Title

AWS re:Invent 2023 - Scaling FinOps with a holistic approach to resource optimization (COP214)

Summary

• Andrew Midgley, Oved Lurie, and Sam Keller presented their insights on FinOps and resource optimization.
• The session focused on the FinOps lifecycle phases: inform, optimize, and operate, with an emphasis on the optimize phase.
• The speakers discussed the importance of resource optimization in on-premises environments, Kubernetes, and public cloud.
• They highlighted the need for engineers to act on cost optimization and the challenges associated with it.
• Turbonomic and Cloudability were presented as tools that help in resource optimization and cost allocation.
• The session covered the importance of right-sizing, placement, and scheduling of workloads across different cloud environments.
• Sam Keller shared Hyland Software's journey in optimizing their infrastructure using Turbonomic, including the financial and operational benefits.
• The speakers emphasized the importance of automation in FinOps, especially for non-disruptive actions like EBS volume scaling.
• The concept of unit economics was introduced as a way to measure business value against resource costs.
• The session concluded with a call to action for FinOps practitioners to leverage tools like Turbonomic and Cloudability to amplify their impact on efficiency and cost savings.

Insights

• The FinOps lifecycle is a structured approach to managing cloud costs and operations, and optimizing resources is a critical part of this lifecycle.
• Resource optimization is a consistent challenge across different environments, including on-premises, Kubernetes, and public cloud.
• Automation is key to scaling FinOps practices, especially when dealing with a large number of resources.
• Tools like Turbonomic and Cloudability can provide actionable insights and automate optimization tasks, leading to significant cost savings and operational efficiencies.
• Measuring unit costs, such as cost per monthly active user or cost per billion lines of billing data processed, can provide a clear picture of the efficiency and business value of cloud resources.
• The session highlighted the importance of collaboration between FinOps practitioners, engineers, and business stakeholders to achieve resource optimization at scale.
• The speakers suggested that even small optimizations, when automated and applied at scale, can lead to substantial cost savings and more efficient use of cloud resources.