Driver for Data-Driven Change
summary
Shows concrete numbers and actions: compares an indexed collection of 2.2 million documents to an unindexed one and demonstrates much higher RU cost and extra document reads when indexes are missing; explains choosing partition keys and autoscale vs shared vs serverless (no minimum RU); demos adding composite indexes reduce RU for server-side pagination; highlights bulk execution capable of bulk imports >1 TB/hour, change feed replacing polling to cut notification latency from minutes to seconds, and reserved capacity up to 60% savings.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:43Who am I
- 1:16DataDriven Economy
- 5:51Choosing the right abstraction
- 11:14Successful optimization
- 12:24Demo
- 20:41Confidentiality
- 21:22Bulk Execution
- 21:54DataDriven Solutions
- 27:09New Features
- 28:02Capacity Calculator
- 29:04Contact Details
- 29:24Conclusion
related talks
53mIndexing Internals for Developers & DBAs- 53mIndexing Internals for Developers & DBAs
- 53mIndexing Internals for Developers & DBAs
14mHow to achieve good cost/performance with Azure Cosmos DB elasticity
53mFrom RDBMS to NoSQL (PRT314)
11mDatabase Evolution: Partitioning and Indexing
41mUsing Databases to Pull Your Application Weight
47mFun with BIG Data with various databases