Introduction to AWS deployment
Understand what an existing AWS deployment includes, what it creates, and what you operate.
읽는 시간 2분최근 업데이트: 6일 전
If your organization runs Unity Asset Manager in Amazon Web Services (AWS) with the AWS Terraform package, this section covers how to install, operate, and maintain that deployment. In an AWS deployment:
- You provision the infrastructure with Terraform and install the platform with Helm, so your assets, metadata, and user activity stay inside your own AWS environment.
- Managed AWS services provide the relational database, Kubernetes, file storage, and secret storage that the platform depends on.
참고
The AWS Terraform package is supported for existing deployments. For new deployments, including deployments on Amazon EKS, use Self-Hosted Deployment on Kubernetes.
This page describes what the deployment creates, what you provide and operate, and what AWS manages for you.
What the deployment creates
The deployment provisions AWS resources that include the following:
- An Elastic Kubernetes Service (EKS) cluster and autoscaling node groups.
- EKS add-ons for the EBS Container Storage Interface (CSI) Driver, EFS CSI Driver, Pod Identity Agent, and kube-proxy.
- A Relational Database Service (RDS) Postgres instance.
- An Elastic File System (EFS) file system.
- A Data Lifecycle Manager (DLM) lifecycle policy for EBS volumes.
- A Secrets Manager secret for configuration.
- An EC2 instance for a FlexLM license server.
The deployment enables node and pod autoscaling by default.
For the full list of resources and secure exposure, refer to Overview of the infrastructure.
Your responsibilities
You provide and operate the following:
- An AWS account, and a role with sufficient permissions to create and manage resources.
- An Amazon Virtual Private Cloud (VPC) with at least two subnets, preferably private.
- The command-line tooling that the deployment requires, including Terraform, Helm, the AWS CLI, ORAS, GNU Make, yq, and jq.
- A Unity Version Control (UVCS) server and its license. The deployment can run UVCS inside the cluster, or connect to an existing external server.
- Ongoing operations, including upgrades, backup and recovery, secret rotation, health checks, and service quota increases.
For the full list of what you need before you start, refer to Prerequisites.
What AWS manages
AWS manages the relational database, the Kubernetes control plane, file storage, and secret storage.
Operational considerations
Consider the following for this deployment type:
- Amazon charges you for the deployed services. For an estimate, refer to Service requirements.
- Your team needs working knowledge of Terraform, Kubernetes, and Helm.
- You must supply a VPC with at least two subnets.
- The deployment runs in a single Availability Zone by default. For high availability, set the Terraform variable to true. Note that this might increase the cost.
enable_multi_az - Default AWS service limits might need increases. For details, refer to Increase service limits.
- You integrate your identity provider (IdP) or single sign-on (SSO) with the platform's bundled Keycloak.