Cloud Digital Leader · Domain 6 of 6

Cloud Digital Leader Domain 6: Scaling with Google Cloud Operations

The smallest section covers cost control and operations: CapEx to OpEx, financial governance, the resource hierarchy, quotas and budgets, Google Cloud Observability tools, resilient design, the four golden signals, and SLIs, SLOs and SLAs.

Exam weight
~10%
Official objectives
2
Estimated questions estimate
5–6
Suggested study time estimate
4–6 hrs

About 11 min read Reviewed October 2026

Overview

Section 6 covers running cloud well once you are there. Objective 6.1 is cost control: how moving to the cloud shifts CapEx to OpEx and affects TCO, Google-recommended financial governance practices, the roles of people, process and technology, the resource hierarchy of resources, projects, folders and the organization node, and consumption controls such as quotas, budget thresholds, Cloud Billing reports, Dynamic Workload Scheduler and Spot VMs.

Objective 6.2 is operations and reliability: Google Cloud Observability tools, operational excellence and high availability, resilient design through redundancy, replication, scalable infrastructure and backups, the golden signals of latency, traffic, saturation and errors, and SLIs, SLOs and SLAs from DevOps and SRE. In the August 2026 guide this section fell to about 10% and lost sustainability and Customer Care.

How it's tested

Section 6 items are usually short and definitional, wrapped in a business setting. A finance lead wants spending alerts, an operations manager wants fewer outages, or a team must measure reliability. Identify whether the question is about preventing cost, seeing cost, organizing resources, observing systems or setting reliability targets.

  • Cost control tool

    A spending concern points to budgets with threshold alerts, quotas, Cloud Billing reports, Spot VMs or Dynamic Workload Scheduler.

  • Hierarchy and inheritance

    A company wants consistent policies across departments, and you apply the resource hierarchy so policies set higher up propagate down.

  • Reliability terms

    A team defines targets or contracts, and you distinguish SLIs, SLOs and SLAs or choose the right observability tool.

Reasoning it through

A finance director wants an email when a project's monthly spend reaches 80% of plan, without blocking any workloads. A resource quota would cap usage and could stop services. Cloud Billing reports show trends but do not alert by themselves. A budget with an 80% threshold rule sends the alert without limiting resources, so it is the best fit.

Objectives 2

Official objectives from the Google Cloud outline, in outline order.

6.1 Recognize how Google Cloud supports an organization's ability to control their cloud costsCapEx to OpEx and TCO, financial governance, people, process and technology, resource hierarchy, consumption controls.
What it means
This objective covers cloud financial management. You explain how moving from on-premises to cloud shifts capital expenditure to operational expenditure and changes TCO, describe Google-recommended governance practices such as deciding who owns cloud costs and using cost tools, recognize the roles of people, process and technology, describe the resource hierarchy and its benefits, and recognize consumption controls including quotas, budget thresholds, Cloud Billing reports, Dynamic Workload Scheduler and Spot VMs.
What to understand
CapEx is upfront spending on assets such as servers; OpEx is ongoing pay-as-you-go spending, which the cloud favors. TCO includes all costs over time, including staff, power and maintenance, not just purchase price. Governance means clear ownership of spending, shared visibility between finance and engineering, and regular review. The hierarchy runs from the organization node to folders, projects and resources; policies and IAM permissions set higher up are inherited below, which improves access control, compliance and auditing. Quotas cap how much of a resource can be used. Budgets with threshold rules send alerts at chosen percentages. Cloud Billing reports visualize spend. Dynamic Workload Scheduler helps obtain accelerator capacity efficiently, and Spot VMs cut cost for interruptible work.
How it shows up
Expect a spending or governance concern and a choice of the control, hierarchy level or financial concept that addresses it.
Common confusion
Believing budgets stop spending automatically; by default they alert, while quotas limit usage.
6.2 Describe the fundamental concepts of modern operations, reliability, and resilience in the cloudGoogle Cloud Observability tools, operations terms, resilient design, golden signals, SLIs, SLOs and SLAs.
What it means
This objective covers how organizations keep cloud services healthy. You describe modernizing operations with Google Cloud Observability, including Cloud Monitoring, Cloud Logging, Cloud Trace, Cloud Profiler and Error Reporting; recognize operational excellence, reliability and high availability; design resilience with redundancy, replication, scalable infrastructure and backups; measure performance with latency, traffic, saturation and errors; and recognize DevOps and SRE fundamentals such as SLIs, SLOs and SLAs.
What to understand
Cloud Monitoring tracks metrics and alerts, Cloud Logging collects logs, Cloud Trace shows request latency across services, Cloud Profiler finds code that consumes CPU or memory, and Error Reporting groups application errors. High availability keeps services running through failures, usually by running across zones or regions. Redundancy duplicates components, replication copies data, and backups allow recovery. The four golden signals are latency, traffic, saturation and errors. An SLI is a measured metric such as successful request rate, an SLO is the internal target for that metric, and an SLA is a contract with customers that carries consequences if missed. SRE applies engineering practices to operations.
How it shows up
Scenarios ask which tool reveals a problem, which design improves resilience, or which reliability term a statement describes.
Common confusion
Mixing up SLO and SLA; the SLO is an internal goal, the SLA is a customer contract.

High-yield concepts

The terms, rules and formulas this domain turns into questions most often.

  • CapEx vs. OpEx

    Upfront asset purchases vs. ongoing pay-as-you-go costs.

    Remember
    Cloud shifts spending toward OpEx.
    Tested as
    Financial model questions.
    Trap
    Thinking OpEx always means a lower total cost.

  • Total cost of ownership

    All costs over an asset's life, including staff and maintenance.

    Remember
    Compare TCO, not just list prices.
    Tested as
    Business case questions.
    Trap
    Comparing only server prices.

  • Financial governance

    Clear cost ownership, shared visibility and regular review.

    Remember
    People, process and technology together.
    Tested as
    Governance questions.
    Trap
    Treating cost control as a tooling problem only.

  • Resource hierarchy

    Organization node, folders, projects, resources.

    Remember
    Policies inherit downward.
    Tested as
    Access and policy scenarios.
    Trap
    Setting the same policy on every project by hand.

  • Quotas

    Limits on how much of a resource can be consumed.

    Remember
    Can prevent runaway usage.
    Tested as
    Hard-limit scenarios.
    Trap
    Confusing quotas with budget alerts.

  • Budgets and threshold rules

    Alerts when spending reaches set percentages.

    Remember
    Notify, do not cap, by default.
    Tested as
    Spending alert scenarios.
    Trap
    Expecting budgets to shut down resources.

  • Cloud Billing reports

    Visualize cost trends by project, service and more.

    Remember
    Visibility, not enforcement.
    Tested as
    Cost analysis questions.
    Trap
    Choosing reports when an alert is required.

  • Spot VMs and Dynamic Workload Scheduler

    Discounted interruptible VMs; efficient access to accelerator capacity.

    Remember
    Consumption options that lower cost.
    Tested as
    Batch and AI workload cost questions.
    Trap
    Running critical services on Spot VMs.

  • Google Cloud Observability

    Monitoring, Logging, Trace, Profiler, Error Reporting.

    Remember
    Each tool answers a different question.
    Tested as
    Tool selection questions.
    Trap
    Using Logging to find slow request paths.

  • Golden signals

    Latency, traffic, saturation, errors.

    Remember
    Core measures of service health.
    Tested as
    Monitoring questions.
    Trap
    Leaving out saturation.

  • SLI, SLO, SLA

    Measured metric; internal target; customer contract.

    Remember
    SLA usually less strict than the SLO.
    Tested as
    Reliability terminology.
    Trap
    Calling the SLA an internal goal.

  • Redundancy and replication

    Duplicate components; copied data across locations.

    Remember
    Remove single points of failure.
    Tested as
    Resilience design scenarios.
    Trap
    Relying on backups alone for high availability.

Side by side

Concepts candidates mix up, compared on the point that decides the answer.

Cost control mechanisms

ConceptMeaningKey differenceCommon trap
Quotas Caps on resource usage Enforce limits Expected to send spending emails
Budgets with thresholds Spending alerts at set percentages Notify without blocking Expected to stop workloads
Cloud Billing reports Visual cost breakdowns Insight into trends Treated as a control
Spot VMs Discounted interruptible compute Lower cost for fault-tolerant jobs Used for critical services

Reliability terms

ConceptMeaningKey differenceCommon trap
SLI A measured indicator such as error rate What you measure Confused with the target
SLO Internal target for an SLI What you aim for Treated as a legal contract
SLA Customer agreement with consequences What you promise externally Set stricter than the SLO

Observability tools

ConceptMeaningKey differenceCommon trap
Cloud Monitoring Metrics, dashboards and alerts Tracks health over time Used to read raw log lines
Cloud Logging Central log collection and search Records events Expected to show request latency paths
Cloud Trace Request latency across services Finds slow calls Confused with Profiler
Cloud Profiler and Error Reporting Code resource use; grouped app errors Pinpoint inefficient code and crashes Assumed to monitor infrastructure metrics

Exam traps

Each trap is an answer that looks right. Learn the clue that should change your mind.

  1. Budget as a brake

    What candidates pick
    Create a budget to prevent a test project from exceeding its limit.
    Why it looks right
    Budgets sound like spending caps.
    Why it's wrong
    Budgets send alerts at thresholds; quotas or automated responses are needed to actually limit usage.

    Clue to watch for Prevent, cap or stop usage.

  2. Per-project policy

    What candidates pick
    Apply the same security policy to each of 300 projects individually.
    Why it looks right
    It feels thorough and explicit.
    Why it's wrong
    Setting the policy at the organization or folder level lets it propagate to all projects through inheritance.

    Clue to watch for Many projects, consistent policy.

  3. SLA as internal goal

    What candidates pick
    Call the team's 99.95% internal availability target an SLA.
    Why it looks right
    SLA is the most familiar reliability term.
    Why it's wrong
    An internal target is an SLO; an SLA is an external agreement with customers and consequences.

    Clue to watch for Internal, team target, no customer contract.

  4. Backups for uptime

    What candidates pick
    Rely on nightly backups to keep a service available during a zone outage.
    Why it looks right
    Backups protect data.
    Why it's wrong
    Backups help recovery but not continuous availability; redundancy across zones keeps the service running.

    Clue to watch for No downtime allowed.

  5. Logs for latency

    What candidates pick
    Search logs to find which microservice slows down checkout.
    Why it looks right
    Logs contain lots of detail.
    Why it's wrong
    Cloud Trace follows requests across services and shows where latency occurs, which logs alone do not reveal clearly.

    Clue to watch for Slow requests across multiple services.

Answer framework

Use these steps for Section 6 questions on cost or operations.

  1. Decide whether the question is about money, organization of resources, visibility into systems, resilience design or reliability targets.
  2. For money, choose budgets for alerts, quotas for hard limits, Cloud Billing reports for insight, and Spot VMs or Dynamic Workload Scheduler for cheaper capacity.
  3. For organization and policy, place controls at the highest sensible level of the resource hierarchy so projects inherit them.
  4. For visibility, pick Monitoring for metrics and alerts, Logging for events, Trace for latency paths, Profiler for code efficiency and Error Reporting for crashes.
  5. For reliability, apply redundancy and replication for availability, then express targets as SLIs, internal SLOs and customer-facing SLAs.

Practice question

A media company's platform team says its internal goal is that 99.9% of video start requests succeed each month, measured by the ratio of successful to total requests. In SRE terms, what is the 99.9% figure?

Want more like this? The Cloud Digital Leader practice test has questions for every domain.

Checklist

0 of 11

Tick each one you can do without notes. Your ticks stay in this browser.

Study strategy

This section carries the least weight, so aim for fast, reliable recall rather than depth. Start with cost: write the CapEx and OpEx definitions, a TCO example, and a three-row table separating quotas, budgets and Cloud Billing reports by whether they limit, alert or show. Draw the resource hierarchy from the organization node down to resources and note that policies flow downward.

Then cover operations. List the five observability tools with the question each one answers, memorize the four golden signals, and practice labeling statements as SLI, SLO or SLA until it becomes automatic. Add one resilience example that uses multiple zones and replication.

Skip sustainability and the Customer Care support process unless your training predates the August 2026 change and you are sitting the older version. Plan about 4 to 6 hours, then fold this section into mixed practice tests so the definitions stay fresh alongside the larger sections.

FAQ

How much of the Cloud Digital Leader exam is Scaling with Google Cloud Operations?

Scaling with Google Cloud Operations is about 10% of the exam in the August 2026 guide, roughly 5 to 6 questions, down from about 17% in the previous version.

Is sustainability still tested?

No. The August 2026 guide removed the sustainability subsection. It is no longer listed among the topics that may be assessed.

Is Google Cloud Customer Care still on the exam?

The Customer Care process and the life of a support case were removed from the current guide, so they are not listed topics.

Do budgets stop resources when the limit is reached?

Not by default. Budgets send alerts at threshold percentages. Quotas cap usage, and automated responses to budget alerts would need extra setup.

What is the difference between an SLO and an SLA?

An SLO is an internal reliability target for a measured indicator. An SLA is an agreement with customers that includes consequences, such as credits, if it is not met.

What is Dynamic Workload Scheduler?

It is a Google Cloud capability that helps organizations obtain GPU and TPU capacity efficiently for AI and batch work, and the guide lists it as a way to control consumption and cost.

Other Cloud Computing certifications

Where the same subject shows up in other exams. Cards marked "Same subject" open the matching domain.

Test your Domain 6 knowledge

Confirm that you can reliably pick the right cost control, observability tool and reliability term before your exam.

Take the Cloud Digital Leader practice test Check your Cloud Digital Leader readiness

Keep going