Business Objective
The client is a leading global pharmaceutical company headquartered in the United States. The company focuses on the research, development, and commercialization of innovative medicines, vaccines, and animal health products. They currently has their infrastructure set up on-premise which takes weeks for the infra enablement process.
The client wanted to build a centralized, robust, and standardized cloud data ecosystem.
Challenges
- Infrastructure enablement process is ad-hoc and takes weeks
- Lack of standardized data and ML operations
- No unified experience in building models
Solution Methodology
- Established Data Foundation – Evaluated basic requirements, reviewed existing data and analytics landscape – tools, frameworks, use cases, etc., selected appropriate tools and services, etc
- Leveraged Automated DataOps Framework using AWS Redshift – Developed a solution for automated provisioning, Setup of the Unified Control Environment, Validation and testing of the solution
- Operationalized/ Modernized – Onboarded 2 local models to the cloud using best practices, enabled data science tools to collaborate, enabled visualization tool (QuickSight) to provide insights, etc
Business Impact
- Reduced time for provisioning environments, services, roles across various regions, markets, and verticals from 1 month to 1 day
- Built a robust framework with a centralized data lake using Amazon S3 & AWS Redshift to enable existing data assets to be repurposed for other projects
- Standardized data structures along with role-enabled access allowed faster onboarding and turnaround time for new projects
- Serverless architecture of AWS Redshift led to cost savings by taking advantage of a “Pay as You Go” model for infrastructure
