An enterprise organization operates multiple Google Cloud projects under a single Cloud Billing account. The central Site Reliability Engineering (SRE) team must provision guaranteed compute capacity for three separate service projects (project-core, project-analytics, and project-services) in us-central1-a ahead of a high-traffic seasonal event. The capacity planning strategy must satisfy the following requirements: Guarantee 60 n2-standard-16 virtual machines (VMs) in zone us-central1-a across the three projects. Allow VMs across all three consumer projects to dynamically consume the reserved pool on demand. Ensure each consumer project is billed directly for the reserved capacity it actively consumes, while applying existing organizational Committed Use Discounts (CUDs). Restrict the authority to create and manage shared capacity pools strictly to the central infrastructure management project (project-infra). Which combination of configuration steps meets these requirements?

Add `project-infra` to the allowlist for the `compute.sharedReservationsOwnerProjects` organization policy constraint, create a shared on-demand reservation in `project-infra` targeting `us-central1-a` with consumer projects added, and enable CUD sharing on the Cloud Billing account.: What Shared Reservations and Organizational Policies Are Shared reservations allow a central owner project in Google Cloud to allocate and guarantee zonal Compute Engine capacity that multiple consumer projects within the same organization can consume. The compute.sharedReservationsOwnerProjects organization policy constraint provides governance by strictly restricting which projects have permission to create and manage shared reservations. How It Meets the Requirements Guarantee capacity dynamically across projects: Provisioning a shared on-demand reservation in project-infra for 60 n2-standard-16 VMs in us-central1-a and specifying project-core, project-analytics, and project-services as consumer projects enables VMs in any of those projects to consume the shared capacity pool dynamically. Accurate billing and attribution: By default, unconsumed reserved capacity is billed to the owner project (project-infra). However, as soon as a consumer project runs a matching VM that consumes reserved capacity, Compute Engine switches the billing of that consumed portion directly to the consumer project. Committed Use Discount (CUD) alignment: Enabling CUD sharing at the Cloud Billing account level ensures that resource-based or spend-based commitments automatically apply discounted rates to the billed project consuming the reservation. Administrative governance: Adding only project-infra to the allowlist for the compute.sharedReservationsOwnerProjects constraint prevents unauthorized teams from creating shared reservations elsewhere in the organization. Key Advantages and Capabilities Centralized governance: Centralizes capacity procurement while distributing usage across teams. Dynamic workload sharing: When one consumer VM stops, its reserved capacity immediately returns to the pool for other consumer projects. Billing attribution: Accurately charges active consumption to individual service teams while holding unconsumed capacity costs in the central infrastructure project. Why This Is Optimal This architecture directly adheres to Google Cloud's multi-project capacity planning best practices by pairing organization policy controls, shared zonal reservations, and billing-level CUD sharing.

Configure Shared Compute Engine Reservations Cuds | Professional Cloud DevOps Engineer

StudyCerts

Explanations

What Shared Reservations and Organizational Policies Are

Shared reservations allow a central owner project in Google Cloud to allocate and guarantee zonal Compute Engine capacity that multiple consumer projects within the same organization can consume. The compute.sharedReservationsOwnerProjects organization policy constraint provides governance by strictly restricting which projects have permission to create and manage shared reservations.

How It Meets the Requirements

Guarantee capacity dynamically across projects: Provisioning a shared on-demand reservation in project-infra for 60 n2-standard-16 VMs in us-central1-a and specifying project-core, project-analytics, and project-services as consumer projects enables VMs in any of those projects to consume the shared capacity pool dynamically.
Accurate billing and attribution: By default, unconsumed reserved capacity is billed to the owner project (project-infra). However, as soon as a consumer project runs a matching VM that consumes reserved capacity, Compute Engine switches the billing of that consumed portion directly to the consumer project.
Committed Use Discount (CUD) alignment: Enabling CUD sharing at the Cloud Billing account level ensures that resource-based or spend-based commitments automatically apply discounted rates to the billed project consuming the reservation.
Administrative governance: Adding only project-infra to the allowlist for the compute.sharedReservationsOwnerProjects constraint prevents unauthorized teams from creating shared reservations elsewhere in the organization.

Key Advantages and Capabilities

Centralized governance: Centralizes capacity procurement while distributing usage across teams.
Dynamic workload sharing: When one consumer VM stops, its reserved capacity immediately returns to the pool for other consumer projects.
Billing attribution: Accurately charges active consumption to individual service teams while holding unconsumed capacity costs in the central infrastructure project.

Why This Is Optimal

This architecture directly adheres to Google Cloud's multi-project capacity planning best practices by pairing organization policy controls, shared zonal reservations, and billing-level CUD sharing.

View source documentation

Professional Cloud DevOps Engineer

Configure Shared Compute Engine Reservations Cuds