GARP ICBRR Exam Prep
International Certificate in Banking Risk and Regulation (ICBRR) (Page 5 )

• Use one Azure AD directory for the organization.
• Separate divisions with subscriptions and/or resource groups.
• Assign each division’s administrator appropriate Azure role-based access control (RBAC) permissions, scoped to their subscription or resource group.

• A — Add a Path for Stage: A Path lets users update the Opportunity’s Stage directly from the record page, without opening the full edit form. It can also display guidance and key fields.
• B — Use a Kanban list view: Users can move Opportunities between stages by dragging records between columns, which is ideal when Stage is the only field they normally change.

• Honeypots: Decoy systems or services designed to attract attackers. Legitimate users should not access them, so activity on a honeypot is a strong indicator of unauthorized activity. Security teams can also observe attacker behavior.

• DNS sinkhole: Redirects requests for known malicious domains to a controlled internal server or blocks them. If an infected workstation attempts to contact command-and-control infrastructure, the sinkhole records the request and helps identify the compromised device.

• Tokenization: Protects sensitive data by replacing it with tokens.
• CI/CD: Automates software development and deployment.
• Threat modeling: Identifies potential risks before or during system design.
• Data obfuscation: Makes data harder to understand, but does not detect intruders.

• The Balanced Scorecard translates an organization’s vision and strategy into a set of concrete objectives and metrics across four perspectives (financial, customer, internal processes, learning & growth). This framework explicitly links IT goals to business goals and provides a way to measure and monitor alignment over time.

• Enterprise architecture (EA): Great for governance, standards, and roadmaps to ensure IT assets support business needs, but it’s more about structure and alignment at design time rather than directly measuring ongoing alignment of goals.
• KPIs: Useful metrics, but they don’t by themselves provide a full framework to ensure IT goals are aligned with organizational strategy.
• Enterprise dashboard: A visualization tool for metrics; it helps report status but doesn’t inherently establish or assess alignment.

• A. Extranet
• B. VLAN
• C. Intranet
• D. VPN

• Answer: D. Select an IdP

• Why: To reduce the number of credentials across multiple SaaS apps, implement a single sign-on approach via an Identity Provider. The first step is choosing the IdP; once selected, you can enable SAML (A) or OAuth as supported by apps. Password vaulting (C) is not a true SSO solution, and B is not the foundational step.

• If the account name includes the parsed search string, show the element; otherwise hide it.
• In CSS, to show an element you can set display to a visible value (commonly 'block' for block-level elements), and to hide you set display to 'none'.

• The ternary should map true to a visible display value and false to 'none'.
• Therefore the placeholder should be replaced with 'block' : 'none' (so the code becomes: accountName.includes(parsedSearchString) ? 'block' : 'none').

• There are several typos in the snippet (e.g., quesrySelectorAll, toLOwerCase, Style.display). In a working version you’d also fix those to document.querySelectorAll, account.innerHTML.toLowerCase(), and account.style.display.

• Correct answer: Power BI dataset

• The scenario says the data model already contains several measures and you want to create a report from the existing data with minimal development. Reusing a published Power BI dataset lets you inherit the existing measures and data model, so you don’t have to recreate them.
• How to do it: In Power BI Desktop, use the connector Power BI datasets (Get Data > Power BI datasets) and connect to the existing dataset, then build visuals using the pre-defined measures.
• Why the other options are less suitable:

• Why: After listing all resources, you need to assign who is responsible for each task. A RACI chart (Responsible, Accountable, Consulted, Informed) documents these roles for every activity.

• What RACI means:

• Why the others aren’t suitable here:

• Quick example: Task "Create user authentication module" — Responsible: Developer, Accountable: Tech Lead, Consulted: Security Architect, Informed: PM.

• A — IDS attack signatures: An IDS generally detects suspicious network activity but does not necessarily scan uploaded PDF files or prevent them from being accepted.
• C — Load balancer: Distributes traffic among servers; it does not inspect files for malware.
• D — WAF command-injection blocking: A WAF can help block web attacks such as command injection, SQL injection, and cross-site scripting, but this option does not specifically address malicious PDF uploads.

• Vertical scaling (scale up) means increasing the size of an existing VM—for example, adding CPU or RAM.
• Vertical scaling down means reducing the VM’s size when less capacity is needed.
• Horizontal scaling (scale out) means adding more VM instances rather than making one VM larger.
• Elasticity is the broader ability to automatically adjust resources based on demand; it does not specifically mean adding CPU or memory to one VM.
• Agility refers to the ability to provision or change resources quickly.

• Correct answer: B) Context window.

• Why: The context window is the number of tokens the foundation model can process in a single inference. It sets the maximum length of the combined input (prompt) and the model’s output. A larger context window allows more content to be included in one prompt.

• How to use: Check the FM’s context window value and ensure your prompt length plus the desired completion stays within that limit. If needed, summarize or split input across multiple prompts.

• Why the other options aren’t correct:

• Quick tip: Verify the specific FM’s context window for Bedrock, since different models have different token limits.

• I can’t view the images, so I can’t tell which table is labeled as A in your screenshot. But here’s how to decide:

• How to identify the fact table:

• How to identify the surrogate key in the Patient table:

• Relationship:

• Correct answer: only new data added to the eventhouse (option E).
• Why: When you enable OneLake availability for the eventhouse, OneLake starts syncing data going forward. It does not automatically backfill or copy over data that already exists in the eventhouse before the enablement.
• What that means in practice: Any data added to the eventhouse after you enable OneLake will appear in OneLake (and be visible from OneLake file explorer). Pre-existing data in the eventhouse stays there unless you perform a separate backfill/copy operation outside the standard OneLake enablement.
• How to verify: add a small new dataset to the eventhouse after enabling OneLake availability, then check OneLake to confirm the new dataset appears.

• AWS IoT Greengrass supports local ML inference using models trained in Amazon SageMaker.
• The model and inference code run on the factory’s local server, so processing continues even when internet connectivity is unavailable.
• A Greengrass component can:

• This provides low latency and avoids sending images to AWS for every inference.

• A: Uses AWS-hosted services and therefore depends on internet connectivity. It would fail during an outage.
• C: Snowball Edge is unnecessary for this use case and adds deployment complexity. A local Greengrass server is the intended edge-inference solution.
• D: Amazon Monitron is designed for monitoring industrial equipment health, not analyzing camera images with a custom SageMaker model.

• Correct answer: C: Amazon Comprehend.

• Why: The library wants to classify books based on their text content. This is an NLP (natural language processing) task. Amazon Comprehend is AWS’s managed NLP service that can perform content classification (including creating custom classifiers) and extract meaningful information from text. The other services are not focused on text classification:

• Quick note: In practice, you’d store the books in S3, extract text if needed (e.g., from PDFs with Textract), and then use Comprehend to classify the documents into topics or genres.

• Correct answer: C — The model is deployed at UK border checkpoints.

• The EU AI Act applies to AI systems used within the EU or by entities established in the EU. So:

• The answer: Question 51: C — Request trigger.

• To invoke a Logic App from an external service (here, the Azure Application Insights availability test) you need a trigger that starts on an incoming HTTP request. In Logic Apps, this is the trigger that exposes an HTTP endpoint — typically shown as "When a HTTP request is received" and referred to as the Request trigger.
• The other options aren’t appropriate:

• Create a Logic App with the Request trigger to generate a URL.
• Configure Application Insights to post the degraded-availability payload to that URL.
• Add actions (e.g., send email, follow-up) in the Logic App to notify the customer services team.

• Why: The data is unlabeled, so you can’t train a model with known input-output pairs. Unsupervised learning discovers structure in the data by grouping similar customers into clusters or tiers without labels.
• What it does: Clustering and segmentation (e.g., k-means, hierarchical clustering, DBSCAN, Gaussian mixtures) to form meaningful groups.
• Why the other options aren’t as suitable:

• Two VMs
• Distributed across two separate availability zones

• Self-signed certificate: Causes a certificate warning, not incorrect IP resolution.
• Nameserver record: A DNS record problem would typically affect multiple users, not just one.
• IP helper: Relays DHCP or other broadcast traffic between networks; it does not control website name resolution.

• R-squared (R²): mede quanto da variação dos valores reais é explicada pelo modelo. É mais usado em regressão, não em classificação.

• Accuracy (acurácia): mede a proporção de previsões corretas em relação ao total. É adequada para a classificação de imagens da questão.

• RMSE (Root Mean Squared Error): mede o tamanho médio dos erros entre valores previstos e reais, penalizando erros maiores. É usado principalmente em regressão.

• Learning rate (taxa de aprendizado): controla o tamanho dos ajustes feitos pelo modelo durante o treinamento. Não é uma métrica de avaliação; é um hiperparâmetro.

• The target API uses HTTPS, so Mule must validate the server’s TLS certificate.
• The TLS configuration is set to None, meaning no custom truststore is configured.
• Mule therefore uses the JRE’s default truststore to determine whether the target certificate is trusted.
• If the issuing CA certificate is in that truststore, the TLS handshake can succeed.
• If it is absent—or the certificate is otherwise invalid—the request fails with a TLS/handshake error.

• Sustainability is not a CAF perspective (it's not one of the CAF perspectives).
• Performance efficiency and Reliability are pillars of the AWS Well-Architected Framework, not CAF.

• A separate scope using a /32 subnet: A /32 identifies one host, but it is not a practical DHCP scope for this purpose.
• A static IP within the DHCP range: This can cause an IP address conflict if DHCP assigns that address to another device.
• SLAAC: Used for automatic IPv6 address configuration, not for assigning a predictable IPv4 address through DHCP.

• Startup scripts run at computer startup, before any user logs on, making them ideal for machine-wide configuration on each new VM.
• They execute under the Local System account, which is appropriate for initial setup tasks.
• Linking a GPO to the OU ensures every new VM in that domain automatically receives the configuration—great for scalability.

• A) SetupComplete.cmd runs only once, right after Windows Setup, not on every boot, so it isn’t ideal for provisioning multiple VMs consistently.
• B) Logon scripts run per user, not per machine, so they aren’t reliable for initial VM configuration before a user signs in.
• D) Placing scripts in a VHD doesn’t automatically execute them.

• Cutover: The old system is replaced immediately. It may be fast, but any failure can cause significant disruption.
• Phased: The system is introduced in stages, which reduces implementation risk but may prolong the transition.
• Pilot: The system is first deployed to a limited group or location. This is useful for testing but does not necessarily minimize downtime for the entire organization.

• IPv6 address types you’ll typically see:

• Unique Local Addresses (ULAs)

• If an address begins with fdXX:… or fc00:…, place it under “Unique Local Address” (ULA).
• If it begins with 2000::/3, place it under “Global Unicast.”
• If it begins with fe80::, place it under “Link-local.”
• If it begins with ff00::, place it under “Multicast.”

• Step 1: From Power BI Desktop, select Get Data, choose Folder, enter the folder path, and click OK, then choose Transform data to open Power Query Editor.

• Step 2: In Power Query Editor, expand the Attributes column.

• Step 3: From the Power Query Editor, select the Content column and use Home > Combine Files (aka “Combine binaries”) to consolidate the files.

• Why: A multi-modal embedding model maps both text and images into a shared vector space, enabling cross-modal similarity and retrieval (e.g., find images matching a text query or text matching an image).

• What it does: Enables efficient similarity search using a vector database or nearest-neighbor indexing.

• Why the other options aren’t suitable:

• Quick implementation note: you’d typically pair the embeddings with a vector store (e.g., ANN index) and possibly fine-tune on domain data for better alignment across modalities.

• Answer: No.

• Why:

• If you need such an assessment while controlling costs, you’d generally pursue a paid MCS engagement or another option outside of the basic Premier entitlement.

• Amazon Kendra is a managed enterprise search service that can index content stored in S3 and support natural-language queries to find text inside documents (PDFs, Word files, etc.).

• Rekognition analyzes images/videos, not document text search.
• Polly is a text-to-speech service.
• Lex is a chatbot service.

• Create a Kendra index.
• Add a data source of type S3 and point it to your bucket (with an IAM role that allows access).
• Configure/document conversion if needed, then index the data.
• Use the Kendra API or console to run natural-language queries against your indexed documents.

• Disabling GPS: Affects location services, not email connectivity.
• Configuring 4G usage only: Does not enable international data access and could prevent connectivity if 4G is unavailable.
• Connecting using a hotspot: Could provide data, but it is not the most direct solution and depends on another network or device.

• IaaS: You deploy and manage virtual machines, operating systems, disks, and networking. Azure manages the physical datacenter and host infrastructure.
• PaaS: Azure manages the operating system and runtime so you can focus on deploying applications, such as with Azure App Service or Azure Functions.
• SaaS: You consume a complete, provider-managed application.

• In Azure, the typical cost approach for pay-as-you-go VMs is an Operational Expenditure (OPEX) model. You’re billed for actual usage each month.
• The term “elastic expenditure model” isn’t a standard Azure cost category. Elasticity describes scaling capacity, not the billing category itself.
• So recommending the elastic expenditure model is not correct for this goal.

• The solution does not meet the goal. Answer: No (B).

• Pay-as-you-go aligns with OPEX because costs vary with usage.
• Elasticity is about scaling resources, which affects performance and capacity, not the fundamental cost model.

• Answer: D) Out-of-band management

• Why this is correct:

• Why the other options aren’t suitable here:

• 6-phone boxes: 6x
• 8-phone boxes: 8x

• D. Provide public endpoints to store and retrieve data

• E. Manage the infrastructure layer and the operating system

• Classifying data
• Configuring access permissions, such as IAM policies
• Selecting or configuring available encryption options, depending on the requirement

• A recipient who already has a Microsoft account or an organizational account can authenticate by signing in.
• A recipient without a usable Microsoft identity may receive a one-time passcode by email.
• The passcode is sent to the same email address that was invited and is used to verify ownership of that address.

• Assign a Teams license to a user — Licenses are assigned through the Microsoft 365 admin center, Microsoft Entra ID, or PowerShell.
• Deploy the Teams client — The Teams admin center manages Teams policies and apps, but it is not the primary tool for deploying the desktop client.
• Prevent users from creating teams — Team creation depends on controlling the creation of Microsoft 365 groups, generally configured through Microsoft Entra ID and PowerShell rather than directly in the Teams admin center.

• wyjasnic istniejace pytania z tej strony,
• wskazac, dlaczego dana odpowiedz jest poprawna lub bledna (z uwzglednieniem celów egzaminu),
• omówic koncepcje zwiazane z Google Cloud AI.

• Entradas grandes, de até aproximadamente 1 GB.
• Processamentos demorados, que podem levar até cerca de 1 hora.
• O cliente envia a solicitação e consulta o resultado depois, sem precisar manter uma conexão aberta durante todo o processamento.

• Real-time inference: indicada para respostas imediatas, com entradas menores e processamento rápido.
• Serverless inference: adequada para cargas variáveis e solicitações menores; não é ideal para processamento longo e arquivos grandes.
• Batch transform: processa grandes volumes em lote, mas não atende à necessidade de respostas próximas do tempo real.

• Date: 2018-12-31
• Time: 08:59

• B: The original value contains the text at, so it cannot be converted directly to a date.
• A: The last characters contain the time, not the date.
• D: Extracting the first 11 characters can leave an unwanted space and does not properly separate the time component.

• The app on EC2 connects to Aurora using DB credentials stored on the instance. Secrets Manager centralizes these credentials and can rotate them automatically, eliminating manual updates on each EC2 host.
• Automatic rotation reduces operational overhead and enhances security because credentials are rotated on a schedule without code changes.
• The app can retrieve the current secret at runtime (e.g., via the AWS SDK) and use it to authenticate to Aurora, avoiding reliance on locally stored files.

• B (Parameter Store) can store secrets but lacks seamless built-in rotation for database credentials like Secrets Manager; rotation often requires more custom work.
• C (S3) would require complex tooling to rotate and securely fetch credentials and still risks exposure if access policies aren’t perfect.
• D (Encrypted EBS volumes) would store credentials locally on each instance and does not automate rotation or centralized management.

• The correct option is B: AWS Storage Gateway.

• Why: The scenario requires backing up on-prem data to AWS without changing existing backup workflows. AWS Storage Gateway provides a hybrid storage solution; specifically, the Tape Gateway emulates a physical tape library, allowing your existing backup software to write to virtual tapes that are stored in AWS (in S3 or S3 Glacier). This lets you back up 10 TB to the cloud while preserving current backup workflows and on-prem infrastructure.

• How it would work: Deploy a gateway appliance or VM on-site, connect your tape-based backup software to the virtual tape library, and start writing to virtual tapes. The data is then archived in AWS, freeing up local storage.

• Why the others aren’t suitable:

• Quick tip: If you weren’t using tape, File Gateway could be another option for file-based backups; Tape Gateway is the right pick for tape-based workflows.

python 
import azure.cognitiveservices.speech as speechsdk 


speech_config = speechsdk.SpeechConfig( 
    subscription="YOUR_KEY", 
    region="YOUR_REGION" 
)

Important Azure Speech SDK classes

• SpeechConfig

• AudioConfig

python 
  audio_config = speechsdk.audio.AudioConfig( 
      use_default_microphone=True 
  )

• SpeechRecognizer

python 
  recognizer = speechsdk.SpeechRecognizer( 
      speech_config=speech_config, 
      audio_config=audio_config 
  )

• SpeechSynthesizer

• AudioOutputConfig

 Bogotá, Colombia