migrations, agencies are now in a much better position to determine which “cloud candidates” will produce the best cost/benefit once they move into a production cloud environment—which is one of the ultimate goals of cloud migrations.  

Rick Haggart is senior vice president of professional services for Virtual Instruments.