How AWS Savings Plans and Reserved Instances differ, when each fits, a commitment mix that avoids over-committing, and where database Savings Plans fall short.
A case study cutting ElastiCache cost over 6x across 200+ clusters through rightsizing, Redis to Valkey conversion, and reserved instances, with the numbers for each lever.
A real Aurora Serverless to provisioned conversion that cut database cost from $450 to $100 a day and improved query latency, plus when Serverless is the right call.
How old logs sitting in S3 Standard quietly cost thousands a month, and how a few lifecycle rules on the right buckets and prefixes cut S3 cost by over 90%.
How we find zero-connection and idle RDS databases across dozens of instances and multiple regions, and the savings that come from deleting and rightsizing them.
Cross-AZ data transfer between instances and their databases can cost tens of thousands a year. How we find it by tag and fix it with a same-AZ replica and failover.