Azure Cosmos DB Performance Tuning - Sonal Pathak & Amresh Krishanmurthy
summary
Provisioning uses Terraform to create Azure Cosmos DB accounts with the Mongo API and automate settings (DB vs collection throughput, autoscale 400→4000 RU). Monitoring uses the portal Insights and Metrics to spot RU spikes and throttles; fixes include server-side retry, upgrading Mongo versions 3.6→4.2, temporary bump to 20,000 RU for bulk loads then back to 4,000, TTL for ephemeral records, careful partition key choice to avoid hot shards, and a concatenated synthetic key (category+subcategory). Java reactive and non-reactive SDKs are used for integration and these changes reduce throttling and improve steady throughput.
AI-written summary · about
chapters (6)
- 0:00Introduction
- 0:48Agenda
- 3:05Cosmos DB Features
- 9:31Terraform Setup
- 13:15Insights Monitoring
- 19:04Performance Measures
related talks
48mPerformance tuning for Azure Cosmos DB
01h 00mCost cutting strategies for Azure Cosmos DB
15mTurning Growing Pains To Growing Wins With Azure Cosmos DB
18mIntelligent Resource Optimization: Mastering Azure Cosmos DB for Enterprise Workloads
08mOptimize for the Best Cost and Performance for Azure Cosmos DB NoSQL API
46mDesign and build multi-tenant SaaS apps at scale with Azure Cosmos DB
01h 00mAzure Cosmos DB : A brief Note for Developers
14mHow to achieve good cost/performance with Azure Cosmos DB elasticity