Modernize Your Retail Data Stack: Replacing Legacy ETL and EDWs with Open Source
Key Speaker
Senior Software Engineer, Ksolves India Limited
He is a Senior Software Engineer with hands-on experience in data engineering, real-time streaming architectures, open-source data platforms, and building scalable Data Lakehouse solutions for enterprise environments.
In this session, Rahul Prakash Surti, Senior Software Engineer at Ksolves India Limited, addresses a modernization challenge that many retail organizations are actively facing: legacy ETL pipelines and Enterprise Data Warehouses that were built for a different era of data volume, velocity, and analytical demand.
The problem with legacy data infrastructure is not just cost, although licensing and maintenance expenses are significant. The deeper issue is architectural. Traditional ETL pipelines and EDWs were designed for batch processing and structured data, and they struggle to keep pace with the real-time demands of modern retail operations, including Point of Sale transaction streaming, IoT data ingestion, inventory analytics, and customer behavior tracking. Teams end up working around their own data infrastructure instead of building on top of it.
This webinar presents a practical proof of concept demonstrating how a modern, open-source Data Lakehouse can replace legacy components entirely. Rahul walks through an architecture built on Apache NiFi, Spark Structured Streaming, Apache Iceberg, MinIO, Trino, and Apache Superset, showing how these technologies work together to deliver real-time analytics at scale. The session covers streaming POS and IoT data in real time, handling schema evolution without pipeline downtime, running federated queries across the platform using Trino, and building real-time dashboards for retail analytics, all without vendor lock-in.
If your organization is modernizing its retail data platform, this session provides practical, implementation-focused insights into building a scalable, open-source data architecture that is built for today’s demands.
Key Takeaways
- How to replace legacy ETL pipelines and Enterprise Data Warehouses with an open-source Data Lakehouse architecture.
- Building real-time retail data pipelines using Apache NiFi and Spark Structured Streaming.
- Managing data at scale with Apache Iceberg and MinIO, including schema evolution without downtime.
- Running federated queries across the data platform using Trino.
- Building real-time retail analytics dashboards using Apache Superset.
- Reducing infrastructure costs and eliminating vendor lock-in through open-source modernization.
