Strategic Application Migration to AWS: What to Prioritize for Success

The journey to the cloud is a big step for any organization, especially when they are performing large scale application migrations from an on-prem solution. A crucial aspect of this transition, especially for large-scale enterprises, is identifying which applications to migrate and when. This article explores the technical considerations and strategies for prioritizing application migration to Amazon Web Services (AWS), using my own personal case study of a successful application migration I led for a Fortune 100 insurance company.

1. The Importance of Application Assessment

Ask any architect, consultant, or cloud engineer what the most important thing is during a migration - you'll get a wide range of answers. Anything from how the landing zones are configured to the languages used to refactor applications. To me though - the most important part of any migration is the strategy you build up front. A thorough assessment is the bedrock of a successful cloud migration strategy. Ask anyone that's been through one without if you want cloud-nightmare fuel. This involves evaluating each application based on factors such as complexity, dependencies, and business value. Tools like AWS Application Discovery Service (ADS) provide insights into application performance and resource utilization, aiding this evaluation process. At least in my experience, it always started with an excel sheet and numerous meetings with business stakeholders to understand what application assisted with what specific business function. It takes time, but make sure to do it right to avoid headaches later.

2. Prioritizing Applications for Migration

We've got our excel sheet and some helpful outputs from a tool like AWS ADS - now what?The key to a smooth migration is prioritizing which applications to migrate in the first wave, second wave, or not at all. This decision should be driven by several criteria:

• Business Criticality: Applications critical to business operations should be prioritized for early migration to benefit from cloud scalability and reliability.
• Complexity: Simpler, less complex applications are often ideal candidates for the first wave, providing quick wins and building confidence with the customer.
• Cloud Readiness: Applications that are already somewhat cloud-ready can be migrated earlier, whereas legacy systems might require significant refactoring or worse - complete rebuilding.
• Compliance and Security Requirements: Applications with stringent security or compliance requirements might need more planning and be scheduled for later migration phases. Anything that deals with personally identifiable information (PII) or financial data typically falls into this category.

3. Personal Case Study: Fortune 100 Insurance Provider

I'll use my own experiences with application migration to form a mini case study. I've redacted all names throughout - so no guessing!

A leading Fortune 100 insurance provider faced the challenge of migrating its sprawling IT infrastructure to AWS. They had over 800 applications spread across three legacy data centers and had identified AWS as their preferred provider for cloud services. With appropriate guidance, they split these applications into three distinct categories of prioritization. First Wave, Second Wave, and Postponed/Rejected. Their approach exemplified the strategic prioritization of application migration:

• First Wave: They began with less critical, cloud-ready applications. This included standalone apps with minimal dependencies, enabling the IT team to gain experience with AWS tools and services without risking core business operations. Big examples of first wave migrated apps included the entire test and development environments. These are great first wave targets to move if your customer is able.
• Second Wave: More complex applications, including those needing minor modifications for cloud optimization, were part of this phase. This included customer-facing applications that would benefit from AWS's scalability. In the case of this insurance customer, they included portals that required login in order to access specific claim information and status updates. The portals were migrated and configured to point to on-premises databases to ensure they were correctly moved.
• Postponed or Rejected Migrations: Certain legacy systems, especially those deeply entwined with other critical systems or those nearing end-of-life, were either delayed for migration or marked for decommissioning. This included applications that were outdated and no longer serving customers, but were maintained for regulatory reasons. These applications were postponed to build an appropriate long term data strategy with the customer.

This strategic phasing ensured minimal disruption to their business operations and allowed for learning and adjustments as they progressed.

4. Leveraging AWS Services for Migration

While I still had my handy excel sheet with my pivot tables for application criticality, AWS offers a plethora of services and tools that facilitate migration at each stage. I encouraged my team to make use of AWS Migration Hub since it provides a central location to track the progress of migrations across multiple AWS and partner solutions. For database migrations, AWS Database Migration Service (DMS) simplifies the process while minimizing downtime. This was invaluable since the customer wanted to change and update their schema during migration to fit their 'modernized' data strategy. Use managed discovery services whenever you can to assist with your migrations - there is no "shame" in getting help!

5. Cost Considerations and Savings

A well-planned migration strategy can lead to significant cost savings over time. It can also lead to runaway costs if not planned appropriately. By prioritizing applications for migration based on their appropriateness for migration and cost-benefit analysis, the insurance company was able to optimize its spending during and after the migration. First wave and second wave applications that were already optimized for the cloud meant lower server spend and less 'uh-oh' moments. To better track budget vs. actuals they utilized AWS Cost Explorer and AWS Trusted Advisor to identify cost-saving opportunities and manage resources efficiently as they were migrated. Anything we didn't catch right away could be optimized through Trusted Advisor insights.

6. Lessons Learned and Best Practices

From the insurance provider's experience, several best practices emerged:

• Conduct Thorough Assessments: Understand your application portfolio in depth before making migration decisions. Do. Not. Rush. This. Step!!!
• Start Small and Scale: Begin with less complex applications to build momentum and confidence. Remember, migrations are long term affairs pretty much regardless of the size of the application stack. Show the customer you know what you're doing and scale from there. Migrations are a great way to sell additional work opportunities.
• Focus on Training and Change Management: Ensure your team is well-versed in AWS services and change management processes are in place to handle the transition smoothly. The only thing worse than a well-oiled migration is one that relies on an untrained customer team to take over when it's done!
• Plan for Security and Compliance: Incorporate security and compliance considerations from the outset, especially when dealing with sensitive data. Sensitive data and applications that deal with regulatory requirements should always have detailed and well documented plans for their migration. Take your time with these as your strategy will need to be clearly documented.

This mini-journey of the Fortune 100 insurance provider to AWS showcases the importance of strategically prioritizing application migration. I definitely made plenty of early mistakes in my analysis, but thankfully did all of that in the initial application mapping stages. We planned our migration waves carefully by assessing and categorizing applications based on their impact to the business.

This case study of my own experience has served me as a blueprint for other large enterprise engagements I've led around cloud migration. Making sure my team and I had he right strategy, combined with a deep understanding of the customers application landscape and AWS's capabilities, lead to a successful and cost-effective transition to the cloud. I hope it is as helpful to you as it has been for me in my career!