Big Data Platform Implementation in Banking - Case Study

Norway’s Largest Bank, DNB ASA, was looking to set-up a Big data platform to be able to ingest large amounts of data (both from on-prem and external, structured & unstructured, batch as well as real-time), have the ability to process this and make data available enterprise-wide in the data lake. This was required to provision data, for specific users/user group, for specific business purpose and to provide a platform for advanced analytics. This included downstream ODS layer with transactions data to provision & categorize data for the bank’s mobile application. Another use-case included implementing a data capture solution to ingest click stream data for analytics

Key Challenges

DNB ASA’s existing EDW platform was expensive and time consuming to scale in order to provision the required data. Provisioning of test environments for any use-case was a time consuming task. Considering heavy initial investment by on-prem systems, exploring newer data opportunities became a challenge

Ready to experience?

TALK TO EXPERTS

The Solution

Infosys implemented data lake based on serverless architecture using services such as AWS s3, glue, lambda and fully automated processing was achieved based on an event-driven architecture

Execution using Agile methodology & DevOps practice significantly enhanced time-to-market capabilities

Key governance considerations such as metadata management using Glue catalog, security governed using bucket policies, KMS and IAM roles, and appropriate lifecycle management setup
Infra as a code (IAAS) based platform to be able to spin-up or tear-down the environment with short turnaround times also help developers improve productivity
Fully automated deployment on AWS cloud with CI/CD approach, along with other solutions for end to end testing and quicker environment provisioning
Real-time monitoring of CI/CD pipelines and automated functional tests led to reduced defects