DRAG DROP (Drag and Drop is not supported)
Drag and drop the characteristics from the left onto the correct data processing techniques on the right, in the context of GDPR.

Select and Place:

See Explanation section for answer.

Answer(s): A

Explanation:

Show Answer Next Question

QUESTION: 18

Which two data encoding techniques are supported by gRPC? (Choose two.)

XML
JSON
ASCII
ProtoBuf
YAML

Answer(s): B,D

Reference:

https://grpclib.readthedocs.io/en/latest/encoding.html

Show Answer Next Question

Refer to the exhibit. Which two functions are performed by the load balancer when it handles traffic originating from the Internet destined to an application hosted on the file server farm? (Choose two.)

Terminate the TLS over the UDP connection from the router and originate an HTTPS connection to the selected server.
Terminate the TLS over the UDP connection from the router and originate an HTTP connection to the selected server.
Terminate the TLS over the TCP connection from the router and originate an HTTP connection to the selected server.
Terminate the TLS over the TCP connection from the router and originate an HTTPS connection to the selected server.
Terminate the TLS over the SCTP connection from the router and originate an HTTPS connection to the selected server.

Answer(s): C,D

Show Answer Next Question

QUESTION: 20

Which transport protocol is used by gNMI?

HTTP/2
HTTP 1.1
SSH
MQTT

Answer(s): A

Show Answer Next Question

Download PDF Next Page

Viewing page 5 of 78
Viewing questions 21 - 25 out of 507 questions

Post your Comments and Discuss Cisco 350-901 exam prep with other Community members:

Comments:

Name:

350-901 Exam Discussions & Posts (Share your experience with others)

Alex on: March 02, 2025
Great material
UNITED KINGDOM

Remy on: May 28, 2026
The AI check is a really good addition as some questions might be confusing and/or false.
Villepinte, France

AI Explanation - verified by User5825 on: May 28, 2026
Question 11:

The goal: Give a user permission to upload (push) container images to an Azure Container Registry with the least privileges.

Why the other options are less appropriate:

- Owner or Contributor: provide broad control over the resource and beyond what’s needed. - AcrPull: allows only downloading images, not uploading. - AcrPush: specifically allows pushing images to the registry, which is exactly the required capability with minimal privileges.

Answer: AcrPush (option C) is the correct role.

Quick deployment note:

- Assign at the registry level (or a specific repository if supported) to limit scope, e.g. via CLI: - az role assignment create --assignee <user> --role "AcrPush" --scope /subscriptions/<sub>/resourceGroups/<rg>/providers/Microsoft.ContainerRegistry/registries/<registry>
Villepinte, France

Sree Parvathy J on: May 28, 2026
Qn answers are provided with clear explanations.
Bengaluru, India

AI Explanation - verified by crazyboymech on: May 28, 2026
Question 21:
Question 21 is about reducing unintended bias when predicting UK driver insurance prices.

Correct answer: B

Why: Taking a training sample that is representative of the UK population ensures the model learns from data that reflects the target users. This helps reduce sampling bias and improves generalization across that population.

Why the other options are weaker:

- A: Removing protected characteristics from the data before sampling doesn't guarantee fairness and may remove useful information needed to model legitimate differences; it’s not sufficient on its own. - C: Using data from global insurers may skew the model away from the UK population, introducing distributional bias. - D: A completely random sample may not be representative of the UK population, so bias can persist.
Chennai, India

Emil on: May 28, 2026
Yes i just want to practice to understand overall sturcture of exams.
Tbilisi, Georgia

AI Explanation - verified by vjsri1616 on: May 27, 2026
Question 69:
Question 69 asks for the least-implementation-effort way to tailor the model’s style based on the user’s age.

Correct answer: B.

Why: Adding a role description in the prompt is prompt engineering—you tell the model the target age range and it adapts style without changing the model itself or collecting new data.

Why the others are less suitable:

- A (fine-tuning) requires collecting data, training, and deployment—heavy effort. - C (chain-of-thought) adds unnecessary reasoning steps and isn’t specifically about style control. - D (summarizing by age) might adjust length but not reliably adjust overall style or depth; it’s a post-processing approach, not a direct style cue.
Example prompt snippet:

"You are a helpful tutor explaining concepts to a reader aged 10. Use simple language and shorter sentences."

Plano, United States

AI Explanation - verified by vjsri1616 on: May 27, 2026
Question 65:
Question 65 asks how to get more deterministic (consistent) outputs from an LLM on Bedrock when doing sentiment analysis.

Correct answer: A (Decrease the temperature value).

Why: The temperature parameter controls sampling randomness. Lowering temperature makes the model’s outputs more deterministic, so the same input prompt is more likely to produce the same response across runs.

Why the others are less suitable:

- Increase the temperature would make outputs more random (less consistent). - Decreasing the length of output tokens or increasing the maximum generation length changes how much text is produced, not how consistent the content is.
Tip: If you need even more consistency, you can also stabilize prompts or adjust other decoding settings (e.g., top_p) alongside a low temperature.
Plano, United States

AI Explanation - verified by vjsri1616 on: May 27, 2026
Question 59:
The correct answer is Confusion matrix (A).
Why:

A confusion matrix is a table that shows, for each material class, how many images were correctly classified (true positives) and how many were misclassified (false positives, false negatives, true negatives). This is the standard tool for evaluating classification performance, especially with multiple classes.

From it you can compute per-class precision, recall, F1, and overall accuracy.

Why the others aren’t ideal:

Correlation matrix isn’t a primary performance metric for classifiers; it shows relationships between variables, not classification errors.

R2 score and Mean Squared Error (MSE) are regression metrics, not suited to discrete class labels.

Practical takeaway:

Use the confusion matrix to see which material types are often confused with each other and to compute overall accuracy and per-class metrics (precision/recall/F1) to judge model performance.

Plano, United States

AI Explanation - verified by User7098 on: May 27, 2026
Question 42:
Question 42 asks you to identify two things in a star schema:

Which table is the fact table?

Which column in the Patient table is the surrogate key (the dimension’s primary key used inside the fact table)?

How to determine:

The fact table is the one that contains the measurable facts (metrics) and the foreign keys to the dimension tables. Look for columns that are numeric measures (e.g., counts, costs, quantities) along with keys that reference dimensions (Date_SK, Patient_SK, Procedure_SK, etc.).

The surrogate key in the Patient dimension is the key that uniquely identifies each patient inside the data warehouse (not the business/natural key). This column is typically named something like Patient_SK or PatientKey. The natural key (e.g., PatientID) may also exist, but the surrogate key used in joins to the fact table is the surrogate key.

What to do with the image:

If the image shows a table with measures and FKs, that’s the fact table.

If it shows a Patient table, the surrogate key column is the one named something like Patient_SK.

If you share the exact table names from the image, I can confirm which option corresponds to A.
Lagos, Nigeria

AI Explanation - verified by juliefe.bagnes on: May 27, 2026
is Contact Key same as Subscriber Key?
Not exactly. They’re two different identifiers, though they can be mapped.

Contact Key: The unique identifier for a contact across all channels in Marketing Cloud (Contact Builder). It’s the cross-channel identity.

Subscriber Key: The unique identifier for a subscriber within Email Studio / All Subscribers. It’s the email-send identity.

Often people set them to the same value to keep things simple, but it’s not required. If you want true omni-channel consistency, you’ll typically align them (or maintain a clear mapping) so one contact maps to the same subscriber key across systems.
Calamba, Philippines

AI Explanation - verified by juliefe.bagnes on: May 27, 2026
what are hard bounce and soft bounce
In email marketing, a message can “bounce” if it can’t be delivered. There are two types:

Hard bounce: Permanent delivery failure. Examples include an invalid email address, a non-existent domain, or a recipient server that blocks delivery. Action: remove or suppress the address from your list and don’t keep trying.

Soft bounce: Temporary delivery failure. Examples include a full mailbox, server downtime, or a message that’s too large. Action: retry delivery (Marketing Cloud will typically try again); if it continues to bounce, the address may eventually be treated as a hard bounce or removed after multiple attempts.

Key takeaways:

Hard bounces signal bad addresses and hurt sender reputation if ignored.

Soft bounces are temporary; monitor and re-send as appropriate, but don’t overdo it.

Calamba, Philippines

AI Explanation - verified by fadouayahia on: May 27, 2026
the answer is webhook instead of service endpoint which one is accurate
Short answer: For Question 334, the accurate choice is the webhook (Option A).

Why webhook is correct here:

- The drag-and-drop steps explicitly describe: create an Azure Function, refactor code, publish, then register a webhook in the Plug-in Registration tool, and register a step on the webhook. This clearly follows the webhook pattern. - The approach reuses the existing plug-in logic by moving it into the Azure Function and invoking it via the webhook after record creation.

Why not Service Endpoint here:

- The Service Endpoint pattern (with an Azure Service Bus queue) would involve registering a service endpoint, enqueuing messages, and having a separate queue-processing flow. That wiring and the required steps don’t match the sequence given in this question.

When to choose each:

- Webhook: simpler, quicker to implement, good for async invocation using existing logic. - Service Endpoint + queue: more resilient to outages and supports buffering/retries, but requires additional components and a different setup.
If you want, I can map the exact steps for both patterns to show where they diverge.
Inlet Beach, United States

Jessica on: May 27, 2026
Explanation of each answer. When there is an help reference is better
Anonymous

AI Explanation - verified by jessicaztto on: May 27, 2026
Question 47:
Question 47 asks about what Territory Management can grant in terms of record access.

Correct answer: A. grants access to records regardless of ownership.

Why:

In Territory Management, access to records is determined by territory membership rather than who owns the record. If a user is in a territory that has access to a record (through territory-based sharing), they can view/edit it even if they’re not the owner.

This is in contrast to access that’s based solely on ownership, or on profiles, or on the role hierarchy (which are other sharing mechanisms).

Key takeaway:

Territory Management shifts access logic from owner-based to territory-based, so membership in the appropriate territory grants access to the associated records.

São Paulo, Brazil

AI Explanation - verified by User2464 on: May 27, 2026
Question 1:
The correct answer is B: a reduced workload for the customer service agents.
Why:

A webchat bot handles many routine, repetitive customer questions automatically, 24/7. This reduces the number of inquiries that human agents need to handle.

As a result, agents can focus on more complex issues, which improves efficiency and reduces staff workload.

Increasing sales or improving product reliability aren’t direct, primary benefits of a chat bot in this scenario. The bot’s main value in this context is operational efficiency and workload reduction.

Lagos, Nigeria

AI Explanation - verified by User4419 on: May 27, 2026
Question 19:
Question 19 asks: A security analyst received a malicious binary file to analyze. Which of the following is the best technique to perform the analysis?

Answer: Reverse engineering (option C)

Why:

A malicious binary is compiled code. To truly understand its behavior, payload, and how it operates (including any obfuscation or anti-analysis tricks), you typically perform reverse engineering. This disassembles or decompiles the binary to reveal the logic and flow.

Static analysis (option B) can reveal strings, imports, and indicators without executing code, but may miss hidden logic or runtime behavior.

Fuzzing (option D) tests how the program handles inputs and is not aimed at understanding the malware’s internal mechanisms.

“Code analysis” (option A) is vague and usually overlaps with static analysis; it doesn’t inherently provide the deep insight into binary behavior that reverse engineering does.

Practical note: conduct reverse engineering in a controlled, isolated environment to avoid accidentally triggering the payload on production systems.
Hermanus, South Africa

AI Explanation - verified by User9860 on: May 27, 2026
Question 68:
Question 68 asks for a shared, editable set of test events for Lambda in a development environment.

Correct answer: B

Why B fits: DynamoDB is serverless and scales automatically. It supports fine-grained IAM access, so all members of an IAM developer group can read and write test events. You can store each test event as an item (or multiple items) and update them easily with PutItem/UpdateItem, making collaboration straightforward and cost-effective.

Why not A (S3): S3 is great for storing objects, but editing JSON objects is more manual (download, modify, re-upload) and there’s no native, easy “multi-user editing” workflow. The wording about “event sharing settings” isn’t a standard feature of S3.

Why not other possibilities: ElastiCache is for in-memory caching, not durable test data. If the data needs to persist and be edited by many users, a database like DynamoDB is more appropriate and cost-efficient for this use case.

Summary: DynamoDB provides a simple, scalable, cost-effective way to share and collaboratively edit test events among a group of developers using IAM permissions.

Toronto, Canada

AI Explanation - verified by User4930 on: May 27, 2026
Question 1:
Great question. In Question 39, Box 1 is Knowledge mining, not NLP, because:

The scenario says: “use Azure Cognitive Search's knowledge mining results and populate your knowledge base of your chatbot.” This is about extracting, structuring, and indexing knowledge from large datasets to build a searchable knowledge base.

Knowledge mining focuses on turning diverse, unstructured data into a structured knowledge base for retrieval by a chatbot.

Natural language processing (NLP) is about understanding and analyzing text to extract meaning, sentiment, or intents (e.g., sentiment analysis, topic detection). That’s not the primary task described here.

So while NLP might be involved somewhere in building a chatbot, the specific task in Box 1 is to mine knowledge and populate a knowledge base, which is why the correct match is Knowledge mining. If you want, I can walk through another example to contrast NLP vs knowledge mining.
Secaucus, United States

AI Explanation - verified by User8972 on: May 27, 2026
Question 3:
I can’t view the interactive hotspot image, but here’s how to approach Question 3 conceptually.
What it tests

End-to-end incident analysis: identify which host originated the infection, and classify the remaining hosts as clean or infected based on logs.

How to approach it

Gather logs from multiple sources: network firewall logs, IDS/IPS alerts, DNS logs, DHCP, and host-level logs (process trees, startup items, AV alerts).

Establish timing: synchronize times and look for the earliest indicator of compromise across devices.

Trace the infection chain: find the first host that initiated malicious activity (e.g., outbound connection to a malicious external host, delivery of a payload, or anomalous process creation).

Determine origin: the host that shows the initial infection activity or first contact with the malware/command channel.

Classify others: mark hosts as infected if they show IOCs (unusual outbound traffic, known malware artifacts, lateral movement patterns) or as clean if they lack such indicators.

Validate with cross-checks: ensure your conclusion is supported by multiple logs and that time/sequence is consistent.

If you can share the log details or describe which hosts show what events, I can help confirm which one is the origin and which hosts are infected or clean. The page’s key lists A as the answer, which means the origin host corresponds to option A on the hotspot map.
La Puente, United States

AI Explanation - verified by User7166 on: May 27, 2026
Question 31:

Correct answer: B

Explanation:

In Palo Alto logs, the status "incomplete" for a TCP session means the session did not complete the TCP handshake, so the firewall cannot identify the application from the traffic.

If the three-way handshake were observed and then the app couldn't be identified, the log would typically reflect an unknown/unidentified application, not necessarily "incomplete."

UDP traffic does not use a TCP handshake, so "incomplete" wouldn’t apply to UDP.

Data arriving and being discarded due to a Deny policy before App-ID could be applied would show a denial/policy drop, not an "incomplete" session classification.

Jammu, India

AI Explanation - verified by User2696 on: May 27, 2026
Question 2:
Question 2 explains which connector to use when your Power Apps project management app is hosted in Teams.
Key point

The app’s data is stored in Dataverse (often Dataverse for Teams when inside Teams). To pull that data into Power BI, use the Dataverse connector.

Why not the other options

Microsoft Teams Personal Analytics is for Teams usage analytics, not app data.

SQL Server database would apply only if the data lived in SQL Server, which isn’t indicated here.

Dataverse Dataflows is about connecting to dataflows, not directly the app’s Dataverse data source.

How to use

In Power BI Desktop: Get Data > Dataverse > sign in with your Microsoft account > pick the Teams Dataverse environment and the project management tables you need.

Notes

If your data is indeed in Dataverse for Teams, the standard Dataverse connector will work (you just need appropriate permissions).

Anonymous

khushi on: May 27, 2026
real time questions
Dehradun, India

AI Explanation - verified by florenceamarachi90 on: May 27, 2026
Question 61:
Here’s the logic and the three actions in the right order, with why each step is needed.
Goal: Produce a denormalized table that includes only managers who are associated with a sales region.

Step 1: Merge Sales_Region and Sales_Manager using an inner join.

- Why: Keeps only regions that have a corresponding sales manager. Filters out regions without a manager, so you’re starting with the valid region-to-manager mappings.

Step 2: Merge Region_Manager and Manager using an inner join.

- Why: Keeps only region-managers that exist in the Manager table. Removes any Region_Manager rows that don’t have a base Manager record.

Step 3: Merge the result with Sales_Region and Region_Manager using a right join (new query named Sales_region_and_Region_Manager).

- Why: Combines the validated region-manager data into a single denormalized table, ensuring you retain all relevant region-manager rows and their region context.
This sequence yields a single table containing only managers who are associated to a sales region.
Port Harcourt, Nigeria

AI Explanation - verified by User7973 on: May 27, 2026
Question 50:
Here’s a focused explanation of Question 50.

The correct choice is A: Disaster recovery.

Why: Azure Site Recovery (ASR) is a disaster recovery service. It keeps business apps and workloads available during outages by replicating workloads from a primary site to a secondary location and orchestrating failovers. This is exactly the core goal of DR—maintaining business continuity after a disruption.

What ASR does (high level):

- Replicates workloads running on physical or virtual machines to a secondary site (on-premises, another data center, or Azure). - Enables planned and unplanned failovers, plus failback after the issue is resolved. - Supports recovery plans and testing to validate DR readiness without affecting production.

What it is not:

- Not fault tolerance (it doesn’t keep individual components up during hardware failures). - Not about latency reduction or dynamic scaling. - Not a crowd-sourced or public-access resource.
Key concepts to remember:

Business continuity = keeping critical services available during outages.

DR is about recovering from a disaster; ASR provides replication, failover, and recovery orchestration to achieve that. If you’d like, I can walk through a simple failover/failback flow using ASR.

Hyderabad, India

AI Explanation - verified by User4223 on: May 27, 2026
Question 10:

Answer: No

Why:

The Azure pay-as-you-go model is an operational expenditure (OPEX) model, not an “elastic expenditure model.”

Paying for VM usage under pay-as-you-go means costs vary with actual resource consumption, which is the essence of OPEX.

“Elastic expenditure model” is not a standard Azure cost model. Elasticity refers to scaling resources up/down, not to a distinct way of billing.

In short, migrating VMs to a pay-as-you-go Azure subscription should be described as an OPEX approach; labeling it as “elastic expenditure” does not meet the goal. If you want, I can explain how OPEX vs CAPEX applies to cloud migrations and how to estimate costs.
Hyderabad, India

AI Explanation - verified by User8114 on: May 27, 2026
Question 10:

The correct answer is D. RegEx pattern matching.

Why:

Data Loss Prevention (DLP) in Microsoft 365 uses Sensitive Information Types (SITs). If the 32-character customer account number isn’t covered by built-in SITs, you create a custom SIT. A custom SIT is defined using a regex pattern to match the specific data format.

How to implement at a high level:

In the Microsoft 365 Compliance Center, create a new custom Sensitive Information Type that uses a regex pattern to detect 32 alphanumeric characters, e.g. a pattern like [A-Za-z0-9]{32} (refine with word boundaries or context if needed to reduce false positives).

Create a DLP policy that uses this custom SIT and apply it to SharePoint Online locations containing the documents.

Test with sample documents to tune the pattern and reduce false positives.

Why the other options are less suitable:

SharePoint search: not how DLP detects or blocks data.

Hunting query in Microsoft 365 Defender: for threat hunting, not DLP data detection.

Azure Information Protection: focuses on labeling/encryption, not detecting sensitive content for DLP.

Indore, India

AI Explanation - verified by User8114 on: May 27, 2026
Question 1:

Answer: D — review the certainty of the source computer.

Why: This Defender for Identity alert (Suspected DCSync) can fire in noisy situations. To minimize admin effort, focus on how certain Defender is about the source of the alert. If the source computer is a domain controller, Defender may not be able to confidently identify a DCSync event, so you can treat it as a lower-confidence finding (often closing or triaging as a benign activity). The goal is to quickly determine whether the alert is trustworthy based on the source’s credibility.

How this plays out:

- Check the source computer’s identity and its role (is it a DC?). - If the source is a DC, the alert may have low certainty and can be closed or marked as a B-TP activity. - If the source is not a DC, you may need to investigate further, but the certainty metric guides rapid triage.

Why not the others:

- Status update time, alert status, or resolution method of the source computer are less indicative for quickly filtering false positives in this scenario.

Key term: Suspected DCSync alert and certainty of the source computer. Reference: Defender for Identity domain-dominance alerts guidance.

Indore, India

Pradeep on: May 26, 2026
this UI is clean, provides relavent answers and the experience is very good. WIll definitly use this portal as my go to point for all my certification needs.
Bengaluru, India

AI Explanation - verified by kellum2010 on: May 26, 2026
Question 84:
The correct answer is B.
Why: The client project manager has no agile experience, but agile daily meetings require informed participation. Proposing a session for essential agile knowledge transfer helps the client team understand agile ceremonies, roles, and decision rights so they can effectively participate in daily stand-ups and align with the vendor’s goals.
Why the other options are weaker:

A: While training is good, this option is vague about immediate representation in daily meetings and doesn’t directly ensure the client can participate right away.

C: Realigning to a predictive approach would undercut the agile structure of 2–4 month iterations and reduce effective client involvement.

D: Providing documents/resources is passive and unlikely to ensure active, capable participation in daily meetings.

Practical takeaway:

Conduct a focused onboarding/coach session for the client PM and stakeholders.

Define client roles in the agile process (e.g., decision-making authority for daily issues).

Align on ceremonies, artifacts, and how the client will engage in daily meetings.

Anonymous

AI Explanation - verified by User4855 on: May 26, 2026
Question 18:
Question 18 asks you to create a BusyBox pod and run a sleep command inside it.
Key points about the solution

Command: kubectl run busybox --image=busybox --restart=Never -- /bin/sh -c "sleep 3600"

kubectl run with BusyBox creates a Pod by default.

--restart=Never ensures you get a Pod (not a Deployment or ReplicaSet).

-- separates kubectl options from the container command.

The container runs /bin/sh -c "sleep 3600", which makes the container sleep for 3600 seconds.

Pod lifecycle outcome

The container sleeps for 1 hour, then exits.

With --restart=Never, Kubernetes will not restart the container, so the Pod transitions to a Completed/Succeeded state after the sleep finishes.

Verification tips

Check status: kubectl get po busybox

Details: kubectl describe po busybox

Note: If you want the Pod to stay running, you’d use a command that doesn’t exit (e.g., a loop).
Correct answer: A
Anonymous

AI Explanation - verified by ohfrrl on: May 26, 2026
Question 388:
Question 388 asks how to reduce the attack surface in data centers.

Correct answer: D) Upgrade end-of-support operating systems.

Why:

End-of-support OSes no longer receive security patches or updates, leaving known vulnerabilities unaddressed. Upgrading to a supported OS reduces these weaknesses and closes exploitable attack vectors, thereby lowering the attack surface.

Why the other options are less effective for reducing attack surface:

Implement a honeynet: Helps with detection and research, not primarily about reducing the number of vulnerabilities.

Define Group Policy on the servers: Can enforce security settings, but it doesn’t fix underlying software vulnerabilities or missing patches.

Configure the servers for high availability: Improves uptime and resilience, not the number of vulnerabilities.

In short, keeping systems within vendor support and receiving patches is the most direct way to minimize attack surface.
Anonymous

AI Explanation - verified by User4210 on: May 26, 2026
Question 45:
Question 45 asks how to ensure a parent case can’t be closed until all its child cases are closed.
Why D (Apex trigger to check child case status before closing it) is the right approach:

Cross-record logic: You need to check related child records when closing the parent. A trigger on the parent can query its child cases and block the update if any are not closed.

Bulk-safe and reliable: Apex can handle multiple records at once and enforce the rule consistently, even in bulk operations.

Why the others are less suitable:

Validation rule on the child: This would only prevent closing a child, not enforce the rule that the parent can’t close while children are open.

Workflow on the parent to check child status: Workflows can’t reliably enforce this cross-record gating in a simple, bulk-safe way.

Rollup summary on the child: Rollups require a master-detail relationship; Case Parent is a lookup, so standard rollups don’t apply without additional tools.

So, the correct approach is to use an Apex trigger (D) to enforce the rule.
Anonymous

AI Explanation - verified by User4210 on: May 26, 2026
Question 44:
Question 44 asks how to automatically change the Account status from Prospect to Customer when an Opportunity is won, choosing two valid options.
Two valid approaches:

B. Use Apex trigger to update account status field: A trigger on Opportunity (after update) checks if StageName is 'Closed Won' and then updates the related Account.Status to 'Customer'. Pros: robust, bulk-safe, works for all updates. Cons: requires code and testing.

C. Create Opportunity workflow rule to update account status field: A workflow rule on Opportunity fires when the record becomes Closed Won and uses an Update Records action to update the related Account.Status to 'Customer'. Pros: no code, admin-friendly. Cons: limited to workflow capabilities (older automation; Flow/Process Builder offer more flexibility).

Why the others aren’t appropriate:

A. Visualforce: Not an automatic automation solution; Visualforce is UI code and won’t run by itself on data changes.

D. Create account workflow rule: An Account workflow won’t fire from an Opportunity change unless you have cross-object automation elsewhere; standard Account workflow rules aren’t triggered by Opportunity updates.

Anonymous

AI Explanation - verified by User4210 on: May 26, 2026
Question 40:
Question 40 asks about replacing a manual contract lookup with a Salesforce data model.
Key concepts:

Entitlements and Service Contracts model post-sales service terms.

Service Contract Lines (line items) let you specify coverage for individual products or services under one contract.

Option analysis:

A: Setup an entitlement with Service Contract — creates the contract, but no line items. Good for a single-term contract but doesn’t scale to multiple products/services.

B/C: Not addressing data model needs; they don’t automate the contract lookup.

D: Setup Entitlements with Service Contract and Line Items — most robust. One contract can have multiple line items describing coverage for each product/service, enabling easier, consistent access to a user’s contract without manual lookups.

Recommended: D. It covers multi-product coverage and aligns with best practices for service coverage. If you only need a single contract with no line items, A could be enough, but D is the safer, more scalable choice.
Anonymous

AI Explanation - verified by User4210 on: May 26, 2026
Question 37:
Here’s what you need to merge accounts with related opportunities and cases:

To merge, you must delete the duplicate (non-master) accounts and you may also need to update the master account during the merge. That requires both:

- Delete on Account - Edit on Account

You also need to reassign related records to the master account, so you must be able to update those child records:

- Edit on Opportunity - Edit on Cases
Putting it together, the correct permission set is: Edit and Delete on Account, Edit on Opportunity and Cases.
Answer: D
Why: You need to delete duplicates and edit the master account during the merge, plus edit permissions on related opportunities and cases to reparent them to the master.
Anonymous

AI Explanation - verified by User8593 on: May 26, 2026
Question 26:
Question 26: The correct answer is No.

Why: The goal is to ensure administrative credentials are encrypted during deployment. Azure Information Protection (AIP) is not a secrets-management solution. AIP is used to classify and protect documents and emails, not to securely store or deliver deployment secrets at runtime.

What to use instead: Use Azure Key Vault to securely store credentials and access them during automation. Combine with a service principal or Managed Identity for your deployment tooling to retrieve secrets at runtime without exposing them in code or logs. This provides proper secrets management and encryption in transit at the point of use.

Abuja, Nigeria

AI Explanation - verified by User8593 on: May 26, 2026
Question 22:

Answer: No

Why:

- Microsoft Managed Desktop is a managed endpoint service for Windows devices, not a tool for provisioning or lifecycle-managing virtual machines. - It does not support Linux VMs, and it isn’t designed to create and delete large numbers of VMs on a weekly basis. - To reduce administrative effort for weekly VM deployment/removal (with a mix of Windows and Linux), use options like: - Azure DevTest Labs (templates, schedules, cost controls for rapid VM provisioning/deprovisioning) - or Azure Automation with runbooks, or ARM templates with a deployment pipeline for lifecycle management.
Abuja, Nigeria

fadoua on: May 26, 2026
explanations
Inlet Beach, United States

AI Explanation - verified by User8593 on: May 26, 2026
Question 19:

Answer: B (No)

Why:

- The target is 99.99% availability with as few VMs and AZs as possible. - Azure SLA basics: - 1 VM: ~99.9% uptime. - 2+ VMs in an Availability Set (single data center): ~99.95%. - 2+ VMs across two or more Availability Zones: ~99.99%. - The proposed design uses two VMs in one Availability Zone. This yields ~99.95% SLA (not 99.99%), so it does not meet the goal. - To achieve 99.99% with minimal resources, you should deploy at least two VMs across two Availability Zones (zone-redundant) and place a load balancer in front of them to handle traffic failover.
Abuja, Nigeria

AI Explanation - verified by User8593 on: May 26, 2026
Question 16:
Question 16 asks whether recommending the Azure API Management service (APIM) is a valid strategy to automatically create a large number of identical Azure resources for multiple business units.

Correct answer: No (B).

Why:

APIM is for API management, not for provisioning or automating the creation of resources.

To automatically deploy identical resources for each business unit, use infrastructure-as-code (IaC) tools such as:

- ARM templates or Bicep - Terraform

Pair IaC with automation (CI/CD pipelines in Azure DevOps or GitHub Actions) and optionally use Azure Blueprints to deploy repeatable, compliant resource sets across units.

Key takeaways:

APIM = API management, not provisioning.

For automatic resource creation, prefer IaC and automation tooling.

Abuja, Nigeria

AI Explanation - verified by User8593 on: May 26, 2026
Question 15:
Question 15 asks if using Azure Machine Learning Studio satisfies the goal of deploying an AI solution with the ability to build, test, and deploy predictive analytics.

Correct answer: Yes (A).

Why:

The objective is to build, test, and deploy predictive analytics in Azure.

Azure Machine Learning Studio provides a visual, end-to-end environment for:

- Building experiments and pipelines (data prep, feature engineering, model training). - Running and evaluating models to assess predictive performance. - Deploying models as web services for real-time or batch predictions.

This covers the full lifecycle required for a predictive analytics solution.

Notes:

In modern practice, you might use the newer Azure Machine Learning workspace (not just Studio), but the Studio option in the question clearly maps to the required capabilities (build/test/deploy) for predictive analytics.

Abuja, Nigeria

AI Explanation - verified by User1193 on: May 26, 2026
Question 22:
The correct feature is VNF Service Chaining (option D).
Why:

The question describes branch routers examining traffic and steering it to the appropriate network function, i.e., chaining virtual network functions (VNFs) to form a service path. That’s exactly what “VNF Service Chaining” provides in Cisco SD-WAN.

Regionalized service chaining means applying these chains in specific regions or sites to optimize cost and user experience by selecting the nearest or most cost-effective VNF path.

Incorrect options:

Cloud onRamp for Colocation: On-ramps to VNFs hosted in a colocated data center; focuses on cloud/provider connectivity rather than chaining VNFs across the fabric.

Cloud onRamp for IaaS: On-ramp to IaaS clouds; not about steering traffic through a network function chain.

Cloud Services Platform: Platform for integrating cloud services; not specifically the traffic-steering VNF chaining feature.

Note: The exam page shows A, but the concept that matches the described regionalized service chaining is D — VNF Service Chaining.
Hamburg, Germany

AI Explanation - verified by User2551 on: May 26, 2026
Question 50:
Question 50 asks how to use a private/custom model with Amazon Bedrock. Here’s how to think about the options:

A. Purchase Provisioned Throughput for the custom model.

- Not a typical requirement for Bedrock integration. Bedrock is a managed inference service; you don’t usually buy throughput separately for a private model.

B. Deploy the custom model in an Amazon SageMaker endpoint for real-time inference.

- Could be part of a private model workflow, but Bedrock itself handles how your private model is invoked. This isn’t necessarily a mandatory step to use the model through Bedrock.

C. Register the model with the Amazon SageMaker Model Registry.

- Helpful for model versioning and governance, but not strictly required to "use" the model via Bedrock.

D. Grant access to the custom model in Amazon Bedrock.

- This is the most relevant mandatory step: Bedrock must be authorized to invoke your private model.
Bottom line:

The most correct single action is D (grant access in Bedrock). The provided answer key showing A is likely incorrect. C could be useful for governance, and B could be part of some setups, but they aren’t strictly required to run the custom model through Bedrock.

Bengaluru, India

AI Explanation - verified by User2551 on: May 26, 2026
Question 20:
Question 20: The correct answer is C) Develop an anomaly detection system.

Why: An anomaly detection system identifies unusual or suspicious behavior, such as IP addresses that deviate from normal traffic patterns, and can trigger alerts to protect the application.

Why the others are not suitable:

- A. Build a speech recognition system: unrelated to IP threat detection. - B. NLP named entity recognition: extracts entities from text, not for network security. - D. Fraud forecasting system: predicts future fraud events, not real-time detection of suspicious IPs.
Tip: In practice, you’d model normal IP traffic using features like frequency, geo/location, ASN, and access patterns, then flag deviations with methods such as isolation forests or autoencoders.
Bengaluru, India

AI Explanation - verified by pebefah636 on: May 26, 2026
Question 169:
The correct answer to Question 169 is Elasticity (B).

What elasticity means: The cloud automatically scales resources up or down based on demand to handle short-term fluctuations. This lets you quickly respond to spikes (e.g., a sale or event) and then scale back down when demand falls.

Why not the others:

- Efficiency: about using resources effectively, not specifically auto-scaling. - Multitenancy: sharing resources among many tenants; not about scaling with demand. - Scalability: the capacity to handle growth in load, but not necessarily automatically or for short-term fluctuations.
In short, elasticity is the dynamic, automatic adjustment of resources to match varying workload levels over short periods, while scalability is more about long-term capacity.
Anonymous

AI Explanation - verified by User6888 on: May 26, 2026
Question 26:
Here’s a targeted explanation of Question 26.
What the question is asking

You need to design a custom RBAC role (CR1) that

- can be assigned only to resource groups in Subscription1 - prevents management of access permissions (i.e., no role to grant other permissions) - allows users to view, create, modify, and delete resources inside those resource groups
Key RBAC concept in play

The role’s assignable scope determines where the role can be assigned. To limit the role to “the resource groups in Subscription1,” you want the scope to be at the resource-group level (not at the subscription or management group level). This lets you grant the role to specific resource groups or to all resource groups in the subscription, depending on how you apply the role.

The permission elements (the Actions) define what you can do inside those scopes. Since you need to manage resources inside the groups (view/create/modify/delete), you need actions that cover resource operations inside the resource groups, while not including capabilities to manage role assignments.

What to select (conceptually)

Assignable scopes: Resource group scope(s) within Subscription1

- Example format: /subscriptions/{subscriptionId}/resourceGroups/* (applies to all RGs in the subscription) - If the UI requires a single scope, you’d use a resource group scope for each RG you want CR1 to apply to; the key is that the scope is at the resource-group level, not the entire subscription.

Permission elements (Actions): allow read/write/delete for resources inside the groups

- Examples: - Microsoft.Resources/subscriptions/resourceGroups/resources/read
Anonymous

AI Explanation - verified by User5463 on: May 26, 2026
Question 95:

Correct answer: Accounts.

Why: A queue is a way to assign ownership of records to a group. In Salesforce, queues are supported on standard objects like Leads and Cases and on Custom Objects. Accounts do not support queues, so they cannot be added to a queue.

Quick check tips:

- To see if an object supports queues, go to Setup > Object Manager > [Object] and look for a Queues option. - If an object has Queues, you can create a queue and assign records to it; otherwise, you cannot.
Mumbai, India

AI Explanation - verified by cymamac on: May 26, 2026
Question 115:
Question 115: How is Primary Key used?

The correct answer: C

Why:

- A is not accurate: Primary Key is not specifically about defining a Send Relationship. - B is incomplete: while a primary key can be the subscriber identifier, the primary purpose is not “identifying a subscriber” in general. - C is correct: a Primary Key identifies a row (or a combination of fields) in a data extension as unique. It enforces row-level uniqueness. - D is incorrect: a primary key enforces uniqueness, but it does not inherently prevent overwriting of data.
Note: The provided answer key on the page lists D, but that is not the best description of how Primary Key works in Marketing Cloud data extensions.
Los Angeles, United States

unknown gun man on: May 26, 2026
Yes, it's best for practicing and learning.
Pune, India

AI Explanation - verified by User8731 on: May 26, 2026
Question 16:

Correct answer: D

Why: Evaluation metrics measure how well a model generalizes to unseen data. You review them after you have trained the model and evaluated it on a validation set (or test set) to judge performance and compare models or tune hyperparameters.

Why the others are not correct:

- A: Before you train a model — you don’t have metrics to review yet. - B: After you clean the data — cleaning affects data quality, not the act of evaluating model performance. - C: Before you choose the type of model — you typically compare models by evaluating them on data, not by reviewing metrics before training.

Quick nuance: In practice, you may compute and review metrics multiple times (e.g., during cross-validation), but the point of reviewing metrics is always after evaluation on data the model hasn’t been trained on.

Willowdale, Canada

AI Tutor 👋 I’m here to help!

AI Tutor — 350-901

Hello,
I’m your AI study assistant for the 350-901 exam. Ask me anything about the exam or the questions on this page, and I’ll break it down so it actually makes sense.
You want it in a different language? Just say it.

350-901 5

Community Feedback

💬 Did you find this helpful?

Your name:

Thank you for sharing! Your feedback helps the community.

Contact Us

Exam Vendors

Hay AI, Learn About Us

Customer Reviews

DMCA

Terms of Service

Privacy Policy

© 2026 Free-Braindumps.com - A Community of Learners.
Free-Braindumps.com is owned by Xùnliàn Limited, a company registered at 15 Wang Chiu Rd, Kowloon Bay, Hong Kong.

The resources, practice tests, and study guides on Free-Braindumps.com are strictly for educational and research purposes. We are an independent community platform and are not affiliated with, endorsed by, or sponsored by any official certification vendors, including Microsoft, Cisco, Amazon, CompTIA, or others. All certification names, trademarks, and logos are the property of their respective owners. Our content is user-contributed or independently developed to assist with exam preparation and skill assessment; we do not host or provide access to proprietary, confidential, or live exam questions.

Cisco 350-901 Exam Questions Developing Applications using Cisco Core Platforms and APIs (DEVCOR) (Page 5 )

QUESTION: 16

QUESTION: 17

Explanation:

QUESTION: 18

Reference:

QUESTION: 19

QUESTION: 20

350-901 Exam Discussions & Posts (Share your experience with others)

Cisco 350-901 Exam Questions
Developing Applications using Cisco Core Platforms and APIs (DEVCOR) (Page 5 )