VCAP · VCF Operations · Exam 3V0-22.25 · VMware Cloud Foundation 9

Every objective in this exam lives in one section of the blueprint. Sections 1, 2, 3 and 5 are empty, and all sixteen objectives sit under Install, Configure, Administrate, carrying eighty-three sub-objectives between them. That is the densest advanced VCF blueprint published so far, and it is why reading the sixteen headings tells you almost nothing about what you will be asked. Hence a workbook of 204 tasks, written against the eighty-three.

The examWhat you are signing up for

From the published exam guide, last updated 11 November 2025:

Source: the official exam guide for 3V0-22.25, published by Broadcom. Read it yourself before you book. Everything below is written against the objectives in that document.

ShapeWhere the eighty-three actually are

Sixteen objectives sounds evenly spread. It is not:

Capacity alone is eighteen per cent of the blueprint, and capacity with forecasting and cost together is more than a third of it. Objectives 4.11, 4.13 and 4.15 have one sub-objective each, which is not an invitation to skip them: a single sub-objective can still be several questions, and super metrics in particular turn up inside dashboard and alerting work as well.

Three productsIt is not one tool

The exam name says Operations, singular. The blueprint tests three separate products, and expects you to know which one answers which question:

Two of those carry one objective each and are easy to leave until the week before. Do not. They are separate products with their own interfaces, their own sizing and their own vocabulary, and a question about content packs or flow analysis is not something you can reason your way to from the main product.

PracticeWhere to get your hands on it

This is the most practisable of the advanced VCF exams, and by some distance. Almost every objective is a thing you click, and most of it works against whatever you already run:

• Your own environment, right now. Capacity, reclamation, rightsizing, policies, dashboards, views, reports and super metrics all work against a real estate with real history. You do not need a lab for sections D through R; you need permission and an afternoon.
• VMware Hands-on Labs, free and in a browser, for the parts you cannot touch in production. Check the catalogue for current VCF 9 Operations labs and confirm the version before you rely on one.
• A read-only account on production is enough for a surprising amount of this, and it is usually easier to get than a lab. Everything in sections D, E, F, H and Q can be rehearsed by looking.

One warning. Sections E and F change running workloads, and section U changes certificates and passwords on a live fleet. Do those in a lab or in a change window, with an owner who agreed first. A reclamation exercise that deletes somebody’s standby server is not a learning experience anybody enjoys.

Workbook204 tasks, in build order

Sections A to C are the deployment and its access control. D to G are capacity, the biggest block. H is forecasting, I is cost, J is policy, K is optimization. L to N are application monitoring and management packs. O and P are alerting and diagnosis, Q and R are dashboards and super metrics, S and T are the other two products, U to V are fleet management and governance, and W is a set of timed drills. The numbers in grey at the right are the objectives, so a weak score report points you back at a group of tasks.

Objective 4.1. What you are operating, before you operate anything with it.

1. 1Draw your VCF Operations deployment: nodes, roles, and what each one does.4.1›
2. 2Record the node types and which of them can be added after the first deployment.4.1›
3. 3Write what a Unified Cloud Proxy is and what it exists to solve.4.1›
4. 4Configure a Unified Cloud Proxy and confirm it reports as connected.4.1›
5. 5Create a Collector Group and add collectors to it.4.1›
6. 6Record what a Collector Group gives you that a single collector does not.4.1›
7. 7Move an adapter instance onto a Collector Group and confirm collection continued.4.1›
8. 8Scale a VCF Operations deployment by adding a node, and record what it required.4.1›

Still 4.1. The work that keeps the platform usable after week one.

1. 9List the Day 2 operational tasks VCF Operations offers and write one line on each.4.1›
2. 10Scale a VCF Operations for Logs deployment and record the sizing you applied.4.1›
3. 11Scale a VCF Operations for Networks deployment and record what changed.4.1›
4. 12Write the sizing inputs each of the three products asks for, side by side.4.1›
5. 13Open Automation Central and record what kinds of job it can schedule.4.1›
6. 14Configure a job in Automation Central and let it run.4.1›
7. 15Check the job history and record where a failed job reports its reason.4.1›
8. 16Write which recurring tasks you would move into Automation Central first, and why.4.1›

The last of 4.1, and the part a provider gets asked about first.

1. 17List the built-in roles in VCF Operations and what each one may do.4.1›
2. 18Create a custom role by cloning one and removing a single privilege.4.1›
3. 19Assign the role to a group and sign in as a member to test the boundary.4.1›
4. 20Record the error a user sees when a role denies an action.4.1›
5. 21Restrict a user to one object scope and confirm they see nothing else.4.1›
6. 22Configure a new integration within VCF Operations and confirm it collects.4.1›
7. 23Record what the integration added: objects, metrics, dashboards, or all three.4.1›
8. 24Remove an integration and record what stops being collected.4.1›

Objective 4.2 begins. Fifteen sub-objectives, the largest single block in the blueprint.

1. 25Write what the demand capacity model measures, in one sentence.4.2›
2. 26Write what the allocation capacity model measures, in one sentence.4.2›
3. 27Build a table comparing the two: what each counts, and what each ignores.4.2›
4. 28Record which model your environment uses now, and where that is set.4.2›
5. 29Find a cluster and record its remaining capacity under each model.4.2›
6. 30Explain why the two numbers differ, using that cluster as the example.4.2›
7. 31Write which model fits a service provider selling committed resources, and why.4.2›
8. 32Write which model fits an internal platform chasing efficiency, and why.4.2›
9. 33Record what changes in the interface the moment you switch model.4.2›

Still 4.2. Where the capacity you already paid for is hiding.

1. 34Open the reclamation view and record every category it reports.4.2›
2. 35Record the total reclaimable capacity in your environment, by category.4.2›
3. 36Configure the resource reclamation settings and record each threshold.4.2›
4. 37Reclaim resources from a powered off virtual machine.4.2›
5. 38Record how long a machine must be powered off before it is listed, and where that is set.4.2›
6. 39Identify an idle virtual machine and record the metrics that made it idle.4.2›
7. 40Reclaim resources from an idle virtual machine.4.2›
8. 41Reclaim space from virtual machine snapshots, and record the age threshold used.4.2›
9. 42Find orphaned disks and reclaim the space they hold.4.2›
10. 43Exclude a virtual machine from reclamation and prove it disappeared from the list.4.2›
11. 44Write the approval process you would put around reclamation before running it for real.4.2›

The other half of 4.2, and the one that changes running machines.

1. 45Write the process to rightsize a workload, from recommendation to change.4.2›
2. 46Find an oversized virtual machine and record what made it oversized.4.2›
3. 47Find an undersized virtual machine and record the symptom in the metrics.4.2›
4. 48Record the recommended size against the current size for both.4.2›
5. 49Configure the policy settings that drive rightsizing, and record each one.4.2›
6. 50Change a rightsizing setting and record how the recommendation moved.4.2›
7. 51Exclude a virtual machine from rightsizing and confirm it is no longer recommended.4.2›
8. 52Record which rightsizing actions need downtime and which do not.4.2›
9. 53Write what you would check before accepting a rightsizing recommendation on a database server.4.2›

The advanced end of 4.2.

1. 54Write the process steps to configure the allocation capacity model.4.2›
2. 55Set the allocation ratios for CPU, memory and storage, and record each.4.2›
3. 56Record what happens to remaining capacity when you change an overcommit ratio.4.2›
4. 57Configure the advanced capacity management features and record what each adds.4.2›
5. 58Write what a custom profile is and the question it answers.4.2›
6. 59Create a custom profile describing a typical workload in your estate.4.2›
7. 60Run the profile against a cluster and record how many instances it says will fit.4.2›
8. 61Create a second profile and compare the two answers on the same cluster.4.2›

Objective 4.3, seven sub-objectives, all of them planning questions.

1. 62Write what a what-if scenario is for, and what it does not change.4.3›
2. 63Run a capacity forecast and record how far ahead it projects.4.3›
3. 64Build a what-if scenario adding workloads to a VCF private cloud.4.3›
4. 65Build a what-if scenario removing workloads, and record the capacity returned.4.3›
5. 66Build a what-if scenario adding hosts to a private cloud.4.3›
6. 67Build a what-if scenario removing hosts, and find the point where capacity runs out.4.3›
7. 68Build a what-if scenario adding HCI hosts and record what it asks for that a standard host does not.4.3›
8. 69Build a what-if scenario migrating workloads to or from a VMware Cloud.4.3›
9. 70Build a what-if scenario migrating workloads to or from a public cloud, and record the cost comparison.4.3›
10. 71Commit more than one scenario at once and record what committing actually did.4.3›

Objective 4.4. The part that gets shown to people who do not log in.

1. 72Complete the cost management configuration and record what it needed first.4.4›
2. 73Change the currency setting and confirm the figures reported in it.4.4›
3. 74Write the difference between cost and price in one sentence each.4.4›
4. 75Write the difference between the provider view and the consumer view.4.4›
5. 76List the cost drivers and record where each default value comes from.4.4›
6. 77Update a cost driver with your real figure and record how the cost per workload moved.4.4›
7. 78Create a pricing or rate card.4.4›
8. 79Update the rate card and confirm the change reached the consumer view.4.4›
9. 80Configure chargeback and produce a figure for one tenant or business unit.4.4›
10. 81Configure billing for resources and record what a recipient would actually receive.4.4›

Objective 4.5, three sub-objectives that quietly control the other fifteen.

1. 82Draw the policy hierarchy in VCF Operations, from default downwards.4.5›
2. 83Record which policy applies to an object when two could, and why.4.5›
3. 84Create a policy and record every group of settings it contains.4.5›
4. 85Update a policy and record what recalculated afterwards.4.5›
5. 86Assign a policy to a specific inventory object.4.5›
6. 87Confirm from the object side which policy it is now using.4.5›
7. 88Change one threshold in a policy and find the alert that changed because of it.4.5›

Objective 4.6. Mostly differentiate and identify verbs, so precision matters.

1. 89Write what business intent is, in one sentence.4.6›
2. 90Write what operational intent is, in one sentence.4.6›
3. 91Give one concrete example of each, and check you have not swapped them round.4.6›
4. 92Write the difference between host-based and cluster-based business intent.4.6›
5. 93Write what the Moderate operational intent mode optimises for.4.6›
6. 94Write what the Consolidate mode optimises for, and what it costs.4.6›
7. 95Write what the Balanced mode optimises for.4.6›
8. 96Write what Buffer does in operational intent and what happens when it is too small.4.6›
9. 97Record the process stages for configuring operational intent.4.6›
10. 98Record the process stages for configuring business intent, and run one.4.6›
11. 99Write the benefit of Advanced Workload Placement when VCF Automation is in play.4.6›

Objective 4.7 begins. Grouping is what makes everything else readable.

1. 100Create a Custom Group with a static membership.4.7›
2. 101Create a Custom Group with a dynamic membership rule.4.7›
3. 102Add a machine that matches the rule and confirm it joined on its own.4.7›
4. 103Record what a Custom Group lets you do that an inventory folder does not.4.7›
5. 104Apply a policy to a Custom Group and confirm the members picked it up.4.7›
6. 105Create a Business Application and add its tiers.4.7›
7. 106Open the Business Application view and record what it shows that a group does not.4.7›
8. 107Break one member deliberately and record how the application view reported it.4.7›

The application monitoring half of 4.7.

1. 108Write the difference between Service Discovery, Managed Telegraf and open-source Telegraf.4.7›
2. 109Record what each one requires on the guest, and what it does not.4.7›
3. 110Configure Service Discovery and let it run against a group of machines.4.7›
4. 111Record the services it found, and one it did not.4.7›
5. 112Deploy the Managed Telegraf agent to a virtual machine.4.7›
6. 113Record the prerequisites the deployment needed: credentials, ports, guest tools.4.7›
7. 114Configure the agent to collect operating system metrics and confirm they arrive.4.7›
8. 115Configure the agent to collect metrics from one application and confirm they arrive.4.7›
9. 116Monitor an application end to end using Application Monitoring.4.7›
10. 117Record what you would do for an application Telegraf has no plugin for.4.7›

Objective 4.8, including the one everybody forgets, the builder.

1. 118Record what a Management Pack adds, in objects, metrics and dashboards.4.8›
2. 119Install a Management Pack.4.8›
3. 120Configure the Management Pack with its endpoint and credentials.4.8›
4. 121Confirm it is collecting and find the objects it created.4.8›
5. 122Monitor Kubernetes through a Management Pack and record what becomes visible.4.8›
6. 123Write the process steps to create a Management Pack with Management Pack Builder.4.8›
7. 124Record what a Management Pack costs you at upgrade time.4.8›

Objective 4.9 begins. Nothing here works until you have triggered it once.

1. 125Open an existing alert definition and record its symptoms, impact and recommendations.4.9›
2. 126Configure an alert and set its criticality.4.9›
3. 127Create a Custom Alert from your own symptom definition.4.9›
4. 128Trigger the alert deliberately and confirm it fired.4.9›
5. 129Clear the condition and confirm the alert cancelled itself.4.9›
6. 130Configure an outbound plugin and test the connection.4.9›
7. 131Record the outbound plugin types available and what each one integrates with.4.9›
8. 132Configure a notification and bind it to your alert.4.9›
9. 133Trigger it again and confirm the notification was delivered.4.9›
10. 134Record what happens to notifications when the outbound target is unreachable.4.9›

The diagnostic half of 4.9, and the only troubleshooting in this blueprint.

1. 135Open the Troubleshooting Workbench on an object and record the four things it shows.4.9›
2. 136Use the workbench to find what changed on an object around a known time.4.9›
3. 137Use the workbench on a real or staged performance problem and write your conclusion.4.9›
4. 138Record which evidence the workbench gave you, and which you still had to find elsewhere.4.9›
5. 139Write the order you would work a slow virtual machine complaint in this tool.4.9›
6. 140Upload a VCF log bundle using Log Assist.4.9›
7. 141Record what Log Assist collected, and how long the upload took.4.9›
8. 142Record when you would use Log Assist rather than collecting bundles by hand.4.9›

Objective 4.10. Three sub-objectives, and the most visible work you will do.

1. 143Record the difference between a dashboard, a view and a report.4.10›
2. 144Create a custom dashboard and add three different widget types.4.10›
3. 145Configure a widget interaction so one widget drives another.4.10›
4. 146Share the dashboard with a group and confirm they can see it.4.10›
5. 147Create a custom view and choose its subject and columns.4.10›
6. 148Add the view to a dashboard and confirm it renders.4.10›
7. 149Create a custom report from one or more views.4.10›
8. 150Schedule the report and confirm it was delivered.4.10›
9. 151Write the one question your dashboard answers, then remove every widget that does not help.4.10›

Objective 4.11. One sub-objective, and more marks than that suggests.

1. 152Write what a super metric is and when a normal metric will not do.4.11›
2. 153Create a super metric with a simple formula.4.11›
3. 154Associate the super metric with an object type and enable it in a policy.4.11›
4. 155Confirm it is collecting values, and record how long before data appeared.4.11›
5. 156Build a super metric that rolls a child metric up to a parent object.4.11›
6. 157Put the super metric on a dashboard and confirm it reads correctly.4.11›

Objective 4.12. A separate product with its own habits.

1. 158Confirm VCF Operations for Logs is receiving events, and from what.4.12›
2. 159Configure a source for event monitoring and confirm its events arrive.4.12›
3. 160Run an interactive query and narrow it to one event type.4.12›
4. 161Create a log event alert query.4.12›
5. 162Trigger the condition and confirm the alert fired.4.12›
6. 163Manage dashboards in Operations for Logs: create one and add a chart.4.12›
7. 164Install a content pack and record what it added.4.12›
8. 165Manage content packs: clone one, change it, and record why cloning was needed.4.12›
9. 166Write which question you take to Logs rather than to Operations.4.12›

Objective 4.13. One sub-objective, but a whole product behind it.

1. 167Confirm VCF Operations for Networks is collecting, and from which sources.4.13›
2. 168Find a virtual machine and trace its network path end to end.4.13›
3. 169Record what the flow analysis shows that NSX alone does not.4.13›
4. 170Use it to answer one real question about east-west traffic.4.13›
5. 171Record what it tells you about a firewall rule that is never hit.4.13›
6. 172Write when you would reach for this product rather than Operations or Logs.4.13›

Objective 4.14. Five sub-objectives, and the most consequential buttons in the product.

1. 173Record what Fleet Management covers, and which components appear in your fleet.4.14›
2. 174Write the process steps for scaling VCF Fleet Management lifecycle.4.14›
3. 175Check for available updates and record what the fleet reports.4.14›
4. 176Run a pre-check on an upgrade and record everything it flagged.4.14›
5. 177Upgrade one component of a VCF fleet and record the order it enforced.4.14›
6. 178Manage SSL certificates using Certificate Management, and replace one.4.14›
7. 179Record which services had to be re-registered after the certificate changed.4.14›
8. 180Manage passwords using Password Management and rotate one account.4.14›
9. 181Record what a failed rotation looks like and how you would recover it.4.14›
10. 182Configure VCF SSO through Identity and Access Management, and keep a way back in.4.14›

Objectives 4.15 and 4.16, the governance end of the product.

1. 183Open License Management and record every license the fleet is using.4.15›
2. 184Add or update a license and confirm it applied to the right components.4.15›
3. 185Record what the product tells you as a license approaches expiry.4.15›
4. 186Run a Compliance Benchmark and record the score it returned.4.15›
5. 187Pick one failed control, fix it, and re-run the benchmark to prove the change.4.16›
6. 188Record which benchmarks are available and which applies to your estate.4.16›
7. 189Open Config Drift and record what it is comparing against.4.16›
8. 190Change a setting deliberately and find it reported as drift.4.16›
9. 191Use Audit Users to find who logged on, and when.4.16›

Do this section twice in the final week, from a clean start, timing yourself.

1. 192Configure a Collector Group and move an adapter onto it. 15 minutes.4.1›
2. 193Configure resource reclamation and reclaim from three categories. 25 minutes.4.2›
3. 194Rightsize one oversized and one undersized workload. 20 minutes.4.2›
4. 195Run a what-if scenario adding hosts, and commit it. 20 minutes.4.3›
5. 196Configure a rate card and produce a chargeback figure. 25 minutes.4.4›
6. 197Create a policy, assign it to an object, and prove it applied. 15 minutes.4.5›
7. 198Configure operational intent and business intent on one cluster. 20 minutes.4.6›
8. 199Deploy Managed Telegraf and collect one application metric. 30 minutes.4.7›
9. 200Create a custom alert, an outbound plugin and a notification, then fire it. 25 minutes.4.9›
10. 201Build a dashboard, a view and a scheduled report. 30 minutes.4.10›
11. 202Create a super metric and show it on a dashboard. 15 minutes.4.11›
12. 203Replace a certificate and rotate a password through Fleet Management. 25 minutes.4.14›
13. 204Run a compliance benchmark and resolve one finding. 20 minutes.4.16›

Four weeksIf that is what you have

Two pieces of advice. The allocation against demand comparison in section D settles more questions than anything else in this blueprint, so learn it until you can write the table from memory. And objective 4.6 is almost entirely differentiate and identify verbs: business intent against operational intent, and the three operational modes. Those are definition questions, they are easy marks, and they are also the easiest thing to get backwards under time pressure.

Exam details are from the published Broadcom exam guide for 3V0-22.25, last updated 11 November 2025, and are accurate at the time of writing. Broadcom revises blueprints regularly, so confirm the current version before you book. Practice tasks here are my own, written from the published objectives; they are not exam content.