Astro Pricing
The fully-managed platform to take Apache Airflow to the next level.
Available on AWS, Azure, and Google Cloud.
Developer
For developers and data teams that are getting started. Pay-as-you-go starting at $0.35/hr.
- Flexible, scale-to-zero compute
- API Access
- 14-day free trial including $300 credit
Team
For teams with pipelines in production that require Airflow support.
- Everything in Developer
- Network Isolation
- Audit Logging
- 24 x 5 Support Availability
Plan Features
Straightforward Pricing
Astro offers transparent pricing tailored to your team’s needs. All product tiers use the same dimensions of our usage-based pricing model: your Airflow cluster, deployment sizing, and worker compute. Networking costs are passed through from the cloud provider.
Cluster Pricing
Configure your cluster type based on networking and security needs.
Type | Price |
---|---|
Standard | Included |
Dedicated | Starts at $2.40 per/hour |
Deployment Pricing
Easy to create, easy to delete, easy to pay for.
Deployment Size | Resources | Developer Price |
---|---|---|
Small | 1 vCPU, 2 GiB memory | $0.35 per/hour |
Medium | 2 vCPU, 4 GiB memory | $0.57 per/hour |
Large | 4 vCPU, 8 GiB memory | $0.77 per/hour |
Worker Pricing
Astro offers the largest worker compute options in the managed Airflow market by 8x.
You only pay for workers when you need them.
Worker Size | Resources | Developer Price |
---|---|---|
A5 | 1 vCPU, 2 GiB memory | $0.13 per/hour |
A10 | 2 vCPU, 4 GiB memory | $0.26 per/hour |
A20 | 4 vCPU, 8 GiB memory | $0.52 per/hour |
A40 | 8 vCPU, 16 GiB memory | $1.04 per/hour |
A60 | 12 vCPU, 24 GiB memory | $1.56 per/hour |
A120 | 24 vCPU, 48 GiB memory | $3.12 per/hour |
A160 | 32 vCPU, 64 GiB memory | $4.16 per/hour |
FAQs
What if I need to run individual tasks on bigger workers?
You might have a large number of tasks that require low amounts of CPU and memory, but a small number of tasks that are resource intensive — e.g., machine learning tasks.
To address this use case, we recommend using worker queues. Worker queues allow you to configure different groups of workers for different groups of tasks. That way, you’ll only be charged for the larger worker type if and when a task that requires that worker type actually runs.
Specifically, you can:
- Create a default queue with a small worker type. For example, A5.
- Create a second queue called
large-task
with a larger worker type. For example, A10. - Set the Minimum Worker Count for the
large-task
queue to 0 if your resource-intensive tasks run infrequently. - In your DAG, assign the larger task to the “large-task” queue.
To learn more about worker queues, see Worker queues in Astronomer documentation.
What if I need additional ephemeral storage for workers?
All Astro workers include an amount of ephemeral storage by default: 10 GiB of for Celery workers, and 5 GiB for Kubenetes Executor and Kubernetes Pod Operator workers. You can configure additional ephemeral storage at a rate of $0.0002 per GiB per/hour.
How will I be charged for the Kubernetes Executor and Kubernetes Pod Operator?
In Airflow, the Kubernetes Executor and the KubernetesPodOperator allow you to run a single task in an isolated Kubernetes Pod. Astro measures the total amount of CPU and Memory that you request in your DAG for any tasks that run with the Kubernetes Executor or the KubernetesPodOperator.
To calculate cost, you will be billed for the total number of A5 workers that are needed to execute that total requested. One A5 worker corresponds to 1 CPU and 2 GiB Memory.
For example:
- If you request 10 CPU cores and 20 GiB memory for 1 task, you will be billed for 10 A5s for the duration of that task run, down to the second.
- If you have 5 concurrent tasks that each request 2 CPU and 4 GiB memory, that is a total of 10 CPU cores and 20 GiB memory. You will be billed for 10 A5 hours.
We round up to the nearest A5 worker type. If your total request totals 0.5 CPU cores and 1 GiB, we will bill you for 1 complete hour of usage for an A5 worker.