CASE STUDY

Food services enterprise modernizes business intelligence ecosystem

A US$ 60+ billion company selling food as well as non-food products to customers and enterprises through a global distribution network.

Key Challenges

The food services enterprise made ad hoc technology investments during a period of rapid business growth, which resulted in a dysfunctional IT landscape –

Data silos for inventory, demand planning, sales and merchandising, and logistics
Data warehouse that could not scale up to emerging business needs
Prolonged back-up schedule and frequent hardware failures

Ready to experience?

TALK TO EXPERTS

The Solution

Scalable and homogenous IT structure

Infosys recommended a cloud-based solution based on a rigorous business and systems assessment. We adopted an ecosystem approach to empower business users with relevant data, and drive advanced data analytics, BI and reporting.

Infosys designed a system to meet the exponential growth in business and corresponding increase in data to be stored, harvested and distilled for business insights. At an enterprise level, our solution delivers compelling business benefits: boosts e-commerce, improves supply chain operations, increases transparency, enhances deliveries, and manages product costs. From a budget standpoint, the new system avoids upfront capital expenditure in favor of timely and relevant infrastructure investments with minimal maintenance costs.

Migration

The Infosys deployment began with migration of on-premise Informatica jobs to an Amazon EC2 instance for concurrent processing of high data volumes. We reengineered the Informatica-based ETL solution for loading 1,100+ tables and 1,400+ mappings. All historical and resource-intensive jobs were identified and rewritten into AWS EMR scripts using Pi Spark to reduce E2E data load execution. An asynchronous process accelerated loading.

Our experts rebuilt the enterprise data warehouse using Amazon Redshift, which uses intense compute six-node clusters. We migrated historical data of the past 10 years from the on-premise systems to AWS Redshift using the Amazon EMR platform.

Orchestration

We created a reusable data orchestration solution by leveraging AWS products – Lambda, EMR, data pipeline, and Aurora DB. Amazon S3 was used as a staging layer to process and store data from various sources, and build a data lake for analytical requirements. The orchestration was restructured from a traditional control and scheduler to AWS data pipeline, which uses Lamda functions for triggering data loads from the source to target. AWS Glacier simplified backup and restoration.

Distinct layers for data collection, job submission and data load were introduced for data ingestion and orchestration. We configured checkpoints and enabled failure notifications through Amazon SNS alerts. In addition, we ensured data synchronization between Amazon Redshift and S3. Our solution for cloud-based event triggering, workflow and orchestration addressed downstream data requirements.

Self-service

The Infosys team reengineered the reporting layer to facilitate ready access to the data layer, and provide business users with one version of the truth. The system allows for a high degree of self-service data capabilities as well as collaboration among users across the enterprise. Business users can now access data within one hour as opposed to the earlier six hour time window, enabling faster decision-making and accelerated turnaround time.

Our cloud solution collapsed business silos, enabled systems to talk with one another, and aggregated data of more than 15 business-critical enterprise applications on one platform. The IT ecosystem helps C-level executives distill business intelligence and make informed decisions. In addition, it enables approximately 7,500 business users to access 9,000 scheduled reports and 3,000+ ad hoc reports.