# Create

## Prerequisites

1. Install and run [Docker](https://docs.docker.com/install) on your machine.
2. Subscribe to the [AMI with GPU support](https://aws.amazon.com/marketplace/pp/B07GRHFXGM) (for GPU clusters).
3. Create an IAM user with `AdministratorAccess` and programmatic access.
4. You may need to [request limit increases](https://console.aws.amazon.com/servicequotas/home?#!/services/ec2/quotas) for your desired instance types.

## Create a cluster on your AWS account

```bash
# install the cortex CLI
bash -c "$(curl -sS https://raw.githubusercontent.com/cortexlabs/cortex/v0.40.0/get-cli.sh)"

# create a cluster
cortex cluster up cluster.yaml
```

## `cluster.yaml`

```yaml
# cluster name
cluster_name: cortex

# AWS region
region: us-east-1

# list of availability zones for your region
availability_zones:  # default: 3 random availability zones in your region, e.g. [us-east-1a, us-east-1b, us-east-1c]

# list of cluster node groups;
node_groups:
  - name: ng-cpu # name of the node group
    instance_type: m5.large # instance type
    min_instances: 1 # minimum number of instances
    max_instances: 5 # maximum number of instances
    priority: 1 # priority of the node group; the higher the value, the higher the priority [1-100]
    instance_volume_size: 50 # disk storage size per instance (GB)
    instance_volume_type: gp3 # instance volume type [gp2 | gp3 | io1 | st1 | sc1]
    # instance_volume_iops: 3000 # instance volume iops (only applicable to io1/gp3)
    # instance_volume_throughput: 125 # instance volume throughput (only applicable to gp3)
    spot: false # whether to use spot instances

  - name: ng-gpu
    instance_type: g4dn.xlarge
    min_instances: 1
    max_instances: 5
    instance_volume_size: 50
    instance_volume_type: gp3
    spot: false
  # ...

# subnet visibility for instances [public (instances will have public IPs) | private (instances will not have public IPs)]
subnet_visibility: public

# NAT gateway (required when using private subnets) [none | single | highly_available (a NAT gateway per availability zone)]
nat_gateway: none

# API load balancer scheme [internet-facing | internal]
api_load_balancer_scheme: internet-facing

# operator load balancer scheme [internet-facing | internal]
# note: if using "internal", you must configure VPC Peering to connect your CLI to your cluster operator
operator_load_balancer_scheme: internet-facing

# to install Cortex in an existing VPC, you can provide a list of subnets for your cluster to use
# subnet_visibility (specified above in this file) must match your subnets' visibility
# this is an advanced feature (not recommended for first-time users) and requires your VPC to be configured correctly; see https://eksctl.io/usage/vpc-networking/#use-existing-vpc-other-custom-configuration
# here is an example:
# subnets:
#   - availability_zone: us-west-2a
#     subnet_id: subnet-060f3961c876872ae
#   - availability_zone: us-west-2b
#     subnet_id: subnet-0faed05adf6042ab7

# restrict access to APIs by cidr blocks/ip address ranges
api_load_balancer_cidr_white_list: [0.0.0.0/0]

# restrict access to the Operator by cidr blocks/ip address ranges
operator_load_balancer_cidr_white_list: [0.0.0.0/0]

# additional tags to assign to AWS resources (all resources will automatically be tagged with cortex.dev/cluster-name: <cluster_name>)
tags:  # <string>: <string> map of key/value pairs

# SSL certificate ARN (only necessary when using a custom domain)
ssl_certificate_arn:

# list of IAM policies to attach to your Cortex APIs
iam_policy_arns: ["arn:aws:iam::aws:policy/AmazonS3FullAccess"]

# primary CIDR block for the cluster's VPC
vpc_cidr: 192.168.0.0/16

# instance type for prometheus (use an instance with more memory for clusters exceeding 300 nodes or 300 pods)
prometheus_instance_type: "t3.medium"
```

The docker images used by the cluster can also be overridden. They can be configured by adding any of these keys to your cluster configuration file (default values are shown):

```yaml
image_manager: quay.io/cortexlabs/manager:0.40.0
image_operator: quay.io/cortexlabs/operator:0.40.0
image_controller_manager: quay.io/cortexlabs/controller-manager:0.40.0
image_autoscaler: quay.io/cortexlabs/autoscaler:0.40.0
image_proxy: quay.io/cortexlabs/proxy:0.40.0
image_async_gateway: quay.io/cortexlabs/async-gateway:0.40.0
image_activator: quay.io/cortexlabs/activator:0.40.0
image_enqueuer: quay.io/cortexlabs/enqueuer:0.40.0
image_dequeuer: quay.io/cortexlabs/dequeuer:0.40.0
image_cluster_autoscaler: quay.io/cortexlabs/cluster-autoscaler:0.40.0
image_metrics_server: quay.io/cortexlabs/metrics-server:0.40.0
image_nvidia_device_plugin: quay.io/cortexlabs/nvidia-device-plugin:0.40.0
image_neuron_device_plugin: quay.io/cortexlabs/neuron-device-plugin:0.40.0
image_neuron_scheduler: quay.io/cortexlabs/neuron-scheduler:0.40.0
image_fluent_bit: quay.io/cortexlabs/fluent-bit:0.40.0
image_istio_proxy: quay.io/cortexlabs/istio-proxy:0.40.0
image_istio_pilot: quay.io/cortexlabs/istio-pilot:0.40.0
image_prometheus: quay.io/cortexlabs/prometheus:0.40.0
image_prometheus_config_reloader: quay.io/cortexlabs/prometheus-config-reloader:0.40.0
image_prometheus_operator: quay.io/cortexlabs/prometheus-operator:0.40.0
image_prometheus_statsd_exporter: quay.io/cortexlabs/prometheus-statsd-exporter:0.40.0
image_prometheus_dcgm_exporter: quay.io/cortexlabs/prometheus-dcgm-exporter:0.40.0
image_prometheus_kube_state_metrics: quay.io/cortexlabs/prometheus-kube-state-metrics:0.40.0
image_prometheus_node_exporter: quay.io/cortexlabs/prometheus-node-exporter:0.40.0
image_kube_rbac_proxy: quay.io/cortexlabs/kube-rbac-proxy:0.40.0
image_grafana: quay.io/cortexlabs/grafana:0.40.0
image_event_exporter: quay.io/cortexlabs/event-exporter:0.40.0
image_kubexit: quay.io/cortexlabs/kubexit:0.40.0
```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.cortexlabs.com/0.40/clusters/management/create.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.