Ready to “cloudify” your apps? Watch the webinar replay here.
Most companies we talk to are not sure of the best way to migrate mission critical legacy apps to the cloud.
First, we should define mission critical legacy apps (MCLAs): these are business applications like Enterprise Resource Planning (ERP) for manufacturers, Electronic Medical Records (EMR) for healthcare providers, and accounting systems for most organizations. These apps typically have a huge impact on delivering products and services, tracking revenue, profitability, managing client relationships, and managing finances. They are often called the core applications of the business.
From an IT perspective, mission critical legacy apps have a few characteristics that make them tricky to move to the cloud:
- MCLAs typically are cloud-unaware because they weren’t born in the cloud. MCLAs are dumb apps on smart hardware. These apps were designed assuming resiliency in the underlying hardware including server, network, and storage failsafes to automatically reduce the risk of an application outage.
- Clouds are different. Some clouds deliver this underlying system resiliency with zero changes to the apps. Other clouds, including the hyperscale clouds like AWS, Azure, and Google Cloud Platform, expect your IT organization to build automation to mimic this smart hardware resiliency, using complex failure plans, automation, and scripting instead.
- Because these MCLAs weren’t born in the cloud, they also weren’t designed with cloud pricing models in mind. Some “chatty” applications generate so much network traffic that they can be cost-prohibitive to maintain on clouds that charge tolls for network traffic like AWS, Azure, and Google.
- MCLAs are often performance-sensitive as well. For decades, IT has thrown high-speed hardware at these applications to make sure that they are fast enough for the end-users to get their work done. Third-party cloud comparisons reveal many of these performance differences across clouds, especially in storage and network performance.
MCLAs are the heartbeat of your organization. But choosing the wrong cloud for MCLAs can be expensive and difficult. What are other companies doing to migrate MCLAs to the cloud?
Some opt for refactoring the application. They either wait for the software vendor to deliver the MCLA from the cloud, or the organization makes software changes itself. The organizations typically make a quantitative decision based on the time and effort required to refactor, to wait for a cloud-native version, or to do it themselves. If no additional features are to be added in a refactor or replatform project, most organizations do not choose this option.
Lift and shift to hyperscale
Some organizations opt to minimize changes to the application using a lift and shift strategy to move the formerly in-house or colocated MCLA to a hyperscale cloud like AWS, Azure, or Google. Unfortunately, performance-sensitive MCLAs, MCLAs that expect underlying hardware resilience, or a lack of hyperscale skillsets among the in-house IT staff can severely limit the migration success. Typical symptoms are high monthly costs, cost variability over time, and application speed issues. Due diligence can prevent these problems, but many organizations skip due diligence thinking they can’t go wrong with cloud brands like Amazon, Microsoft, and Google.
Lift and shift to enterprise cloud
Many are finding success using Enterprise Cloud as part of a sensible multi-cloud strategy to solve MCLA challenges. An Enterprise cloud is a VMware-based cloud designed for speed and resiliency plus multi-cloud capabilities, all using tools familiar to the in-house IT staff and managed service providers.
Why do Enterprise Clouds often fit better for MCLAs?
- Full hardware resiliency with multiple backups and no single points of failure
- VMware-based tooling, familiar because most in-house IT staff have been using it for years
- Designed with application speed in mind
- Zero changes to MCLAs
So what will you do to bring your mission critical legacy apps into the cloud? You could undertake a lengthy and costly refactor or replatform process. You could take a chance on using hyperscale that may or may not fit your requirements. Or, you could move your MCLAs to the Enterprise Cloud and enjoy the benefits of infinite scale, lower staff workload demands, predictable costs, and lower risk.
Watch the webinar replay here: