Centralized Data Lake
InsuranceBusiness Impacts
70%
Reduction in time to market
$15K
Estimated spend reduction per month
95%
Meeting SLA efficiency
Customer Key Facts
- Location : North America
- Industry : Insurance
Problem Context
The customer is a large insurance Fortune 100 company that had its data stored on over twenty systems and in multiple formats which resulted in data duplicity.
Challenges
- Different teams used multiple tools which resulted in increased cost of ownership support resources
- Insurance claim data is highly confidential and needs to be secured in transit
- Integrating data from systems of different sizes and configurations
- Large-scale data ingestion took required higher turnaround time
Technologies Used
Amazon Lambda
Hadoop HDFS
Apache Ranger
Apache Atlas
Apache Knox
Hortonworks Data Platform
Developing a Centralized Data Lake for Scalability, Security and Greater Cost Savings
Solution
Quantiphi migrated their data from multiple on-prem data sources to cloud. A cost-effective and scalable centralized data lake platform was then built on cloud by setting up an ingestion pipeline with security protocols; further enabling them to visualize and draw insights in real time.
In the platform, billing alerts were enabled for any threshold violations, AWS Lambda functions were used to start/stop services as per business hours, and AWS Glue was used for fully automated and highly scalable heavy ETL jobs.
Result
- Improved data security
- Flexible user management
- Foundation for future AI/ML workloads
