For the last several weeks, enterprises have battled the ripple effects of the pandemic. This often meant having employees work at home when it was once not allowed. Cloud computing uses that were still being debated in the halls of corporate IT before the crisis suddenly became reality. One clear theme: The enterprise must rely on public cloud-based platforms and services to ensure its future in the new normal.
With that theme in mind, here are a few steps to implement a successful transition:
Step 1: Realistically assess the existing problems and most probable near-future damages
From a business perspective, we need to identify and understand the negative effects that occurred during the lockdown. What additional damage will likely occur in the short and long terms? This can range from relatively minor problems, such as a slowdown of some customer deliveries or lack of materials for manufacturing, to a complete shutdown of some operations due to on-premises systems that could not be maintained or fixed during the lockdown.
You need to assign dollar amounts to each issue. Keep in mind that some of these will be hard costs, meaning sales and billing. Others will be soft costs, such as reputation. What points hurt the business the most? We need this information to prioritize triage.
Step 2: Determine the short-term and long-term changes required
For most enterprises, this step will immediately identify the need to migrate some assets to cloud. The migration will typically target existing on-premises systems that managed to limp through the crisis. Based on historical migration data, the most common move will involve a “lift and shift” of resources, such as storage and compute, to a public cloud provider. Most enterprises will opt to refactor the applications at a later date; a few will refactor as the applications migrate.
Migrations will include quick fixes to remove vulnerabilities discovered during the crisis that might still be occurring. Don’t lose sight of new issues that can arise during a migration that can affect long-term operations. In a crisis, it’s easier to focus on the flames than the smoldering pile of embers, but both the flames and embers need equal attention to get the fire under control.
The danger here is that some companies will move too fast and be reckless with their cloud selections and other immediate decisions. The flame solution is not necessarily the ember solution. I suspect that many enterprises will migrate twice: once to a failure and once to a success.
Step 3: Determine changes to operations
The key to success is the ability to operationalize all near- and longer-term changes. If you think about businesses that are suffering, it’s likely related to shortcomings in business continuity and disaster recovery planning that never anticipated the removal of humans from data centers.
As we recover from this crisis, ops planning takes on a whole new level of importance. Yes, ad hoc ops changes need to be implemented right now. However, don’t delay planning for longer-term changes to the operational run books that should likely include the addition of public cloud resources.
Although I’ve listed ops planning last, this step is the most important of the three. Make ops planning a priority.
The road forward will be difficult for most companies, no matter how well prepared they were—or thought they were. With a bit of positive planning and diligent execution, we’ll put this crisis behind us.