How ABFL modernized Microsoft applications on AWS to drive efficiency and agility

In this blog post, you will learn how Aditya Birla Finance Limited (ABFL) adopted a progressive and architecture-driven approach to modernize their legacy Windows-based wealth management platform, Aditya Birla MyUniverse ABMU, and achieve 4X total cost of ownership (TCO) savings.

Introduction

ABFL, headquartered in Mumbai, India, offers customized solutions in the areas of personal finance, mortgage finance, SME finance, corporate finance, and wealth management. Its business-to-consumer digital wealth management platform was expected to support growing wealth product portfolio needs, as well as scale with a growing customer base. Legacy infrastructure and technology supporting the wealth platform was impacting business continuity and customer experience. A technology refresh was needed.

ABFL chose Amazon Web Services (AWS) to migrate and modernize its wealth management platform. By using a modern serverless architecture, a managed open-source platform, and prescriptive guidance, ABFL was able to transform a legacy Windows stack platform into a true modern platform offering unmatched agility, performance, and scale at a much lower cost.

To position ABFL for future growth in the wealth management sector, it needed to address immediate platform challenges, as well as improve overall agility, scale, and customer experience. Some of ABFL’s challenges included:

• Technology obsolescence – Legacy hardware and software reaching end-of-life (EOL) were posing security and performance challenges, limiting its ability to scale effectively and threatening business continuity.
• Cost overruns – Rising costs of maintaining legacy hardware, software licenses, indirect costs of energy consumption, manual oversight, and specialized IT personnel were unsustainable.
• Agility issues – The inability to scale the application infrastructure dynamically as business grew and demand fluctuated was a major pain point. Providing more data processing capacity required procuring hardware and reconfiguring existing setups, which was a time-consuming and inflexible process that bottlenecked operations.
• Operational complexities – Managing a myriad of proprietary software licenses was becoming increasingly complex and time-consuming. Outdated tools in managing, monitoring, and securing infrastructure required manual intervention without real-time insights. This complexity diverted valuable resources from core business activities.

In an industry driven by rapid technological advancements, the lack of agility and inability to adopt new services was a critical disadvantage. These challenges ultimately led to a decision for modernization. ABFL started looking for a partner who not only provided a robust, secure, and reliable platform, but more importantly, would work with the ABFL team to build a modernization roadmap and be a part of its transformation journey. That is where ABFL engaged with AWS to begin this journey.

Modernization approach

The ABFL team decided on an incremental approach to modernization. This approach involved breaking down the overall modernization journey into smaller manageable steps, with each delivering tangible business value. Adopting this approach allowed ABFL to achieve quick wins, mitigate risks, and continuously adapt the journey based on results and feedback.

The identified modernization changes were grouped and then packaged into a series of transition architectures (TAs). These TAs were sequenced based on business priority, quick wins, and impact. TAs are intermediary, stable, independent states between the baseline and target architecture. Dividing the modernization journey into a series of TAs helped demonstrate wealth platform capability at an architecturally incremental state. Aligning short-term wins with long-term objectives helped the team develop a non-disruptive, value-focused roadmap where each transition or phase was delivered as an incremental TA.

The ABFL team defined specific, measurable, achievable, relevant, and time-bound (SMART) goals for each TA. For example, post-cloud migration, one of the initial TAs, was identified to replatform Windows to Linux. To make the goal clear, the team calculated the amount of savings after operating system (OS) porting. A proof of concept (POC) was used to build confidence in porting and determine the effort required for this porting.

Leveraging a TA also helped demonstrate the value of the modernization effort to stakeholders and build momentum for the rest of the journey. The ABFL team integrated these TAs into a modernization roadmap with four transitions (Figure 1).

Figure 1. Wealth platform modernization journey

Now, let’s dive deeper into each of these four transitions.

Transition 1 – Infrastructure and platform modernization

The goal of the first transition was to migrate the wealth platform to the AWS Cloud, port the OS to Linux, and optimize the cloud infrastructure. After migrating the entire wealth platform to AWS, the re-platforming of Windows to Linux was performed in small batches. The servers that had dependencies on Windows, like SQL Server, were excluded from OS replatforming.

Infrastructure optimization was taken next and right sizing was carried out per the AWS Compute Optimizer report (Figure 2).

Figure 2. Infrastructure services and OS modernization

This successful transition has helped ABFL move to a stable state with savings on licensing cost and a reduction in maintenance and management efforts. It also improved availability, elasticity, and scalability. The ability to iterate quickly to undertake further modernization changes removed undifferentiated heavy lifting.

Transition 2 – Database modernization and optimization

The goal of the second transition was to move to an open-source database and refactor the database into purpose-built databases to optimize the database infrastructure and reduce cost. The database modernization was carried out in two phases. First, the self-managed SQL Server database hosted on Amazon Elastic Compute Cloud (Amazon EC2) was migrated to a fully managed Amazon RDS for SQL Server. It was modernized further by migrating to open-source Amazon RDS for PostgreSQL. The AWS Schema Conversion Tool (AWS SCT) was used to assess impact, plan, and run the conversion.

Second, to make the database lean and efficient, it was refactored into purpose-built databases. Transactional data was kept in PostgreSQL to maintain data relationships and enable rich querying. Amazon DynamoDB was used for the key-value data to utilize its flexible schema, scalability, and cost efficiency. Documents and application logs were moved to Amazon Simple Storage Service (Amazon S3) object storage to reduce storage cost. AWS Graviton processors were leveraged for Amazon RDS to improve database performance by 15-20% and reduce cost (Figure 3).

Figure 3. Database modernization and optimization journey

Transition 3 – Application modernization

The goal of this transition was to transform the wealth monolith application to a modern application architecture and move to serverless technologies to gain agility without a strain on the cost budget. This transition involved adopting modern mechanisms like microservices, containerization, REST services, and single page application architecture.

The application was refactored into smaller, manageable microservices along product feature boundaries, following the strangler fig pattern and AWS recommendations. Docker, Amazon Elastic Container Service (Amazon ECS) and Amazon Elastic Container Registry (Amazon ECR) were used to package and deploy these services on AWS Fargate, a lightweight serverless compute engine. The core microservices, which were part of the main application flow, were hosted in autoscaling Fargate containers. Supplementary services, like notifications and reconciliations, were hosted in AWS Lambda and triggered as needed.

The microservices were exposed using well-defined REST APIs and deployed on Amazon API Gateway to avoid tight coupling and to leverage built-in API Management features. Finally, the legacy Python web application was modernized into a React-NodeJS Single Page Application (SPA). Figure 4 depicts the changes carried out in this transition.

Figure 4. Application modernization journey

Refactoring to a microservice architecture on AWS Fargate and AWS Lambda improved scalability, resilience, and development velocity. REST APIs enabled the application to become headless and quickly composable. The SPA and REST APIs enabled the team to quickly build and deploy a responsive mobile application. AWS Fargate, AWS Lambda, and Amazon EC2 Graviton optimized costs and improved performance.

Transition 4 – Product-based delivery

The goal of this transition was to enable product-based delivery by adopting DevSecOps practices for faster time to market. This phase involved setting up stages in AWS CodePipeline for each service to retrieve code from the source code repository hosted in AWS CodeCommit. AWS CodeBuild used for compiling source code, runs unit tests, and produces releasable artifacts to deploy to production using AWS CodeDeploy.

AWS Cloud Development Kit (AWS CDK) and AWS CloudFormation used for defining the cloud infrastructure in code and automate infrastructure provisioning. Amazon ECR was utilized to manage container images and scan for vulnerabilities when pushed to repositories. This automation enabled frequent, reliable releases of new features by automating build, test, and deployment processes. It fostered a culture of innovation and experimentation by empowering the product team to release small changes often accelerating the delivery of business value (Figure 5).

Figure 5. Project to product-based delivery

Ongoing optimization

This modernization has helped ABFL reach an architecturally mature state, enabling the architecture to evolve and iterate quickly in alignment with business needs. Recently, the ABFL team incorporated relevant AWS innovations like Graviton processors, the latest instance types, the latest Amazon Elastic Block Store (Amazon EBS) volumes, and new Amazon S3 storage classes. Regular updates to the architecture and leveraging the latest enhanced AWS security, monitoring, and other features keeps the ABFL wealth platform aligned with business needs.

Overall benefits

This modernization journey with AWS has benefited ABFL in six ways:

1. Significant TCO saving – A significant TCO savings of approximately 4X, compared to on-premises spend, which included hardware, licenses and operational expenses. Sixty-five Window Server licenses were eliminated by moving to Linux. Migrating to PostgreSQL helped in saving four SQL Server database enterprise licenses. Adopting AWS managed security and monitoring services helped in saving monitoring, security tooling licenses and freed the ABFL team from license management complexity and administrative burden.
2. Increase in team productivity – Server consolidations, serverless infrastructure, and AWS managed services has reduced infrastructure management to only three servers. Previously managing this large on-premises infrastructure demanded considerable resources and posed procurement and scalability challenges. Infrastructure procurement time reduced from 6 – 8 weeks to near instantaneous.
3. Faster time to market – The deployment and release cycle has been reduced from weeks to hours through the implementation of microservices, automated CI/CD pipeline, and product-based delivery. Continuous monitoring of user behaviors and a tighter feedback loop with customers is enabling more effective tailoring of the user experience via personalized content and features. This is allowing ABFL to make quick product feature improvements and innovations, resulting in faster alignment with customer needs.
4. Enhanced customer experience – On-premises infrastructure had frequent issues, resulting in downtime and performance degradation. Migration to the AWS Cloud resulted in near zero down time and dramatic improvements in page response time, providing an uninterrupted experience to ABFL customers.
5. Optimum resource utilization – Leveraging serverless compute like AWS Fargate and AWS Lambda provided automatic scaling to meet changing compute demands. This helped improve infrastructure utilization and efficiency when compared to the overprovisioned and underutilized on-premises setup.
6. Enhanced security – AWS’s security services Amazon Inspector, AWS Config, Amazon GuardDuty, and AWS WAF have enhanced security posture, minimized vulnerabilities, and reduced security incidents.

Challenges encountered

• To minimize downtime, ABFL team meticulously planned and used blue/green deployment along with weighted target groups, a feature of Application Load Balancer.
• The transition to cloud-built technologies was a cultural shift and required a significant upskilling of the ABFL team. ABFL invested in technical training, as well as setting a proper IT and cloud governance structure with roles, ownership, and outcomes defined upfront.
• Ensuring compliance and implementing proper security postures was another critical task. Close collaboration between the ABFL chief information security officer (CISO) and AWS security team established appropriate security controls from the start.
• Predicting and managing costs in a dynamic cloud environment required an alternative approach to financial planning. Collaboration between the ABFL cloud team and AWS FinOps team created effective FinOps practices.

Recommended best practices

Application modernization requires a customized approach rather than a one-size-fits-all process. The specific approach to your application modernization will depend on factors like application complexity, availability of skilled resources and expected outcomes. Flexibility and adaptability in approach are key when navigating through the modernization journey. Here are some of the best practices that helped guide ABFL’s journey:

• A comprehensive assessment of your application is a must. The inputs from this assessment will assist in planning and developing the modernization roadmap that aligns with your priorities and business needs.
• During the planning phase, it is highly recommended to involve key stakeholders and establish cross-team collaboration and transparent communication.
• Use rapid prototyping to validate and redefine your priorities. AWS provides many ready-to-use building blocks to quickly test your hypotheses and validate your architecture and assumptions. The ABFL team carried out many quick POCs to evaluate different design decisions, select right AWS services, fine tuning timelines, and defining approaches to risk mitigation.
• Documenting architecture states (baseline, target, and interim TAs) is critical to demonstrate incremental capabilities and business value delivery through each transition stage. For large-scale transformation, several interim stages may be required.
• Include implementation governance and delivery plans, demonstrating how a TA that implements an interim stage will be controlled, realized and delivered.
• Have rigorous testing in place. Adopt canary releases to control exposure and DevOps practices to ensure smooth transition and minimize risks and disruptions.
• Continuous monitoring and optimization ensures that the application remains up-to-date and performs well over time.

Conclusion

ABFL’s wealth platform migration and modernization to the AWS Cloud has brought extensive benefits to ABFL by supporting its strategic goals and driving business growth. The financial analysis shows that investing in AWS services yielded 4 times more cost savings compared to maintaining an on-premises infrastructure. In addition, moving to AWS aligned with ABFL’s business objectives.

To realize the full potential of the cloud, you should not only look to migrate, but also modernize your infrastructure, platform, and applications. AWS can help you assess how your company can get the most value from the cloud. Join the millions of AWS customers who trust us to migrate and modernize their critical applications on to the AWS Cloud. To learn more on modernizing Windows Server or SQL Server, visit Windows on AWS. Contact us to start your migration and modernization journey today.

AWS has significantly more services, and more features within those services, than any other cloud provider, making it faster, easier, and more cost effective to move your existing applications to the cloud and build nearly anything you can imagine. Give your Microsoft applications the infrastructure they need to drive the business outcomes you want. Visit our .NET on AWS and AWS Database blogs for additional guidance and options for your Microsoft workloads. Contact us to start your migration and modernization journey today.