Accelerated analytics: Building the next-gen data platform for Hertz
summary
Claims a full next‑gen data platform for Hertz built in 10 months using native AWS services and tooling: Amazon S3, Amazon Kinesis, Amazon EMR, AWS Lambda, Glue, Redshift, Aurora, DynamoDB, Elasticsearch, plus Airflow and Tableau. Architecture is specific: Kinesis→Firehose→S3 raw→curated→conformed with S3 event→Lambda→SQS→EMR Spark orchestration, YAML schedules, CI/CD CFTs and auto‑scaling, restartability and audit logs in Aurora. It supports cross‑functional reporting, runs over half a dozen production ML models, handles 1.5M+ vehicles and 19M+ customers, and is deployed and monitored end to end.
AI-written summary · about
chapters (13)
- 0:00Intro
- 0:24Agenda
- 4:07Opportunity
- 6:25Building from ground up
- 8:24Engagement across users
- 10:01What is a modern data platform
- 11:30What is this solution solving
- 19:03DynamoDB implementation
- 23:16Glue implementation
- 26:16Patterns
- 33:11Pattern design
- 36:01Automation deployment
- 40:16Learnings
related talks
28mTurn Data Into Insights
22mHow Rocket Companies modernized their data strategy on AWS
43mData warehousing reinvented for today's needs
43mManaging a Data Lake on AWS
29mThe need for speed: Data streaming architectures
27mData-driven and cloud-native ecosystems in the automotive industry
29mGrowing insights through hybrid data architectures
49mBuilding an Enterprise Data Lake Architecture on AWS