Courseiva

CCNA Design for New Solutions Questions

21 of 321 questions · Page 5/5 · Design for New Solutions · Answers revealed

301
MCQeasy

A solutions architect is designing a new serverless application using AWS Lambda for business logic, Amazon API Gateway for RESTful APIs, and Amazon DynamoDB for data storage. The application will experience unpredictable traffic spikes. What is the MOST cost-effective way to handle concurrency and scaling?

A.Use Lambda provisioned concurrency to pre-warm instances.
B.Use Lambda reserved concurrency to set a limit on concurrent executions.
C.Configure DynamoDB auto scaling to handle traffic spikes.
D.Set a usage plan in API Gateway with a throttling limit.
AnswerB

Reserved concurrency controls the maximum number of concurrent Lambda invocations, preventing excessive scaling and cost.

Why this answer

Lambda reserved concurrency sets a hard limit on the number of concurrent executions for a function, preventing runaway scaling and controlling costs during unpredictable traffic spikes. It ensures that the function does not consume more concurrency than allocated, which avoids excessive DynamoDB read/write capacity usage and keeps costs predictable without needing to pre-warm instances.

Exam trap

The trap here is that candidates confuse provisioned concurrency (which reduces latency but adds cost) with reserved concurrency (which controls scaling and cost), or they mistakenly think DynamoDB auto scaling or API Gateway throttling directly manages Lambda concurrency.

How to eliminate wrong answers

Option A is wrong because provisioned concurrency pre-warms a fixed number of instances to reduce cold starts, but it incurs additional costs even when idle and does not control scaling or concurrency limits during spikes—it is not cost-effective for unpredictable traffic. Option C is wrong because DynamoDB auto scaling adjusts read/write capacity based on actual traffic, but it does not directly handle Lambda concurrency or scaling; it only manages the database side and can still lead to high costs if Lambda invocations spike. Option D is wrong because a usage plan in API Gateway throttles requests at the API level, but it does not control Lambda concurrency or scaling; it may reject valid requests rather than managing cost-efficient concurrency.

302
MCQhard

A company is migrating a monolithic application to microservices on Amazon ECS with Fargate. The application has variable traffic patterns, with high traffic during business hours and low traffic at night. They want to optimize costs while maintaining performance. Which scaling strategy should they implement?

A.Use target tracking scaling with a schedule to increase minimum capacity during business hours.
B.Use step scaling policies based on memory utilization.
C.Use scheduled scaling to increase capacity during business hours.
D.Use simple scaling policies based on CPU utilization.
AnswerA

Target tracking alone scales on demand but cannot anticipate the predictable daytime peak. Adding a schedule that raises minimum capacity during business hours pre-warms tasks, maintaining performance while target tracking scales down at night to optimise cost.

Why this answer

Combining target tracking scaling with a scheduled action allows the application to dynamically adjust capacity based on actual demand while ensuring a higher baseline during peak business hours. This hybrid approach optimizes costs by scaling down at night and maintains performance by preventing cold starts or lag during traffic spikes, which is ideal for variable patterns on ECS Fargate.

Exam trap

The trap here is that candidates often choose scheduled scaling alone (Option C) thinking it directly handles variable traffic, but they miss that it cannot react to unexpected spikes or lulls within the scheduled window, whereas target tracking with a schedule provides both proactive and reactive scaling.

How to eliminate wrong answers

Option B is wrong because step scaling policies based on memory utilization are less responsive to traffic-driven CPU spikes and can cause thrashing if memory is not the bottleneck; they also lack the predictive baseline needed for variable patterns. Option C is wrong because scheduled scaling alone cannot adapt to real-time fluctuations within business hours, leading to either over-provisioning or under-provisioning if traffic deviates from the schedule. Option D is wrong because simple scaling policies are deprecated in AWS and lack the cooldown and metric stabilization features of target tracking, making them prone to oscillation and inefficient for variable traffic.

303
MCQmedium

A company is designing a new data lake on AWS using Amazon S3. The data will be ingested from various sources, including IoT devices, application logs, and streaming data. The data must be processed in near real-time as it arrives. Which combination of services should be used for ingestion and processing?

A.Amazon S3 Transfer Acceleration and AWS Lambda
B.Amazon Kinesis Data Firehose and Amazon Kinesis Data Analytics
C.Amazon Athena and Amazon S3
D.AWS Glue and Amazon Redshift
AnswerB

Kinesis Data Firehose ingests streaming and log data continuously, while Kinesis Data Analytics runs SQL over the stream for near real-time processing. Together they satisfy the low-latency processing constraint across IoT, logs and streaming sources before landing data in S3.

Why this answer

Amazon Kinesis Data Firehose is the correct ingestion service because it can reliably capture and load streaming data into Amazon S3 in near real-time without custom code. Amazon Kinesis Data Analytics then processes the data using SQL or Apache Flink as it arrives, enabling near real-time transformations and analytics before the data lands in the data lake.

Exam trap

The trap here is that candidates often confuse Amazon S3 Transfer Acceleration (a speed optimization for large file uploads) with a streaming ingestion service, or assume that Athena can process data as it arrives, when in fact Athena only queries data at rest in S3.

How to eliminate wrong answers

Option A is wrong because Amazon S3 Transfer Acceleration is a feature that speeds up uploads over long distances using edge locations, but it does not provide streaming ingestion or near real-time processing capabilities; AWS Lambda alone cannot handle continuous high-throughput streaming ingestion without a buffer like Kinesis. Option C is wrong because Amazon Athena is an interactive query service for analyzing data already stored in S3, not a service for ingesting or processing streaming data in near real-time. Option D is wrong because AWS Glue is a serverless data integration service for batch ETL and cataloging, and Amazon Redshift is a data warehouse for analytics on structured data; neither is designed for near real-time streaming ingestion into a data lake.

304
MCQmedium

A company is building a new order-processing system on AWS. The system uses an Amazon SQS FIFO queue to receive order events from multiple producers. The company requires strict ordering of messages within each order ID and exactly-once processing. To achieve exactly-once processing, the producers must include a deduplication ID with each message. However, the producers currently do not generate deduplication IDs. What should a solutions architect recommend to meet the ordering and exactly-once processing requirements with minimal changes to the producers?

A.Use a standard SQS queue and rely on the order ID as the message deduplication ID.
B.Enable content-based deduplication on the FIFO queue and set the message group ID to the order ID.
C.Configure the FIFO queue with a visibility timeout and use the order ID as the message group ID, but do not set deduplication.
D.Switch to Amazon Kinesis Data Streams with partition key set to the order ID and enable enhanced fan-out.
AnswerB

Content-based deduplication automatically generates a deduplication ID from the message body, so producers do not need to change. Setting the message group ID to the order ID ensures strict ordering within each order. This meets both requirements with minimal producer changes.

Why this answer

For FIFO queues, content-based deduplication automatically creates a deduplication ID from the message body, eliminating the need for producers to supply one. The message group ID ensures that messages within the same group (order ID) are processed in strict order. This solution requires no producer changes and satisfies both ordering and exactly-once processing.

Exam trap

The trap here is assuming that FIFO queues automatically deduplicate without any configuration, when in fact deduplication must be enabled either by providing a deduplication ID or by turning on content-based deduplication.

305
MCQeasy

A company wants to deploy a containerized application on AWS. The application requires persistent storage that can be shared across multiple containers running on different EC2 instances. Which AWS service should be used?

A.EC2 Instance Store
B.Amazon EFS
C.Amazon S3
D.Amazon EBS
AnswerB

Amazon EFS provides a shared, elastic NFS file system mountable concurrently from multiple EC2 instances across Availability Zones, meeting the cross-instance shared persistence requirement. EBS volumes attach to a single instance, and instance store is ephemeral.

Why this answer

Amazon EFS provides a fully managed, scalable, and elastic NFS file system that can be mounted concurrently on multiple EC2 instances. This makes it the ideal choice for persistent storage that must be shared across containers running on different EC2 instances, as it supports the NFSv4.1 and NFSv4.0 protocols and automatically scales storage capacity as files are added or removed.

Exam trap

The trap here is that candidates often confuse Amazon EBS with a shared storage solution, but EBS volumes (except for the limited multi-attach feature) can only be attached to a single EC2 instance at a time, making it unsuitable for multi-instance shared access.

How to eliminate wrong answers

Option A is wrong because EC2 Instance Store provides ephemeral block-level storage that is physically attached to the host computer, and data is lost when the instance is stopped or terminated; it cannot be shared across multiple EC2 instances. Option C is wrong because Amazon S3 is an object storage service accessed via HTTP/HTTPS APIs, not a file system that can be mounted directly by multiple EC2 instances for concurrent read/write access with standard file system semantics. Option D is wrong because Amazon EBS provides block-level storage volumes that can be attached to only one EC2 instance at a time (except for multi-attach EBS io1/io2 volumes, which are limited to a small number of Nitro-based instances and are not designed for general-purpose shared file storage across many containers).

306
Drag & Dropmedium

Drag and drop the steps to set up a cross-region VPC peering connection in the correct order.

Drag or tap steps into the slots.

Steps
Order
1Step 1
2Step 2
3Step 3
4Step 4

Why this order

First initiate, then accept, then add routes, then update security groups, and finally network ACLs.

307
MCQmedium

A company is designing a new microservices architecture on AWS. Each service must be independently deployable and scale based on demand. The company wants to minimize operational overhead for container orchestration. Which AWS service should the company use?

A.Amazon Elastic Kubernetes Service (EKS)
B.Amazon EC2 with Auto Scaling
C.Amazon Elastic Container Service (ECS) with AWS Fargate
D.AWS Lambda
AnswerC

Fargate removes the need to provision or manage EC2 instances for the cluster, so each task runs serverlessly while ECS still provides independent deployment and demand-based scaling per service. This directly satisfies the stem's constraint of minimising container orchestration operational overhead.

Why this answer

Amazon ECS with AWS Fargate is the correct choice because it provides a fully managed container orchestration service that eliminates the need to provision, configure, or scale underlying EC2 instances. Fargate runs each container in its own isolated compute environment, allowing each microservice to scale independently based on demand while minimizing operational overhead for cluster management.

Exam trap

The trap here is that candidates often choose Amazon EKS assuming Kubernetes is the only modern orchestration tool, but the question specifically asks for minimal operational overhead, and Fargate's serverless model removes the need to manage any cluster infrastructure, which EKS does not fully eliminate even with managed node groups.

How to eliminate wrong answers

Option A is wrong because Amazon EKS requires you to manage the Kubernetes control plane (or pay for it) and typically involves managing worker nodes or using managed node groups, which adds operational overhead compared to Fargate's serverless model. Option B is wrong because Amazon EC2 with Auto Scaling requires you to manage the underlying instances, including patching, scaling policies, and container orchestration software, which contradicts the goal of minimizing operational overhead. Option D is wrong because AWS Lambda is designed for short-running, event-driven functions with a maximum execution timeout of 15 minutes and limited runtime environments, making it unsuitable for long-running microservices that require container-level isolation and persistent connections.

308
Multi-Selectmedium

A company is deploying a containerized web application on Amazon Elastic Kubernetes Service (Amazon EKS). The application requires persistent storage that must be shared across multiple pods running on different nodes. The storage must support ReadWriteMany (RWX) access mode and provide high throughput for media processing workloads. The company wants to minimize operational overhead and use AWS managed services. Which two solutions should a solutions architect recommend? (Choose two.)

Select 2 answers
A.Use Amazon Elastic File System (Amazon EFS) with the EFS CSI driver to provide a shared file system for the pods.
B.Use Amazon FSx for Lustre with the FSx CSI driver to provide a high-performance shared file system.
C.Use Amazon Elastic Block Store (Amazon EBS) with the EBS CSI driver to provide persistent volumes for the pods.
D.Use Amazon S3 with the Mountpoint for Amazon S3 CSI driver to provide a shared file system.
E.Use AWS Storage Gateway in file gateway mode to provide an NFS mount for the pods.
AnswersA, B

Amazon EFS supports the ReadWriteMany access mode, allowing multiple pods across different nodes to mount and write to the same file system concurrently. The EFS CSI driver integrates with Amazon EKS to dynamically provision persistent volumes. EFS is a fully managed, scalable file system that requires minimal operational overhead and can deliver high throughput for media processing, making it an ideal choice for this scenario.

Why this answer

Amazon EFS and Amazon FSx for Lustre both support the ReadWriteMany access mode and are managed AWS services that integrate with Amazon EKS via CSI drivers. EFS provides scalable, elastic storage suitable for general-purpose shared file systems, while FSx for Lustre delivers high throughput and low latency for compute-intensive media processing. Together, they offer the required shared storage with minimal operational overhead.

Exam trap

The trap here is assuming that Amazon EBS or Amazon S3 can satisfy a ReadWriteMany requirement, but EBS is limited to single-node attachment and S3 is not a POSIX file system.

309
MCQmedium

A company is designing a new application that will use Amazon DynamoDB as its primary data store. The application requires consistent low-latency read and write performance at any scale, and the company wants to minimize operational effort. The data model includes a table with a partition key and a sort key. The workload is expected to have highly uneven access patterns, with some items being accessed much more frequently than others. Which design approach should a solutions architect recommend to meet these requirements?

A.Use DynamoDB on-demand capacity mode and design the partition key to distribute traffic evenly across partitions.
B.Use DynamoDB provisioned capacity mode with auto scaling and choose a partition key that includes a timestamp to ensure even distribution.
C.Use DynamoDB Accelerator (DAX) in front of the table to cache frequently accessed items and absorb read traffic.
D.Use a global secondary index (GSI) with a different partition key to spread the load and enable queries on alternate attributes.
AnswerA

On-demand capacity mode automatically scales to handle workload changes, eliminating capacity planning and minimizing operational effort. Designing a partition key that distributes traffic evenly prevents hot partitions, ensuring consistent low-latency performance even with uneven access patterns. This combination meets the requirements for scalability and low operational overhead.

Why this answer

On-demand capacity mode removes the burden of capacity planning and automatically scales with workload, minimizing operational effort. A well-distributed partition key is essential to avoid hot partitions and ensure consistent performance, especially when access patterns are uneven. Together, they provide low-latency reads and writes at any scale without manual intervention.

Exam trap

The trap here is assuming that DAX can solve all performance issues, but DAX only accelerates reads and does not help with write scalability or hot partitions.

310
Multi-Selecthard

A company is designing a new multi-tier web application on AWS. The application uses an Auto Scaling group of EC2 instances for the web tier and an Amazon RDS for PostgreSQL DB instance for the database. To improve security, the company wants to ensure that the web tier instances can connect to the database only through a specific port and that the database is not accessible from the internet. Which steps should the company take? (Choose THREE.)

Select 3 answers
A.Launch the database instance in a public subnet.
B.Configure the database security group to allow inbound traffic on port 5432 from the web tier security group.
C.Store database credentials in AWS Systems Manager Parameter Store.
D.Launch the web tier instances in a private subnet.
E.Set the 'Publicly accessible' option of the RDS instance to 'No'.
AnswersB, D, E

Referencing the web tier's security group as the source restricts inbound database traffic to those instances only, on port 5432. This enforces the requirement that connections arrive solely through the PostgreSQL port from the web tier, rather than any broader CIDR range.

Why this answer

Option B is correct because PostgreSQL listens on TCP port 5432, and referencing the web tier's security group as the source in the RDS security group's inbound rule allows only those EC2 instances to reach the database on that port. Option D is correct because placing the web tier instances in a private subnet removes them from direct internet exposure while still allowing outbound access to the RDS endpoint. Option E is correct because setting the RDS instance's 'Publicly accessible' option to 'No' ensures the database receives only a private IP address and cannot be reached from the internet.

Option A is wrong because a public subnet would expose the database to internet routing, contradicting the requirement. Option C, while a good credential-management practice, does not control network access or port restrictions and is therefore not part of the required steps.

Exam trap

The trap here is that candidates may confuse security best practices (like using Parameter Store for credentials) with network-level access controls, leading them to select Option C instead of recognizing that only security group rules and subnet placement directly control connectivity and internet exposure.

311
MCQeasy

A company wants to automate the deployment of a three-tier web application on AWS. The deployment should include the network, security groups, EC2 instances, and an Application Load Balancer. Which AWS service should they use?

A.AWS Elastic Beanstalk
B.AWS CloudFormation
C.AWS CodeDeploy
D.AWS OpsWorks
AnswerB

AWS CloudFormation provisions the entire stack declaratively from a template, defining VPC subnets, security groups, EC2 instances and the Application Load Balancer as interdependent resources. This satisfies the stem's requirement to automate deployment of all three tiers together, rather than configuring each component manually or scripting only parts of the infrastructure.

Why this answer

AWS CloudFormation is an Infrastructure as Code service that lets you define and provision AWS resources — including VPCs, subnets, security groups, EC2 instances, and Application Load Balancers — declaratively in templates. It is the correct choice for automating the deployment of an entire three-tier architecture with all supporting network and security components.

Exam trap

The trap is confusing deployment orchestration (CodeDeploy) or PaaS abstraction (Elastic Beanstalk) with infrastructure provisioning — only CloudFormation declaratively provisions the full stack including network and load balancer.

How to eliminate wrong answers

Option A is wrong because Elastic Beanstalk is a PaaS that abstracts infrastructure for application deployment; it does not give granular control over VPC, subnets, security groups, and ALB configuration in a single declarative template. Option C is wrong because CodeDeploy is a deployment orchestration service for applications onto EC2, Lambda, or ECS — it does not provision network or load balancer infrastructure. Option D is wrong because AWS OpsWorks (Chef/Puppet-based) is a configuration management service, not a declarative infrastructure provisioning tool, and it is legacy/deprecated for new architectures.

312
MCQmedium

A company is designing a new application that will process sensitive financial data. The data must be encrypted at rest and in transit. The application runs on EC2 instances. Which combination of services meets these requirements?

A.Use Amazon S3 with server-side encryption and enforce HTTPS.
B.Use AWS Key Management Service (KMS) to generate keys and enable encryption on EBS volumes.
C.Use AWS Certificate Manager (ACM) to issue TLS certificates and configure the application to use HTTPS.
D.Enable EBS encryption on the volumes and configure the application to use TLS for all network traffic.
AnswerD

EBS encryption protects data at rest on the attached volumes, while TLS secures data in transit between the application and its clients or services. Together they satisfy both stated requirements for the financial data processed on the EC2 instances.

Why this answer

It addresses both encryption requirements: EBS encryption ensures data at rest is encrypted using AWS KMS-managed keys, and configuring the application to use TLS encrypts data in transit between clients and the EC2 instances. This combination directly meets the stated security needs without relying on external services like S3 or ACM for the EC2-hosted application.

Exam trap

The trap here is that candidates often pick options that address only one requirement (e.g., only encryption at rest or only encryption in transit) or confuse storage services (S3) with compute services (EC2), failing to realize both encryption states must be covered for the EC2-based application.

How to eliminate wrong answers

Option A is wrong because Amazon S3 is a storage service, not the compute platform (EC2) where the application runs; using S3 with server-side encryption and HTTPS does not encrypt data at rest on EC2 EBS volumes or in transit to/from the application. Option B is wrong because enabling encryption on EBS volumes only covers data at rest, but does not address encryption in transit for network traffic. Option C is wrong because ACM-issued TLS certificates and HTTPS only encrypt data in transit, leaving data at rest on EC2 EBS volumes unencrypted.

313
MCQmedium

A company is designing a multi-region active-active application using Amazon DynamoDB. They need to ensure low-latency reads and writes globally. Which DynamoDB feature should they use?

A.Auto Scaling
B.DynamoDB Accelerator (DAX)
C.DynamoDB Streams
D.Global Tables
AnswerD

Global Tables replicate data across chosen AWS Regions with multi-active writes, letting each Region serve local reads and writes at single-digit millisecond latency. This satisfies the active-active, low-latency global requirement, whereas single-Region tables or read replicas cannot accept writes everywhere.

Why this answer

DynamoDB Global Tables provide a fully managed, multi-region, active-active replication solution. They automatically replicate data across chosen AWS Regions, allowing low-latency reads and writes from any region. This directly meets the requirement for a multi-region active-active application with global low-latency access.

Exam trap

SAP-C02 often tests the difference between features that improve performance within a region (like DAX) and those that enable multi-region active-active (Global Tables). Candidates may mistakenly choose DAX for global low-latency, but DAX is region-specific and does not replicate data.

How to eliminate wrong answers

Option A is wrong because Auto Scaling only adjusts read/write capacity based on demand; it does not replicate data across regions. Option B is wrong because DAX is an in-memory cache for DynamoDB that reduces read latency within a single region, but it does not provide multi-region replication or active-active writes. Option C is wrong because DynamoDB Streams captures a time-ordered sequence of item-level changes, but it is not a replication feature; it is used for triggers and change data capture, not for global active-active setups.

314
MCQhard

A company is designing a new application that requires a relational database with high availability and automatic failover across multiple Availability Zones. The database must support read-heavy workloads with minimal latency and must be able to scale read capacity independently. The company wants to minimize administrative overhead. Which solution meets these requirements?

A.Amazon RDS for PostgreSQL with Multi-AZ and a standby replica that can serve read traffic to offload the primary.
B.Amazon Aurora with a Multi-AZ deployment and Aurora Replicas in the same region.
C.Amazon DynamoDB with global tables and on-demand capacity mode.
D.Amazon RDS for MySQL with a Multi-AZ deployment and a read replica in a different region.
AnswerB

Amazon Aurora stores data across multiple Availability Zones and supports Aurora Replicas that can serve read traffic with low latency. It provides automatic failover and allows independent scaling of read capacity by adding replicas. Aurora is fully managed, minimizing administrative overhead. This meets all requirements: high availability, automatic failover, low-latency reads, and independent read scaling.

Why this answer

Amazon Aurora is a MySQL- and PostgreSQL-compatible relational database that provides high availability through storage replication across three Availability Zones and automatic failover. Aurora Replicas can serve read-only traffic, enabling independent scaling of read capacity with low latency. It is fully managed, reducing administrative overhead.

Exam trap

The trap here is assuming that an RDS Multi-AZ standby can serve read traffic, but it cannot; only read replicas can offload reads.

315
MCQeasy

A company is designing a new web application that will be accessed by users globally. The application uses Amazon CloudFront as a CDN and stores static content in Amazon S3. The dynamic content is served from EC2 instances in a single AWS Region. Which of the following will improve performance for users in distant regions with the LEAST operational effort?

A.Deploy EC2 instances in multiple AWS Regions and use Route 53 latency-based routing.
B.Use AWS Global Accelerator to route traffic to the EC2 instances via the AWS global network.
C.Move the dynamic content to Lambda@Edge functions running at CloudFront edge locations.
D.Add an Amazon ElastiCache cluster in front of the EC2 instances to cache dynamic responses.
AnswerB

AWS Global Accelerator provides static anycast IP addresses and routes user traffic onto the AWS global network at the nearest edge location, reducing latency to the single-Region EC2 origin. It satisfies the distant-user performance requirement with minimal operational effort, requiring no application changes or multi-Region deployment.

Why this answer

AWS Global Accelerator uses the AWS global network to route traffic from edge locations to the optimal EC2 endpoint, reducing latency and jitter for distant users without requiring multi-region deployments. This minimizes operational effort because it involves only a single configuration change (creating an accelerator and associating it with the existing EC2 instances) rather than managing infrastructure across multiple regions.

Exam trap

The trap here is that candidates often assume Lambda@Edge can handle all dynamic content at the edge, but they overlook its execution time and resource constraints, making it unsuitable for complex backend logic or database queries.

How to eliminate wrong answers

Option A is wrong because deploying EC2 instances in multiple AWS Regions and using Route 53 latency-based routing requires significant operational overhead for managing, patching, and synchronizing infrastructure across regions, which contradicts the 'least operational effort' requirement. Option C is wrong because Lambda@Edge functions are designed for lightweight compute at edge locations (e.g., header manipulation, authentication) and are not suitable for serving full dynamic content from a backend, as they have execution time limits (5 seconds for viewer events) and cannot maintain persistent connections to databases. Option D is wrong because adding an ElastiCache cluster in front of EC2 instances only caches dynamic responses for repeated requests within a single region, but does not reduce network latency for users in distant regions who still traverse the public internet to reach the origin region.

316
MCQeasy

A company wants to decouple a frontend API from backend processing to improve scalability and fault tolerance. The frontend sends requests that can be processed asynchronously. Which AWS service should be used to decouple the components?

A.Amazon Simple Notification Service (SNS)
B.Amazon Simple Queue Service (SQS)
C.Amazon Kinesis Data Streams
D.AWS Step Functions
AnswerB

Amazon SQS provides a fully managed message queue that buffers requests between the frontend API and backend workers, so components communicate asynchronously without direct coupling. This satisfies the decoupling requirement, absorbing traffic spikes and letting backend processing scale or fail independently of the frontend.

Why this answer

Amazon Simple Queue Service (SQS) is the correct choice because it provides a fully managed message queue that decouples the frontend API from backend processing. The frontend can send requests to an SQS queue, and backend consumers can poll and process messages asynchronously, which improves scalability by buffering traffic spikes and enhances fault tolerance by persisting messages until they are successfully processed.

Exam trap

The trap here is that candidates often confuse SNS (push-based notification) with SQS (pull-based queue) for decoupling, but SNS does not provide the durable, asynchronous message buffer required for decoupling frontend and backend processing.

How to eliminate wrong answers

Option A is wrong because Amazon SNS is a pub/sub messaging service that pushes messages to subscribers (e.g., HTTP endpoints, Lambda, SQS) but does not provide a durable buffer for asynchronous decoupling; it is designed for fan-out notifications, not for queuing where consumers pull messages at their own pace. Option C is wrong because Amazon Kinesis Data Streams is optimized for real-time streaming of large-scale data (e.g., clickstreams, logs) with ordered records and replay capabilities, not for simple request/response decoupling where each message is processed independently by a single consumer. Option D is wrong because AWS Step Functions is a serverless orchestration service for coordinating multiple AWS services into workflows, not a message queue; it does not inherently decouple frontend from backend via asynchronous message buffering.

317
MCQhard

A company is designing a new application that will use Amazon RDS for PostgreSQL. They need to implement read replicas to offload read traffic. However, they are concerned about replication lag affecting read consistency. Which action would minimize replication lag?

A.Increase the allocated storage on the primary instance
B.Use an encrypted connection between the primary and replica
C.Enable Multi-AZ on the primary instance
D.Use instance types with higher network performance
AnswerD

Replication lag on RDS read replicas is primarily bounded by the network throughput between primary and replica and by write volume. Larger instance types with higher network performance let the replica apply the primary's WAL stream faster, reducing lag and improving read consistency.

Why this answer

Using an instance type with higher network performance (option D) reduces replication lag because it provides more bandwidth for the replication process. Option A (increasing allocated storage) does not directly affect replication lag. Option B (using an encrypted connection) does not impact lag; encryption adds overhead but is necessary for security.

Option C (enabling Multi-AZ) is for high availability, not for read replicas.

318
Multi-Selecthard

A company is designing a new containerized application on Amazon EKS. The application must be able to access secrets (e.g., database credentials) securely. The company requires that secrets be automatically rotated and audited. Which THREE actions should the company take to meet these requirements?

Select 3 answers
A.Mount the Secrets Store CSI Driver volume directly to the pod without using ASCP
B.Use IAM roles for service accounts (IRSA) to grant pods access to Secrets Manager
C.Store secrets in AWS Secrets Manager and enable automatic rotation
D.Use the AWS Secrets and Configuration Provider (ASCP) for the Secrets Store CSI Driver to inject secrets into pods
E.Store secrets in Kubernetes Secrets and use a ConfigMap to reference them
AnswersB, C, D

IRSA maps a Kubernetes service account to an IAM role via the EKS OIDC provider, letting pods retrieve Secrets Manager values without long-lived credentials. This enables fine-grained IAM policies and CloudTrail auditing of each secret access.

Why this answer

Option B is correct because IAM roles for service accounts (IRSA) associates a Kubernetes service account with an IAM role via the cluster's OIDC provider, giving pods scoped, credential-free access to AWS Secrets Manager without embedding long-lived keys. Option C is correct because AWS Secrets Manager natively supports automatic rotation through Lambda rotation functions, and it logs API calls to CloudTrail for auditing, satisfying both the rotation and audit requirements. Option D is correct because the AWS Secrets and Configuration Provider (ASCP) for the Secrets Store CSI Driver mounts Secrets Manager secrets as volumes into pods and can sync them to Kubernetes Secrets, enabling rotation-aware secret delivery to the application.

Option A is wrong because the Secrets Store CSI Driver alone cannot retrieve AWS Secrets Manager secrets without the ASCP provider plugin, so mounting the volume directly would fail. Option E is wrong because Kubernetes Secrets are not automatically rotated and are only base64-encoded by default, and ConfigMaps are not designed for sensitive credential storage or auditing.

Exam trap

The trap here is that candidates may think mounting the CSI driver without ASCP (Option A) is sufficient, but ASCP is the critical component that bridges the CSI driver to AWS Secrets Manager, and without it, the driver cannot retrieve secrets from AWS.

319
Multi-Selectmedium

A company is designing a new batch processing system that processes large files from Amazon S3. The processing is CPU-intensive and can take up to 2 hours per file. The company wants to minimize cost and avoid idle compute capacity. Which THREE components should the architect include? (Choose THREE.)

Select 3 answers
A.EC2 Spot Instances to reduce compute costs.
B.AWS Lambda functions to process each file.
C.AWS Batch with a job queue and compute environment.
D.Amazon S3 Event Notifications to trigger the batch job when a new file is uploaded.
E.Auto Scaling group with scheduled scaling policies.
AnswersA, C, D

EC2 Spot Instances use spare AWS capacity at steep discounts, directly addressing the minimise-cost requirement. Because jobs run up to two hours and are interruptible batch work, Spot reclaim risk is acceptable, and capacity releases avoid idle compute.

Why this answer

Option A is correct because EC2 Spot Instances can cut compute costs by up to 90% versus On-Demand, and since the workload is a fault-tolerant batch job that can be retried, Spot interruptions are acceptable, directly supporting the goal of minimizing cost. Option C is correct because AWS Batch is purpose-built for batch computing: it manages a job queue and a compute environment (which can be backed by Spot Instances), dynamically provisioning resources so there is no idle compute capacity. Option D is correct because Amazon S3 Event Notifications can publish an event (e.g., to Amazon EventBridge, SQS, or Lambda) when a new file lands in the bucket, automatically submitting the batch job and eliminating polling or manual triggering.

Option B is not appropriate because Lambda has a maximum execution timeout of 15 minutes, far short of the 2-hour CPU-intensive processing time per file. Option E is not appropriate because an Auto Scaling group with scheduled scaling provisions capacity on a time schedule rather than reacting to file arrivals, which risks idle compute and does not provide the job queue and job management that AWS Batch offers.

Exam trap

The trap here is that candidates often choose AWS Lambda for any event-driven processing without considering its 15-minute timeout limit, overlooking that long-running CPU-intensive tasks require a different compute service like AWS Batch.

320
MCQmedium

A company is designing a data lake solution on Amazon S3. Data is ingested from multiple sources and stored in a raw bucket. The data must be processed and transformed before being moved to a curated bucket. The processing logic is complex and includes conditional transformations. Which service should be used to orchestrate the transformation pipeline?

A.AWS Data Pipeline
B.AWS Step Functions
C.AWS Lambda functions
D.AWS Glue ETL jobs
AnswerB

AWS Step Functions orchestrates multi-step workflows with conditional branching via Choice states, satisfying the complex, conditional transformation logic the stem requires. Unlike AWS Glue, which runs Apache Spark jobs, Step Functions coordinates Lambda, Glue, and EMR tasks in sequence, making it the appropriate orchestration layer for moving processed data between raw and curated buckets.

Why this answer

AWS Step Functions is the correct choice because it is designed to orchestrate complex, multi-step workflows with conditional branching, retries, and error handling. It can coordinate AWS Lambda functions, AWS Glue jobs, and other services to process and transform data from a raw S3 bucket to a curated bucket, making it ideal for a transformation pipeline with complex logic.

Exam trap

The trap here is that candidates often confuse AWS Glue ETL jobs as an orchestration tool because it can transform data, but Glue is a processing engine, not a workflow orchestrator; Step Functions is the correct service for coordinating complex, conditional pipelines.

How to eliminate wrong answers

Option A is wrong because AWS Data Pipeline is a legacy service for moving data between sources and destinations, but it lacks native support for complex conditional transformations and is not as flexible as Step Functions for orchestrating custom processing logic. Option C is wrong because AWS Lambda functions are stateless and have a maximum execution timeout of 15 minutes, making them unsuitable for orchestrating long-running or multi-step transformation pipelines; they are better suited for individual processing tasks within the workflow. Option D is wrong because AWS Glue ETL jobs are designed for batch data transformation using Apache Spark, but they are not an orchestration service; they would be a component orchestrated by Step Functions, not the orchestrator itself.

321
MCQhard

A company is migrating a legacy on-premises application to AWS. The application requires a shared file system that can be mounted by multiple EC2 instances concurrently, with strong consistency and low-latency access. Which AWS storage solution should be used?

A.AWS Storage Gateway File Gateway
B.Amazon EFS
C.Amazon EBS with multi-attach enabled
D.Amazon S3
AnswerB

Amazon EFS provides a fully managed NFS file system that multiple EC2 instances mount concurrently across Availability Zones, delivering the strong consistency and low-latency shared access the legacy application demands. Unlike EBS, which attaches to a single instance, EFS satisfies the concurrent-mount constraint directly.

Why this answer

Amazon EFS provides a fully managed, scalable, and elastic NFS file system that can be mounted concurrently by multiple EC2 instances across multiple Availability Zones. It offers strong read-after-write consistency and low-latency access, making it ideal for shared file workloads requiring POSIX permissions and concurrent access.

Exam trap

The trap here is that candidates often confuse Amazon EBS multi-attach with a shared file system, but EBS multi-attach is a block-level shared volume limited to a single AZ and does not provide file-level locking or POSIX semantics, making it unsuitable for concurrent file system access.

How to eliminate wrong answers

Option A is wrong because AWS Storage Gateway File Gateway provides on-premises caching of S3 objects via NFS/SMB, but it is designed for hybrid cloud access with higher latency and not optimized for low-latency concurrent EC2 mounting. Option C is wrong because Amazon EBS with multi-attach enabled supports only up to 16 Nitro-based EC2 instances in a single Availability Zone, lacks cross-AZ support, and does not provide a shared file system interface (it is a block device). Option D is wrong because Amazon S3 is an object storage service accessed via HTTPS API, not a file system that can be mounted with POSIX semantics, and it does not support concurrent file-level locking or low-latency NFS access.

← PreviousPage 5 of 5 · 321 questions total

Ready to test yourself?

Try a timed practice session using only Design for New Solutions questions.