DynamoDB Backups Gone Wild

AWS
by AJ Bothe, Proctor & Gamble

I recently joined a team and one of our accounts was spending an insane amount on DynamoDB (close to $20k / mo) for a set of tables that was under 20GB. Come to find out we had a process to backup the tables every hour of every day with no retention policy. This was implemented before point-in-time recovery was implemented. If you compound the costs of backups, it means that backup storage costs were growing exponentially each month. We checked with the business and asked the basic questions like “Do you really need this much retention?”. Switching to point-in-time recovery and deleting old backups from DynamoDB resulted in significant savings (~$12k / month) and still allowed us to restore to any minute within the last 35 days. Deleting the old backups is the most difficult part surprisingly, due to limitations of the API, so a special script had to be written to loop through each backup.

Related Member Stories

A Guide for Adopting FinOps in Your Organization

AWS
Azure
GCP
Industry: Internet
Persona: FinOps Practitioner
by F2 Working Group, FinOps Foundation

One of the biggest challenges in starting a FinOps practice is getting broad executive support and buy-in to dedicate the time and resources needed for the cultural change.

Read more

Removing AMI Snapshots

AWS
by Stephanie Gooch, AWS

Upon reviewing the amount of snapshots a customer had we found a large proportion of them were created from AMIs. This was found by listing all available amis in and connecting them back to the snapshot using the description. However, many of the AMIs that created them had been released....

Read more

Architecting Cloud Workloads for Financial Reporting

AWS
Azure
GCP
Industry: Information Technology & Services
Persona: FinOps Practitioner
by Rich Hoyer, SADA

A list of best practices for cloud architects to design systems to optimize FinOps.

Read more