Keeping the Noisy Neighbors Happy
summary
Explains concrete choices for SaaS multitenancy: prefer database-per-tenant for security but use Azure elastic pools to share compute and lower cost; use the Microsoft shard map manager to map tenant IDs to per-tenant databases and demonstrated creating four tenant DBs and running schema updates across them with elastic jobs; instrument telemetry with a tenant ID and build per-tenant monitoring in Application Insights to detect noisy neighbors before customers call.
AI-written summary · about
chapters (10)
- 0:00Start
- 1:31The Cloud World
- 14:08Reference Architecture (Monolith)
- 25:35Reference Architecture (Microservices)
- 41:18Onboarding & Provisioning
- 43:18Global Reach
- 44:43Encryption
- 46:44DevOps Issues
- 48:01Monitoring
- 56:44Takeaways
related talks
46mDatabase-Per-Tenant Architecture Using .Net Core and Azure SQL Database
14mBest Practices for Building Multi-Tenant Applications on Azure Cosmos DB
18mBuilding Performant Data Architecture for Multi-Tenant SaaS Apps
25mMulti-Tenancy Architecture Lessons Learned
52mMultitenancy: cost-efficient and scalable patterns for Software-as-a-Service
48mBuilding SaaS solutions on Azure SQL Database
14mEasier Queries for Multi-Tenant SaaS Apps
46mDesign and build multi-tenant SaaS apps at scale with Azure Cosmos DB