Purpose-Built Databases for Modern Applications
summary
Claims that Amazon uses multiple purpose-built stores for scale and gives exact numbers: DynamoDB handles 7.11 trillion API calls with peaks at 45.4 million requests per second on Prime Day; Aurora runs 1,900 Aurora PostgreSQL instances with 148 billion transactions, 609 TB stored and 306 TB transferred. Recommends microservices per-service data stores and lists options: DynamoDB global tables, DocumentDB (MongoDB 3.6 compatible), ElastiCache (Redis/Memcached), Neptune graph, Timestream for time-series (preview), QLDB ledger and managed Cassandra.
AI-written summary · about
chapters (16)
- 0:00Introduction to purpose-built databases
- 1:53What defines a modern application
- 5:25Why use purpose-built databases
- 7:11Microservices vs monolithic apps
- 10:22AWS database categories overview
- 12:30Demo of Amazon Aurora
- 16:01Amazon DynamoDB deep dive
- 18:29Amazon DocumentDB explained
- 19:35Amazon ElastiCache for caching
- 21:19Amazon Neptune graph databases
- 22:43Amazon Timestream for time-series data
- 24:07Amazon QLDB for ledgers
- 25:09Amazon Managed Cassandra Service
- 26:53Benefits of managed database services
- 27:14Database migration tools
- 28:17Resources and next steps
related talks
29mPurpose-Built Databases for Modern Applications
59mModernize your applications with purpose-built databases
24mBuild modern applications with purpose-built AWS databases
38mWhat is the right database for you?
20mAmazon DynamoDB: A Scalable, Predictably Performant, and Fully Managed NoSQL...
21mSave, Grow and Innovate using AWS Databases
30mBuild for Scale, Performance With Feature-packed Amazon DynamoDB
24mDatabases purpose-built for modern applications