Scalable Cloud Migration for Personalized Diabetes Care Platform

About the Client

A global MedTech leader providing diabetes care solutions sought to improve the scalability and reliability of its hypercare platform. This platform plays a critical role in onboarding, monitoring, and delivering personalized services to diabetes patients while processing terabytes of data daily from connected insulin pumps, caregivers, and business systems. The legacy on-premises infrastructure struggled to keep pace with rapid data growth and rising patient volumes, creating risks for downtime and delayed insights. To address these challenges, the client partnered with Zimetrics to migrate the platform to a modern, cloud-native AWS architecture.

Impact Delivered

25%
Reduction in BOM cost
50%
Reduction in wireless charging component costs
17–23%
Improvement in patient recovery metrics
17–23%
Improvement in patient recovery metrics

Standing at a Turning Point

Scalability Limitations: On-prem infrastructure could not handle 22,000+ daily uploads from IoT devices, patient portals, and business data sources like Salesforce and SAP, leading to frequent slowdowns.

Data Flow Management: Efficiently processing terabytes of heterogeneous data while maintaining throughput and concurrency (up to 1,000 concurrent Lambda executions) was difficult.

Interoperability Issues: Multiple data sources like medical devices, patient engagement apps, and enterprise systems

Compliance & Security: Required strict privacy controls to protect PII and PHI while ensuring compliance with healthcare regulations.

Solutioning

Zimetrics designed and executed a phased migration and modernization strategy to transition the client’s hypercare platform from its legacy on-premises environment to a scalable AWS Cloud infrastructure.

Migration Strategy

  • Extracted and loaded data using ETL jobs and AWS Lambda scripts, achieving 230 Lambda executions in 15 minutes.
  • First Migration (April 2024), Migrated data from on-prem Oracle schemas to AWS Oracle RDS,
  • Second Migration (August 2025) migrated an AWS Oracle database to a PostgreSQL database using AWS Database Migration Service (AWS DMS) and the AWS Schema Conversion Tool (AWS SCT). Built automated failure handling and recovery mechanisms.
  • Introduced data quality verification pipelines to ensure migrated data integrity.
  • Implemented masking for PII and PHI data to meet compliance requirements.


AWS Cloud Infrastructure Setup

Refactored backend services to leverage AWS-native components:

  • Amazon S3 for ingestion and data lake storage.
  • PostgreSQL for Analytical and Complex Workloads
  • Amazon ECS on Fargate for microservices containerization.
  • Amazon SQS & SNS for decoupled event-driven processing.
  • Amazon CloudWatch for real-time monitoring and alerting.
  • AWS API Gateway for secure API exposure to external systems.

Modernization of Workloads
Decomposed monolithic applications into containerized microservices:

  • Set up Blue- green deployment strategies for seamless production rollouts.
  • Dual-Write Strategy for data consistency
  • Integrated GitHub, JFrog, Jenkins pipeline, AWS CodeBuild, CodePipeline, Cloud Formation into a DevOps-driven CI/CD pipeline.
  • Work In Progress: GitHub is in process to migrate to GitActions

Data Observability & Insights

  • Integrated Amazon OpenSearch for real-time analytics and search.
  • Developed data pipelines for therapy performance, patient engagement, and churn analytics.

Tech Stack:

  • Data pipelines: AWS Lambda, S3, PostgreSQL
  • Microservices: Amazon ECS (Fargate), Auto Scaling Groups, EFS/S3
  • APIs & Integration: AWS API Gateway, Amazon EventBridge, Amazon SQS, SNS
  • CI/CD: GitHub, JFrog, AWS CodeBuild, CodePipeline, Jenkins pipeline, CloudFormation
  • Monitoring: Amazon CloudWatch, CloudTrail Logs, Errors log
  • Analytics: Amazon OpenSearch, ELK – Kibana tool

Engineering the Transformation

Zimetrics engineers began with a complete teardown of the existing device — from PCB layout to sensor calibration behavior. Using schematic cross-mapping and dependency tracing, the team identified more than 60 EOL or NRND components affecting sensing, BLE communication, and power management circuits. Each dependency was categorized by form-fit-function risk to ensure downstream design decisions didn’t compromise FDA 510(k) clearance

Zimetrics engineers began with a complete teardown of the existing device — from PCB layout to sensor calibration behavior. Using schematic cross-mapping and dependency tracing, the team identified more than 60 EOL or NRND components affecting sensing, BLE communication, and power management circuits. Each dependency was categorized by form-fit-function risk to ensure downstream design decisions didn’t compromise FDA 510(k) clearance

Future Outlook

This project creates a blueprint for sustainable device evolution. With validated component alternatives, stable firmware, and EMI-safe wireless capability, the client is positioned to scale rapidly and lead in wireless medical wearables.
“Every lifecycle challenge is an opportunity to reimagine innovation. This project proves that with the right partnership, even the most complex problems can become growth opportunities” — Zimetrics Team

Related Stories