Organizations are adding public cloud to their IT infrastructures for all the right reasons. Cloud helps organizations achieve operational efficiencies and provides the flexibility to scale capacity up or down quickly. Strategic use of cloud enables forward-thinking organizations to leapfrog competitors by transforming into digital leaders. COVID-19, and the business challenges it creates, has dramatically accelerated cloud adoption. Cloud is now firmly established as an essential ingredient of a modern IT backbone.
Your organization’s current cloud migration strategy might be producing results. But as a CEO or CFO, you might be concerned that it’s taking too long to get legacy applications up and running on cloud, especially applications that support critical processes and information your business depends on. As a CTO or CIO, you know why: migrating legacy applications poses special challenges. This paper examines those challenges and offers a solution for speeding up the cloud migration process.
Even though more than 90% of organizations are using at least one cloud service, just 23% of workloads are running on cloud. This means the vast majority of organizations using cloud are not realizing anywhere close to the value they could, especially when it comes to legacy applications. There are good reasons why.
All successful cloud migrations start with a readiness assessment of applications to determine their suitability for cloud. Legacy applications are hard to assess because they are often decades old, sit on outdated architectures, and rely on obsolete technologies. They’ve grown in size and complexity due to numerous additions and enhancements. The people who wrote them left the organization a long time ago, leaving little or no documentation. Further, many large organizations have hundreds or even thousands of legacy applications.
As part of the assessment, applications are evaluated to identify those with similar requirements that can be grouped into workloads for simpler and easier migration. A workload consists of applications and the underlying resources needed to make them operate efficiently, including compute power, networking, storage, and other factors. Creating workloads also poses challenges. One of the toughest tasks is identifying application dependencies so they can be taken into account. For example, one or more applications in a workload might have the need for consistent access to data housed in a centralized database. If that database is not available to the workload, the applications that rely on it will not work.
Given these challenges, the assessment phase alone of a cloud migration can take a year or more—significantly slowing down or even derailing an organization’s plans to modernize for the digital era. Consider the example of a leading European telco that spent more than a year analyzing its portfolio of almost 600 applications and their dependencies. By the time the company finished its assessment, the portfolio had changed significantly. The company had to start over assessing its application portfolio prior to a cloud migration.
Assessment isn’t the only challenge of migrating legacy applications. Most legacy applications will likely require some degree of restructuring and optimizing of their existing code to take better advantage of cloud attributes without changing their behavior. This is also time-consuming and expensive.
The time and expense associated with assessing and optimizing legacy applications prompt some organizations to delay moving them to cloud. Others sidestep moving them altogether by pursuing a cloud-native strategy. With a cloud-native strategy, only applications specifically written to take advantage of modern development practices, technologies, and cloud infrastructure are hosted on cloud. Cloud-native applications certainly have their place in an overall cloud strategy. The problem with pursuing a cloud-native strategy exclusively is that it limits organizations from participating fully in a world where it’s possible to run even the most complex legacy applications reliably, securely, and cost-effectively on cloud.
For organizations seeking digital transformation, the lost opportunity is too big a price to pay—and unnecessary because there are other options.
Organizations Must Evaluate and Choose a Migration Strategy Carefully
While organizations have migration options, they must choose carefully. The three most common approaches are rehost, rewrite/refactor, and replatform. When applied to legacy applications, all have limitations.
Rehost. Rehosting involves simply taking legacy applications and their associated data from on-site servers and placing them on cloud servers. This is a risky endeavor. Without any modifications at all, the applications will likely not be able to take advantage of cloud attributes such as scalability and resilience and therefore, perform sub-optimally.
Rewrite/refactor. Rewriting legacy applications for cloud is also risky. With potentially millions of lines of code that need to be re-written, the project could take many months or years and be very expensive.
Replatform. Replatforming is the process of restructuring the code of legacy applications to optimize them for cloud without changing their functional logic. This is more efficient than starting from scratch. However, replatforming typically takes a lot of time and skill. It’s also costly because it is mostly a manual process.
Replatforming with Automation Speeds Legacy Application Migration
Amaze® is an enterprise-level cloud adoption platform that automates most of the actions in replatforming. Amaze® is especially useful when applied to migrating legacy applications because it allows organizations to realize the benefits of cloud with far less risk than rehosting and in far less time and with less cost than rewriting/refactoring or replatforming manually.
The two main areas where Amaze® saves the most time are in first intelligently identifying the code blockers and inhibitors in legacy applications that will require modification and second automating the process of making the majority of changes. This is accomplished in the following way:
Initially, Amaze® performs an application analysis on the source code of all applications an organization seeks to migrate to cloud. The applications are then quickly containerized with minimal refactoring and deployed on the chosen cloud platform. On deployment, Amaze® generates an assessment report highlighting the time taken to identify and replatform the applications. The report also identifies the various technologies in the composition of the applications. Based on this report, a replatform option is chosen. As a final step, required code and design patterns are automatically generated and refactored to take advantage of cloud attributes without changing the behavior of the applications.
For migrating legacy applications, this high degree of automation reduces cloud migration time by up to 50% (see Figure 1), reduces cost by 50-60% compared to manual efforts, and increases developer productivity by up to 25% for changes that must be made manually.
Figure 1: Impact of Automation on Legacy Application Cloud Migration Time
Consider the example of a Fortune 100 financial services company seeking to curb the escalating costs of running a portfolio of legacy applications on proprietary licensed software and hardware. To reduce costs, the company decided to leverage open-source technology deployed on cloud. However, the company was unable to execute this project because the application portfolio had grown in functional complexity and technology debt over a period of several years. Further, the subject matter experts who had built these applications were no longer with the company and available documentation was inadequate. The company decided to engage with a replatforming partner to proceed.
After evaluating multiple competing solutions, the company chose Amaze® because of its ability to deliver valuable insights about the applications without documentation or subject matter experts and a high degree of automation as well as versatility and risk mitigation features. Most of the competing options promised to complete the project in 6 to 9 months. Amaze® helped the customer replatform its entire application portfolio in less than 10 weeks.
Amaze® Enables a Cloudify Everything® Strategy
There is a lot of buzz about cloud-first and cloud-smart strategies. With cloud-first, an organization commits to evaluating cloud as a solution first but leaves the door open to alternatives. While this allows for greater flexibility, it inevitably relegates legacy applications to the back of the line for cloud migration because of the greater challenges they pose. A cloud-smart strategy, such as the one adopted by the U.S. federal government in 2019, assumes that organizations must build knowledge, revise security and procurement policies, and upgrade workforce skill levels before attempting cloud migrations. These and other perceived barriers can stop a cloud program in its tracks.
The common flaw in most cloud strategies is that they focus on what is not possible instead of what is possible. As a result, organizations deploying these strategies tend to define cloud opportunity too narrowly and put artificial constraints on cloud adoption.
In contrast, Amaze® helps organizations pursue a Cloudify Everything® strategy by significantly expanding cloud adoption possibilities. Amaze® brings a business-aligned, accelerated, predictable, and outcome-driven approach that facilitates migrating applications to cloud—including business-critical legacy applications—at the fast pace of innovation organizations need in the digital era.
With Amaze®, organizations can embrace the possibilities of cloud for digital leadership and competitive advantage through automation. Automation enables the speed and scale required to maximize the cost and time savings and developer productivity improvements possible with cloud.
Moving to Cloud is More Than a Technology Transformation
Cloud equals capability. Organizations that do not fully embrace the power of cloud will see their competitors pulling away and the gap will continue to widen very rapidly. Further, an organization’s technical leaders need to drive cloud adoption. However, given the magnitude of organizational and culture change required to fully realize the benefits of cloud, business leaders should engage and partner with technical leaders every step of the way. Instead of focusing cloud investments on today’s tactical needs or what is considered possible today, organizations would be better served by adopting a holistic view of an organization’s commitment to cloud and a strategy of funding critical infrastructure investments—people, processes, and technology—that will enable continual expansion and improvements far into the future.
Glossary of Terms (in order of reference)
Public cloud is a scalable computing service offered by third-party providers over the internet. It may be free or available on-demand and gives customers the option to pay per usage for the CPU cycles, storage, and bandwidth they consume.
Legacy applications refer to the software applications organizations have that were written years ago and are based on outdated technologies. Some of the applications support critical processes and information the organizations depend on. These applications might need to be upgraded or rewritten before being moved to cloud for optimum cloud performance.
Cloud migration is the process of assessing, preparing and moving data and applications from on-site data centers to cloud or from one cloud to another.
A workload consists of an application or multiple applications and the underlying resources needed to make them operate efficiently, including compute power, networking, storage. A workload is classified based on its architecture, resource requirements, and other factors. Workloads are containerized before migration to cloud so they can be moved around quickly and easily.
Cloud-native is an approach to building applications that leverages the advantages of the cloud computing model. Cloud-native applications and services can be deployed and managed efficiently as they are specifically designed for the cloud from day one.
Also known as “lift and shift,” rehosting refers to moving servers and applications as-is from the current on-site environment to cloud. This approach has the lowest upfront costs and requires less time and fewer resources.
Refactor / Rewrite
Refactoring involves completely re-evaluating and re-architecting an application to make it more suited for cloud. The application code is modified to leverage cloud capabilities and the overall application is optimized for higher performance.
Also known as the “lift-tinker-and-shift” approach to cloud migration, re-platforming can involve code modifications to modernize and/or add new features to legacy applications. It can be a cost-effective approach when automation is applied.
A cloud-first strategy is an organizational commitment to cloud adoption as a key driver of business transformation. The possibilities of cloud are factored in and evaluated as a solution first before looking at other alternatives.
A cloud-smart strategy focuses on maximizing cloud value by ensuring thorough preparedness before the migration journey. It is aimed at providing definitive guidance on security, procurement policies, and workforce skills for cloud adoption.
Cloudify Everything® is Hexaware’s outcome-driven strategy to enable organizations to reap the full benefits of cloud transformation by significantly expanding cloud adoption possibilities.
Cloudify Everything involves modernizing and transforming the traditional IT ecosystem for speed and scale, resulting in winning outcomes on cloud.