# Underuse of Fargate Spot for interruptible workloads Canonical: https://www.pointfive.co/efficiency-hub/inefficiencies/underuse-of-fargate-spot-for-interruptible-workloads Many teams run workloads on standard Fargate pricing even when the workload is fault-tolerant and could tolerate interruptions. By: Andrew Shieh Updated: 2026-09-28 [Cloud Efficiency Hub](https://www.pointfive.co/efficiency-hub) # Underuse of Fargate Spot for Interruptible Workloads The short version Many teams run workloads on standard Fargate pricing even when the workload is fault-tolerant and could tolerate interruptions. [Andrew Shieh](https://www.linkedin.com/in/shandrew/) Principal Engineer / Director of Engineering, Operations @ SmugMug AWS service [AWS Fargate](https://www.pointfive.co/efficiency-hub/cloud-services/aws-fargate) Category [Compute](https://www.pointfive.co/efficiency-hub/service-category/compute) Reference CER-0181 Type Suboptimal Pricing Model ## Explanation Why the waste happens and who it affects. Fargate Spot provides the same performance characteristics at up to 70% lower cost, making it ideal for stateless jobs, batch processing, CI/CD runners, or retry-friendly microservices. ## Billing model The pricing dimensions that drive this cost. Fargate is billed based on vCPU and memory usage per second. Standard Fargate pricing is significantly higher than Fargate Spot, which offers steep discounts for interruptible workloads. ## How to detect 5 checks to find it in your estate. Identify ECS workloads on Fargate that: - Do not require continuous availability or persistent state - Can tolerate retries or interruptions - Are currently using standard pricing rather than Spot - Run in non-production environments or are scheduled/batch in nature - Have high "Fargate-vCPU-Hours" and "Fargate-GB-Hours" billed at on-demand rates ## How to fix 3 ways to remove the waste. - Add the FARGATE\_SPOT capacity provider to the ECS cluster and set a capacity provider strategy on services or RunTask calls (Fargate Spot is ECS-only, not supported on EKS, and Linux-only) - For mixed environments, use a capacity provider strategy with weights and a base across FARGATE and FARGATE\_SPOT to route eligible tasks to Spot - Monitor interruption rates and implement retry logic where necessary ## Documentation Vendor references for pricing and configuration. [AWS Fargate Spot - Official Docs](https://docs.aws.amazon.com/AmazonECS/latest/userguide/fargate-capacity-providers.html) [AWS Fargate Pricing](https://aws.amazon.com/fargate/pricing/) ## Related inefficiencies [Browse the library](https://www.pointfive.co/efficiency-hub) - AWS Fargate CER-0196 ### [Suboptimal Architecture Selection in AWS Fargate](https://www.pointfive.co/efficiency-hub/inefficiencies/suboptimal-architecture-selection-in-aws-fargate) AWS Fargate supports both x86 and Graviton2 (ARM64) CPU architectures, but by default, many workloads continue to run on x86. Graviton2 delivers significantly better price-performance, especially for stateless, scale-out container... Compute - AWS EC2 CER-0324 ### [Excessive On-Demand Compute Spend Due to Low Savings Plan and Reserved Instance Coverage](https://www.pointfive.co/efficiency-hub/inefficiencies/excessive-on-demand-compute-spend-due-to-low-savings-plan-and-reserved-instance-coverage) AWS compute services charge the full published On-Demand rate when no commitment-based discount - such as a Savings Plan (SP) or Reserved Instance (RI) - is in effect. On-Demand pricing provides maximum flexibility, but it is also the most... Compute - AWS Lambda CER-0296 ### [Inefficient Lambda Pricing Model for Steady High-Volume Workloads (Use Lambda Managed Instances)](https://www.pointfive.co/efficiency-hub/inefficiencies/inefficient-lambda-pricing-for-steady-high-volume-workloads-use-lambda-managed-instances) A function has steady, high-volume traffic (or predictable load) but continues running on default Lambda pricing, where costs scale with execution duration. Lambda Managed Instances runs Lambda on EC2 capacity managed by Lambda and... Compute --- Source: the public page above. Product screenshots and illustrative interfaces are examples, not live customer data.