1. Docs
  2. Administration
  3. Self-Hosting
  4. Operations
  5. Compute Sizing

Compute Sizing

    Self-hosting is only available with Pulumi Business Critical. If you would like to evaluate the self-hosted Pulumi Cloud, sign up for the 30-day trial or contact us.

    The API service and console are stateless containers. This page documents recommended container resource allocations and minimum infrastructure requirements.

    For component configuration details, see Components.

    ServiceCPUMemoryNotes
    API2048m (2 vCPU)4096 Mi (4 GB)Scale horizontally for HA
    Console512m (0.5 vCPU)512 MiStatic web UI, low resource usage
    Migrations128m128 MiRuns once per upgrade before service rollout

    For production, 2 vCPU / 4 GB RAM per API instance is a good starting point. Scale horizontally (more instances) rather than vertically for the API service, since it is stateless and benefits from running behind a load balancer across multiple AZs.

    The ECS installer exposes apiTaskCpu and apiTaskMemory Pulumi config options for customizing API service resources. Other installers can adjust resource requests/limits directly in the Kubernetes manifests.

    Minimum infrastructure requirements

    Production (high availability)

    ComponentSpecification
    DatabaseMySQL 8.0.x, multi-AZ, 2+ instances (writer + reader), 20 GB+ SSD
    Object storageS3-compatible, 200 GB+ SSD, versioning enabled
    Compute2+ instances/pods across 2+ AZs, 2 vCPU / 4 GB RAM per instance minimum
    Load balancerApplication load balancer across 2+ AZs
    NetworkVPC with public and private subnets in 2+ AZs
    DNSTwo A records: api.{domain} and app.{domain}
    TLSCertificates for both API and console domains