
Building a Real-Time F1 Telemetry Dashboard with Apache Pinot and Apache Flink on AWS - Lets Talk About Data
In this show, we explore how to build a real-time analytics solution using Apache Flink and Apache Pinot on AWS. We demonstrate this by simulating Formula 1 race telemetry data, streaming it through Flink for processing, and ingesting it into Pinot for ultra-low-latency querying and live dashboard. We see how Apache Pinot performs real-time aggregations and serves fast OLAP queries to power high-speed, data-driven experiences.
Prasad Matkar
Amazon Employee
Published May 23, 2025
In this episode of Let's Talk About Data, host Prasad Matkar is joined by Ismail Makhlouf and Francisco Morillo . They walkthrough an innovative demonstration of building a real-time analytics pipeline for Formula 1 racing telemetry data. The session showcased how to combine Apache Flink and Apache Pinot on AWS to create a high-performance streaming analytics solution. The architecture begins with collecting live telemetry data from an F1 racing game through a UDP client, which is then streamed to Amazon MSK. Apache Flink processes this streaming data, performing enrichments and applying business rules to track metrics such as driver positions, speeds, brake usage, and tire temperatures. The processed data is then ingested into Apache Pinot, which serves as the real-time serving layer, enabling ultra-low-latency querying and visualization through Grafana dashboards. The system also incorporates AI-powered insights using Amazon Bedrock, providing real-time performance recommendations. Throughout the demonstration, the hosts showed how this architecture achieves sub-second latency for complex analytical queries while maintaining high throughput, making it ideal for real-time applications requiring immediate insights from streaming data.
Key highlights
• Apache Pinot as a real-time OLAP database for serving low-latency queries
• Integration with AWS services like Amazon MSK (Managed Streaming for Apache Kafka)
• Real-time dashboard creation using Grafana connected to Pinot
• Demonstration of sub-second latency for complex analytical queries on streaming data
• Use of Amazon Bedrock for generating AI-powered insights during the race
• Scalability and performance considerations for both Flink and Pinot
• Comparison with traditional batch processing and data warehousing approaches
Prasad Matkar - Database Specialist @AWS
Francisco Morillo - Senior Specialist Solutions Architect@AWS
Ismail Makhlouf - Senior Specialist Solutions Architect@AWS
Ismail Makhlouf - Senior Specialist Solutions Architect@AWS
Links from today's episode
- Batch vs stream data processing - https://www.aboutamazon.in/news/tech-blog/a-guide-to-batch-vs-stream-data-processing-for-developers
- Apache Pinot - https://pinot.apache.org/
- Build a real-time analytics solution with Apache Pinot on AWS - https://aws.amazon.com/blogs/big-data/build-a-real-time-analytics-solution-with-apache-pinot-on-aws/
- Deploy real-time analytics with StarTree for managed Apache Pinot on AWS - https://aws.amazon.com/blogs/big-data/deploy-real-time-analytics-with-startree-for-managed-apache-pinot-on-aws/
Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.