What are the most effective strategies for cost optimization in 2024?

All answers to this question.

Cost creep is a common issue in <cloud computing> if you don't have a strict FinOps culture. First, implement a "Janitor" script using Python and Boto3 to find and delete unattached EBS volumes and aged snapshots. Second, move your non-production workloads to Spot Instances; this can save you up to 90% compared to On-Demand pricing. Also, look into AWS Instance Scheduler or Azure Automation to turn off development environments during weekends and off-hours. Finally, use a tool like CloudHealth or even the native Cost Explorer to set up granular alerts that trigger when a specific service exceeds its daily budget.

   Answered 2023-06-12 by Brenda Harrison

Are your developers empowered to see the costs of the resources they spin up, or is the <cloud computing> bill only visible to the finance department?

   Answered 2023-06-14 by Douglas Perry

• Douglas, that's part of the problem. Only the leads see the bill at the end of the month. I want to integrate cost visibility into our CI/CD pipeline. Have you used tools like Infracost? I heard it can show the cost impact of a pull request before the infrastructure is actually provisioned. That seems like a great way to build cost-awareness in our workflow.

     Commented 2023-06-15 by Stephen Foster

Rightsizing is key. Most teams over-provision "just in case." In <cloud computing>, it's much better to start small and let horizontal auto-scaling handle the spikes.

   Answered 2023-06-17 by Pamela Jordan

• Spot on, Pamela. We cut our bill by 30% just by switching from m5.xlarge to m5.large and letting the auto-scaler do its job during peak hours.

     Commented 2023-06-18 by Sandra Mitchell