On-prem Oracle Exadata data warehouse migration to BigQuery
InsuranceBusiness Impacts
60%
Infrastructure & Maintenance cost savings
3x
Growth in the analytics business
150 TB
Data migrated
21k
Data Mart tables added
Customer Key Facts
- Country : America
- Industry : Insurance
- About : The client is a leading U.S. - based global commercial property and casualty insurance
Challenges
- The client wanted a model orchestrator that will integrate with the Claim Center/Rules Engine and Support the deployment of clients’ flag library & deploying predictive models in Vertex AI.
- Client also wanted a ML model that will give scoring to the claims based on severity (High, medium, low), once the FNOL notification has came
- Preparation of input data for ML model – combining data points from uploaded structured and unstructured documents with legacy data from client database
- Integration with multiple modules
Technologies Used
Cloud Storage
BigQuery
Data Fusion
Dataflow
Composer
Looker
Pub/sub
DLP API
Solution
- Quantiphi migrated 150 terabytes of data from Oracle Exadata to Google’s BigQuery.
- Built a claim center data warehouse on GCP
- Developed an enterprise-grade DLP solution
Results
Significant Cost Savings
60% Reduction in Infrastructure & Maintenance Costs: The migration from Oracle Exadata to Google’s BigQuery resulted in substantial cost savings in infrastructure and maintenance, reducing overall operational expenses.
Improved Claim Processing
- Integrated Claim Center Data Warehouse: Built a comprehensive claim center data warehouse on GCP, enabling better integration with the Claim Center/Rules Engine.
- Predictive Model Deployment: Developed and deployed predictive models in Vertex AI for scoring claims based on severity (high, medium, low) upon First Notice of Loss (FNOL).
Expanded Data Infrastructure
- 21,000 Data Mart Tables Added: The addition of 21,000 data mart tables expanded the client's data infrastructure, providing a robust foundation for advanced analytics and reporting.
Data Integration and Preparation
- Combining Structured and Unstructured Data: Prepared input data for machine learning models by integrating data points from uploaded structured and unstructured documents with legacy data from the client's database.
- Seamless Module Integration: Achieved seamless integration with multiple modules, enhancing the overall functionality and efficiency of the system.
Enhanced Analytical Capabilities
- 3x Growth in Analytics Business: The migration facilitated a threefold increase in the analytics business, allowing the client to leverage enhanced data processing and analytical capabilities.
- 150 TB Data Migrated: Successfully migrated 150 terabytes of data, ensuring seamless continuity and accessibility of historical data.
Enterprise-Grade Data Security
Enterprise-Grade DLP Solution: Developed a robust Data Loss Prevention (DLP) solution, ensuring the security and compliance of sensitive data throughout the migration and beyond.
