For most organizations, the cloud is a promise for better outcomes. Let’s be honest, why would you go through all that effort otherwise? This is especially true for all those applications and workloads that are currently running in data center environments but need to be moved to the cloud to support digital transformation efforts. And yet, in several studies, McKinsey has found that:
- 60% of organizations have migrated less than 10% of their workloads. (source)
- Only 16% of digital transformation efforts have successfully improved performance and equipped the organizations to sustain changes in the long term. (source)
In short, businesses are struggling to migrate to the cloud, and when they get there, they’re struggling to achieve the desired outcomes. This state of affairs is unsustainable. So, what’s the problem?
Let’s consider the migration paths—and there are several. Two approaches scrap everything you have on-premises and start fresh:
- Rebuild by rewriting the application from scratch to be cloud-native.
- Replace it with a new cloud-native application.
And there are three methods which retain some level of the existing software:
- Rehost the application by simply moving the entire thing as is to the cloud (also called “lift and shift”).
- Refactor it by rearchitecting parts of the application so it better supports the cloud environment.
- Replatform it by moving it with minimal upgrades to take advantage of cloud benefits.
A study by 451 Research found that 56% of organizations are focusing on these last three options (rehost, refactor, replatform)—far more than the 31% taking a cloud-first approach and the 13% who aren’t deploying any workloads off-premises (perhaps we should call that option remain).
Whether they’re rehosting, refactoring, or replatforming, over half of enterprises are literally moving workloads from the data center to the cloud. The difference between their migrations is simply how much, if any, of those applications are modified in the process. This could mean that unless workload performance and dependencies are taken into account, how those applications operate on-premises will affect how they run in the cloud.
What happens in the data center doesn’t stay in the data center
Long gone are the days of siloed applications. Data and services are shared, often extensively, creating a complex web of interdependencies. What you have in the data center, including compute, networking, and storage elements—and how they are all interconnected—is critical information you need to understand before you can even begin to think about moving anything into the cloud. The philosophy of “move fast and break things” does not apply to moving your mission-critical applications to the cloud, with often a lot more time spent up front in the planning phases. Simply “not breaking” is a low bar, however, and not enough to justify the effort and expense of migration. Once in the cloud, your workloads need to perform to expectations. And that’s where baselines come in.
The role of baselines in cloud migrations
A baseline assessment provides you with detailed information about the health, utilization, and performance characteristics of your on-premises infrastructure. This is critical for two reasons.
First, it identifies any potential issues you have that need to be addressed prior to migration. You certainly don’t want any sins in your data center to be replicated in the cloud, but you also don’t want to move forward with faulty assumptions. For example, if you have ongoing performance issues such as CPU contention or memory pressure due to oversubscribed hosts or virtual machines, you might presume that these performance-affecting conditions will go away once the workload is running in a scalable cloud. And they could, but you could also see a big cost jump as a result. Wouldn’t you want to identify that and adjust your plan accordingly—before you incur the surprise expense?
Second, the baseline provides you with a reference point for comparing utilization and performance in the cloud. It should reflect any seasonality in the targeted workloads to ensure your post-migration deployment delivers on performance and other KPIs to support the enterprise’s objectives.
The bottom line: A baseline gives you the information you need to understand how the workloads will operate in the cloud so you can make more informed decisions about your cloud configurations, and so you can actually quantify what “better” means for your transformation initiative.
Know before you go—with Virtana Workload Placement
Virtana Workload Placement delivers the application discovery and mapping capabilities you need to capture existing applications and infrastructure topologies along with key performance and operational metrics. With baselines built from high-fidelity data, you’ll fully understand all your application workloads before migration. Try it for free
Randy Randhawa
