Building scalable IoT systems with MQTT and AWS
summary
HiveMQ outlines concrete patterns for MQTT on AWS: brokers on EKS or EC2 bridge local edge brokers to Kinesis, MSK, S3, DynamoDB, Timestream and Lambda for real-time message processing. A single HiveMQ installation reportedly handles 30 million devices using about 10 EC2 instances. Edge gateways keep factories running offline, while a data hub enforces schema validation and transformations before landing in data lakes. BMW uses the pattern to lock/unlock cars in milliseconds.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:54Technology
- 3:51Patterns
- 7:11MQTT and AWS
- 9:27Case study BMW
- 11:38Durable storage patterns
- 12:56Combining live and durable data
- 13:44Data quality
- 14:41Data Hub
related talks
44mBenefits of MQTT for IoT Messaging and Beyond
49mDesign patterns for distributed client server systems via IoT
41mScaling IoT and Web messaging infrastructure for fun and profit: An Introduction to Eclipse Amlen
51mLeveraging Eclipse Sparkplug to Accelerate DataOps and Machine Modeling into AWS Services
28mTransform Tomorrow's Industries With AWS IoT
24mMigration from On-Prem Messaging System to The Cloud
29mData-driven Transformation With AWS IoT
40mBuilding Scalable Resilient IoT Messaging