Case StudyRe-Host: Cloud Migration
Re-Host: Cloud Migration
Cloud migration is pivotal for modern businesses seeking agility and scalability. Our recent project, involving the migration of nearly 70 servers from a traditional data center to the cloud, serves as an exemplary model in this domain. This case study delves into our bespoke approach, combining the AWS Migration Acceleration Program (MAP) with out custom automation, and details the technical intricacies of our process.
Our client’s data center comprised a complex mix of about ~70 servers, hosting varied workloads and applications. The challenge was to migrate these to the AWS cloud without impacting their operational efficiency and ensuring data integrity.
1. Resource Discovery and Assessment
We initiated the project with an extensive discovery phase, using AWS Application Discovery Service (ADS) and AWS Migration Hub. These tools helped us gather detailed information about the on-premises infrastructure, including server utilization, network dependencies, and application configurations. We leveraged the Agentless collector to gather information from VMWare vCenter. We also collected usage data for databases running in the data center. We need data over a period of time to get a decent sample size of the workloads on each server. Depending on the type of workload, we might have to run the collector during specific events like end-of-month financial processing or peak customer activity. Or we can use historical information to extrapolate the peak usage.
2. Workload Classification and grouping:
Network visualization available as part of ADS helped us with identifying workload dependencies. We were able to quickly group the servers into Application groupings.
3. Workload classification: 7 R’s:
Working with the client Product, Dev, and Operations team, we classified the workloads as per AWS 7 R’s. We identified 3 representative workloads for the initial phase.
4. TCO and Business case:
We used the Migration Portfolio Analysis (MPA) tool to generate a TCO analysis and projected cost savings over a 3-year period. We exported the data and prepared a Migration Business case to seek funding approval.
Using our Cloud Enablement tool (qTOOLS) we vended AWS accounts with predefined security and operations best practices. We created a shared network account to centralize routing for east/west and north/south traffic. A Shared Services account was created to host common services like DevOps tools etc.
Terraform was the framework of choice for Infrastructure as Code (IaC). Pulling from our vast library of modules, we were able to quickly compile the IaC solution to do a Lift and Shift (Re-hosting) of virtual machines, and setup Database Migration Service for moving an Oracle database to RDS Postgres. CICD pipelines were created with minimal effort using templates provided by qOPS, our cloud operations accelerator.
7. Monitoring and Observability:
We integrated CloudWatch with clients existing monitoring tools for log aggregation and altering. We later enhanced the solution by implementing observability tools to gather and report on metrics.
After the success of the first phase, we created a repeatable process and customizable CICD pipeline templates to migrate the other workloads which were targetted for Rehosting.
9. Custom Solutions for Complex Workloads
For more complex workloads that required refactoring or re-platforming, we developed custom solutions. This involved modifying applications to leverage cloud-native services. These were handled on separate work orders.
Efficiency in Execution: The integration of AWS MAP guidelines with qTOOLS automation allowed for a streamlined and efficient migration process, significantly reducing manual efforts and potential human errors.
Enhanced Performance and Scalability: Post-migration, the client experienced improved performance, with the scalable AWS environment adeptly handling varying workloads and traffic.
Cost Optimization: Our approach enabled the client to optimize costs by right-sizing resources and utilizing AWS’s pay-as-you-go pricing model.
Robust Security and Compliance: By adhering to AWS’s well-architected framework, we ensured that the migrated infrastructure met rigorous security and compliance standards.