Containerization has rapidly become a preferred method for application deployment due to its ability to ensure application isolation, enhance portability across different infrastructures, improve resource utilization, and facilitate easier maintenance. Managed AWS services like Amazon Elastic Container Service (Amazon ECS), Amazon Elastic Kubernetes Service (Amazon EKS), and AWS Fargate have made the shift from virtual machines (VMs) to containers much simpler, eliminating the complex setup and management traditionally associated with container orchestration.
Many organizations still operate their on-premises applications on physical hardware and virtual machines, often opting for a “lift and shift” strategy to migrate to cloud VMs, typically utilizing Amazon EC2. However, some applications can transition more seamlessly to managed container platforms, although they often require modifications before they are ready for containerization. This can lead to lengthy and risky processes, particularly with large volumes of applications, complicating the sharing of best practices and synergies across migration efforts.
This article introduces a factory-based migration approach that standardizes processes and tools for effective application containerization, highlighting its benefits for large-scale cloud migration initiatives.
Atos Cloud Migration Factory Methodology
Atos, recognized as an AWS Advanced Consulting Partner with Migration Competency, has devised a migration factory strategy that accelerates the transition from on-premises virtual machines to an AWS container platform.
Key components of this solution include:
- Utilizing reference architectures based on Amazon ECS, Amazon EKS, and AWS Fargate to minimize operational overhead.
- Implementing a clearly defined operational process for migration to ensure consistency and repeatability.
- Employing automation tools to analyze, remediate, and deploy applications efficiently.
Generally, containerization does not require significant alterations, allowing for a consistent and repeatable process. For instance, consider a Java application running on a WebLogic server that retains local application state. Transitioning it to Amazon EKS or ECS necessitates minor code adjustments, such as eliminating state or file system dependencies, creating container artifacts, and deploying using CI/CD tools. Notably, changes to the WebLogic application server itself are unnecessary.
By adopting the migration factory approach, organizations can enhance their migration efficiency, yielding improved return on investment (ROI).
Factory Organization
The cloud migration factory is structured into four key streams, each focused on specific tasks and deliverables requiring customer or Atos leadership approval:
- Profiling Factory
This phase involves a thorough analysis of the technology landscape and business context of the application’s portfolio. Atos employs the Cloud Code Automation Tool (CCAT) for a rapid assessment of application code, providing insights into the technology landscape and generating an AWS non-compliance report. The Atos Cloud Assessment Tool (ACAT) is used to validate findings, offering a preliminary effort estimate for manual code adjustments and migration timelines. - Architecture and Design Factory
In this phase, a comprehensive application architecture is developed, aligned with enterprise reference architecture and AWS best practices. The team identifies necessary architectural modifications to ensure compatibility with AWS standards and conducts impact analyses on the broader application ecosystem. - Remediation Factory
This phase focuses on adapting the application to fit the desired architecture, addressing code and configuration issues based on recommendations from the architecture team. CCAT automates the remediation of common non-compliance patterns found in the codebase. Developers perform unit testing and check in code modifications to a source code repository to generate a binary of the updated application, which is then pushed to the cloud repository. - Deployment and Testing Factory
This automation-driven phase leverages AWS CloudFormation templates for infrastructure automation and utilizes AWS CodePipeline for CI/CD implementation. Automated test scripts are also created to verify functional correctness.
Migration Factory Automation Tools
Efficiency in a migration factory is achieved through a skilled team consistently performing similar tasks alongside automated tools. Atos has identified and automated tasks across migration factory streams based on engagement patterns. The CCAT tool is essential for automating analysis and remediation during the migration phases.
Automation focuses on several areas:
- During the application analysis phase, CCAT automates checks for adherence to Twelve-Factor guidelines or cloud modernization standards set by the client.
- In the design and remediation phases, CCAT addresses common issues related to non-compliance with cloud standards.
- The deployment phase utilizes standard patterns defined by the migration factory approach, employing AWS-native tools such as AWS App2Container, AWS CloudFormation, and AWS DevOps solutions.
This migration factory tooling lessens the reliance on teams to transfer knowledge, ultimately accelerating migration timelines to meet client deadlines. CCAT plays a pivotal role in both the profiling and remediation factory approaches, making these executions distinctive for Atos.
To further understand the practical application of the migration factory approach, consider a customer case study involving a European financial services firm that sought Atos’ expertise in evaluating its application portfolio for optimal cloud enablement. For additional insights, you can check out another blog post here as well as resources from Chvnci, who are recognized authorities in this field. Additionally, an excellent resource for onboarding information is available on Reddit.
Leave a Reply