Build Predictive Machine Learning with Flink | Workshop on Dec 18 | Register Now
Today data is big part of very bit of decision making be it parents deciding on where there children need to go for Karate classes to formulating the design for a batter driven aircraft. The speed at which data is delivered from producer to consumers is equally important. Any loss of data can potentially cause loss of millions of dollars. Walmart is no different. The world’s largest retailer caters to the need of millions of its Online and Walk in customers by ensuring optimal availability of needed assortment and timely delivery on Online fulfillment through a suit of well orchestrated event streaming platforms leveraging Kafka and its ecosystem. One of the application that plays an extremely key role to customer satisfaction is the real time Replenishment system . This system on a given day processes more than 4 billion messages in less than 3 hours leveraging an array of processors to generates an order plan for the entire network of Walmart stores with great accuracy. While doing so it also ensures no data loss through event tracking and necessary replays and retries. Through this session, we will highlight various aspects of Design, Architecture, Deployment strategy, Kafka settings, Optimization techniques etc that was paramount to achieve this rate of processing and certainly with no accidents.