Estimate ECS Fargate costs for x86 and ARM with Spot pricing, ephemeral storage, and Compute Savings Plans.
Last verified: May 2026
20 GB included free. Additional storage charged at $0.000111/GB/hour.
Fargate is serverless: you define CPU/memory per task and AWS manages the underlying infrastructure. EC2 launch type gives you control over the EC2 instances in your cluster, offering lower cost at the expense of operational overhead. Fargate is ideal for teams that want to focus on application code without managing servers.
AWS Graviton processors deliver up to 20% lower cost and up to 40% better price-performance compared to x86 instances. Most containerized workloads can run on ARM with multi-arch Docker images. Graviton is recommended for new workloads unless you have x86-specific dependencies.
Fargate Spot runs tasks on spare capacity at up to 70% discount. Tasks may be interrupted with a 2-minute warning. Best for fault-tolerant workloads like batch processing, queue workers, and CI/CD jobs. Not recommended for user-facing services that require consistent availability.
Savings Plans commit to a consistent amount of compute usage ($/hour) for 1 or 3 years. 1-year No Upfront offers ~22% discount; 3-year All Upfront offers ~52% discount. They apply automatically to Fargate (and Lambda/EC2) usage and are flexible across regions, instance types, and OS.
Output will appear here...The ECS/Fargate Cost Estimator calculates monthly costs for running containers on AWS Fargate, covering both x86 and ARM (Graviton) processor architectures. It factors in vCPU and memory pricing, ephemeral storage beyond the 20 GB default, Spot pricing for fault-tolerant workloads, and Compute Savings Plans discounts. The tool provides side-by-side comparisons to help you choose the most cost-effective configuration.
Fargate charges per-second for vCPU and memory resources allocated to your task. You specify the task's CPU and memory at the task definition level, and billing starts when the task image is pulled and stops when it terminates. Pricing varies by region and processor architecture.
Fargate Spot uses spare AWS capacity at up to 70% discount but tasks can be interrupted with a 30-second SIGTERM warning. Use Spot for fault-tolerant workloads like queue processors, batch jobs, and development environments.
Yes. Graviton-based Fargate tasks are approximately 20% cheaper than x86 tasks and often deliver better performance for containerized workloads. Ensure your container images are built for the linux/arm64 platform before switching.
Your team runs 50 Fargate tasks (0.5 vCPU + 1 GB each) 24/7 for a microservice fleet. Current cost: $1,950/month at on-demand x86. The estimator models: switching to Graviton saves $390/month (-20%); adding a 1-year Compute Savings Plan saves another $470/month (-30%); overall = $1,090/month, a 44% reduction. You implement Graviton first (no commitment), then layer the Savings Plan after 2 weeks of confirming performance parity.
The estimator computes Fargate cost as: vCPU-seconds × per-vCPU-second rate + GiB-seconds × per-GiB-second rate, plus optional ephemeral storage above the 20 GB free tier × per-GB-month rate. Spot pricing applies a discount (~70%) to the rate. Compute Savings Plans apply a flat 17-50% discount on the compute portion based on commitment level.
Fargate's per-second billing has a 1-minute minimum charge per task. For short-running batch jobs (<60 seconds), you're paying for 60 seconds regardless. Either batch multiple jobs into a single task or use Lambda for sub-minute work — Fargate becomes wasteful at small task durations.
Compute Savings Plans cover Fargate at the same discount rates as EC2 (up to 50% for 1-year, 66% for 3-year, no upfront option). If you have any always-on Fargate tasks, a Savings Plan covering the baseline is one of the highest-ROI optimizations available — and it doesn't lock you to specific instance types.
Fargate Graviton (ARM) tasks cost ~20% less than x86 Fargate. Most modern containerized apps (Node.js, Python, Java with ARM-compatible JDK) run on Graviton with zero code changes. Check your image: `docker manifest inspect <image>` should show linux/arm64 platform support before flipping the runtime architecture.
Was this tool helpful?
Disclaimer: This tool runs entirely in your browser. No data is sent to our servers. Always verify outputs before using them in production. AWS, Azure, and GCP are trademarks of their respective owners.