on What to Avoid in Public Clouds
summary
Intuit begins cloud adoption with nonproduction mission-critical workloads—build environments and load-test generation—then scales production only after patterns prove out; load-test demand rises about 12× once on demand capacity is available. Intuit open-sources its load generator Tank on GitHub to automate spinning up and tearing down capacity. Two hard lessons: track cloud-native metrics like average instance age to avoid lifting legacy management patterns, and enforce shutdown or cost controls so teams don’t keep paying for idle instances.
AI-written summary · about
chapters (5)
related talks
18mIntuit's journey to the cloud: Protecting native services
27mHybrid-Clouds: How to Go Slow and Haemorrhage Money Doing It
19mFrom discovery to optimization: Four steps to cloud adoption
56mBusy Architect’s Guide to the Cloud Offerings
26mChallenges of Large Enterprises in Moving to the Public Cloud
59mBuilding the Best Cloud Data Architecture: Avoiding Costly Pitfalls
31mOn-Prem is the new Black
20mAre the days of on-premise servers over?