FinOps & Observability: technical implementation
summary
A cloud-cost engine analyzes AWS Pricing API and CloudWatch metrics, runs as scheduled ECS Fargate tasks (Terraform IaC), writes idempotent results to Aurora Serverless, pushes reports to S3 and Snowflake for PowerBI/Tableau. It flags unused EBS, DynamoDB, RDS and EC2 resources, generates team-level evidence (bytes-in/out, last use) and management-level savings estimates plus required engineering hours. A single deployment uncovers $5–7M/month cloud spend with EBS volumes as biggest cost and produces actionable savings that management funds and teams implement.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 5:02Cloud cost control struggles
- 12:34FinOps model for EBS volumes
- 18:12Serverless cost analysis framework
- 24:11Engineers versus management
- 27:05FinOps strategy alignment
- 32:11Custom vs generic FinOps tools
- 36:04Q&A
related talks
19mFinOps for business results
23mFinOps and Cloud Cost Optimization
38mFinOps and Tag Strategy on AWS
43mMigration to the Cloud, DevOps Perspective
19mDoing FinOps Right: Stories of (Anti)Patterns and Spendthrift Architectures
34mOptimizing Cloud Efficiency with Prompt Engineering
27mMastering AWS Cost Optimisation
46mWhen the CFO Comes Knocking: Our Journey to a Cost Efficient Data Platform