Digital
Point of Sale Order Processing with Confluent Kafka
A Fortune 100 Retail company needed to build a message driven event platform which can act as a unified information hub to exchange business information in real-time and enable faster decision making.
Client Challenges and Requirements
- The legacy system was time consuming, not highly available, and data was being lost due to the lack of stream persistence.
- Client required a solution that was cost efficient, provided maximum throughput, achieved zero data loss and maintained message history.
- The platform needed to provide for security, disaster recovery, fault tolerance, scalability and real-time data processing.
Bitwise Solution
Bitwise built a distributed, high performance and highly available data platform with Confluent Kafka to produce real-time alerts for the POS order processing system.
Solution Design
Data Collection - collect data across various digital distribution channels
Data Cleansing - process collected data in real-time
Real-Time Data Operations - apply business rules
Real-Time Data Operations - apply business rules
Analytics & Alerts – provide remote monitoring and implement alert mechanism for ecommerce engine
Tools & Technologies We Used
Confluent Kafka
Kafka Streams
Key Results
High Performance – can sustain >300+k messages /sec
High Availability – runs 24x7 and achieves Uptime SLA of 99.5%
Scalability – 12% rise in boarding new third-party suppliers
Go to market time reduced by 30% and customer satisfaction increased by 40%
