Schroders: Serverless Architecture Migration

01 / Context

The challenge

Schroders Personal Wealth needed to modernize their fraud detection and financial crime prevention systems while reducing operational costs. The legacy system faced: • High cloud infrastructure costs due to inefficient architecture • Scalability limitations during peak transaction periods • Slow response times affecting fraud detection capabilities • Complex maintenance requirements for monolithic architecture • Need for enhanced security and compliance features

02 / Contribution

The work

Led the complete architectural transformation to serverless infrastructure: **Serverless Architecture:** • Designed and implemented AWS Lambda-based event-driven architecture • Migrated from EC2-based infrastructure to fully serverless stack • Implemented auto-scaling capabilities for variable workloads • Created microservices for fraud detection and financial crime analysis **Application Modernization:** • Containerization of legacy components using Docker • Event-driven communication patterns using AWS SNS/SQS • API Gateway implementation for secure service exposure • DynamoDB and RDS integration for optimal data storage **Cost Optimization:** • Detailed cost analysis and optimization strategies • Right-sizing of resources based on usage patterns • Implementation of cost monitoring and alerting • Migration of batch processes to cost-effective Lambda execution

03 / Evidence

Reported results

20%

Reduction in cloud infrastructure expenditure

£2M

Programme value delivered through modernization

100%

Migration to serverless architecture completed

Enhanced

Fraud detection capabilities with improved response times

These results are reported in the existing portfolio and have not been independently verified.