A dedicated cluster exclusively runs Deployments from your Organization within a single-tenant environment on Astronomer’s cloud. Dedicated clusters provide more configuration options for regions, connectivity, and security than standard clusters. You might want to create a dedicated cluster if:
Dedicated clusters offer the self-service convenience of a fully managed service while respecting the need to keep data private, secure, and within a single-tenant environment. If you don’t need the aforementioned features, you can use one of the standard clusters when you Create a Deployment.
Each cloud provider has specific requirements for CIDR size, based on underlying infrastructure behavior and observations from customer deployments. Astro doesn’t accept values smaller than these minimums.
Astronomer can’t provide an exact match between network settings and workload size or the number of Deployments, because many factors influence address space utilization. Astronomer engineers provide specific recommendations and guidance after reviewing your use case requirements.
In the Astro UI, click Organization Settings, then click Clusters.
Click + Cluster.
Configure the following details about your cluster:
172.20.0.0/20, and the minimum size you can use is /22.(Optional) Expand Advanced Configuration to further define Kubernetes Pod Networking:
Astro allocates a secondary VPC CIDR range specifically for Kubernetes Pod IPs within your cluster. Pods use source network address translation (SNAT) for connections outside the VPC, so your data sources see connections from the primary VPC subnet, not the Pod CIDR. Only traffic internal to the VPC, such as to RDS databases, appears from Pod IPs in the secondary CIDR.
Pod CIDR Range: Specify the range for Pod IP addresses. This range must not overlap with your VPC Subnet Range or any connected networks.
The Pod CIDR range isolates Pod networking and mitigates Pod IP exhaustion. SNAT ensures that all external connections use the main VPC subnet, simplifying integration with external services.
(Optional) In the Disaster Recovery section, configure cross-region disaster recovery:
Cross-region disaster recovery requires the Enterprise Business Critical tier.
Cross-Region Disaster Recovery: Enable to create a secondary DR cluster in a different AWS region.
Failover Region: Select the AWS region for the secondary cluster.
DR VPC Subnet Range: Specify a different VPC subnet range for the secondary cluster, or leave blank to use the same range as the primary cluster.
DR Pod CIDR Range: Specify a different Pod CIDR range for the secondary cluster, or leave blank to use the same range as the primary cluster.
Task Logs Replication SLA: Enable to guarantee a 15-minute recovery point objective (RPO) for task logs. Additional charges apply.
See Disaster recovery for details on how these options affect your DR configuration.
Click Create cluster. After Astro finishes creating the cluster, users in your Organization can select the cluster when they create a Deployment.