Quibank app: now on the App Store — free, works offline. Get the app ↗

24 Technology: Compute & Networking Practice Questions & Answers

Every Technology: Compute & Networking practice question from the AWS Cloud Practitioner Practice Test, with the correct answer and a short explanation.

Start practice test →
  1. 1. A company runs an in-memory database that must keep a very large data set in memory for fast processing. Which EC2 instance category best fits this workload?

    • A.Compute optimized, built around high-performance processors
    • B.Storage optimized, built for heavy local disk reads and writes
    • C.General purpose, built for an even mix of compute and memory
    • D.Memory optimized, built to process large data sets held in memory✓ Answer

    AWS describes memory optimized instances as delivering fast performance for workloads that process large data sets in memory, such as in-memory databases. Storage optimized instances target local disk throughput rather than memory capacity.

    Ref QB-CQ3QZWSource: Amazon EC2 Instance Types page, Memory Optimized categoryReport a problem with this question

  2. 2. A studio runs media transcoding and dedicated game servers on Amazon EC2. Which instance category does AWS list for these workloads?

    • A.Storage optimized, because the jobs read large files from local disk
    • B.Compute optimized, because the jobs demand high-performance processors✓ Answer
    • C.Memory optimized, because the jobs keep huge data sets in memory
    • D.General purpose, because the jobs need an even mix of resources

    Compute optimized instances are designed for workloads that benefit from high-performance processors, and AWS lists batch processing, media transcoding and dedicated game servers among their uses. Memory optimized suits large data sets processed in memory, and storage optimized suits heavy low-latency disk I/O.

    Ref QB-V79VKXSource: Amazon EC2 Instance Types page, Compute Optimized categoryReport a problem with this question

  3. 3. A data platform runs a high-throughput database that performs heavy sequential reads and writes against very large data sets on local disks. Which EC2 instance category best fits?

    • A.Memory optimized, since the database keeps its full data set in memory
    • B.Compute optimized, since the database depends on the fastest processors
    • C.Accelerated computing, since the database offloads its reads to co-processors
    • D.Storage optimized, since the database drives high local disk throughput✓ Answer

    Storage optimized instances are designed for high sequential read and write access to very large data sets on local storage. An in-memory database would point to memory optimized instead, so the local disk cue is what decides this one.

    Ref QB-DQE27YSource: Amazon EC2 Instance Types page, Storage Optimized categoryReport a problem with this question

  4. 4. A company already runs its applications on Kubernetes on premises. It wants to move to AWS, keep its Kubernetes tooling and avoid refactoring. Which service best fits?

    • A.AWS Fargate, an orchestrator that takes over from the Kubernetes API
    • B.Amazon ECR, a registry that runs Kubernetes workloads from images
    • C.Amazon EKS, a managed service that runs conformant Kubernetes✓ Answer
    • D.Amazon ECS, an AWS-native orchestrator that accepts Kubernetes files

    Amazon EKS is managed Kubernetes: AWS runs the control plane, and because EKS is certified conformant, existing workloads and community tooling keep working without refactoring. Amazon ECS is AWS's own orchestrator, AWS Fargate is a compute layer, and Amazon ECR only stores images.

    Ref QB-S77V3USource: Amazon EKS User Guide, What is Amazon EKSReport a problem with this question

  5. 5. A development team builds container images and needs a managed place to keep them so that Amazon ECS can pull them at deployment. Which service fits this need?

    • A.Amazon EKS, which stores the images in its control plane
    • B.AWS Batch, which stores the images until a job needs them
    • C.AWS Fargate, which stores the images on its compute layer
    • D.Amazon ECR, which stores and manages the container images✓ Answer

    Amazon ECR is the managed container image registry: it stores images, and services such as Amazon ECS and Amazon EKS pull from it to run them. It does not run containers itself, and the other services run workloads rather than store images.

    Ref QB-AQC2PUSource: Amazon ECR User Guide, What is Amazon Elastic Container RegistryReport a problem with this question

  6. 6. A team already uses Amazon ECS to orchestrate its containers but wants to stop provisioning and scaling the virtual machines underneath them. What should the team adopt?

    • A.Amazon ECR, as a registry that runs the containers from stored images
    • B.AWS Lambda, as a function service that hosts its ECS tasks unchanged
    • C.AWS Fargate, as the serverless compute layer beneath its ECS tasks✓ Answer
    • D.Amazon EKS, as a replacement orchestrator that removes every server

    AWS Fargate is not an orchestrator or an alternative to Amazon ECS; it is the serverless compute layer that runs under it. The team keeps ECS, specifies CPU and memory for each task, and no longer provisions, configures or scales virtual machines.

    Ref QB-T7J79VSource: Amazon ECS Developer Guide, AWS Fargate for Amazon ECSReport a problem with this question

  7. 7. Each time a photo is uploaded, an app must run a few seconds of code to create a thumbnail. Uploads arrive irregularly, and the team does not want to manage servers. Which compute service best fits?

    • A.AWS Lambda, running a short function only when an upload event arrives✓ Answer
    • B.Amazon EC2, running an instance around the clock to watch for uploads
    • C.Amazon Lightsail, running a fixed-price virtual server for all uploads
    • D.AWS Fargate, running a long-lived container that waits for each upload

    AWS Lambda runs code as functions in response to events with no servers to manage, which suits short, irregular tasks like this one. Options that keep a container or server running would mean paying for and tending capacity that sits idle between uploads.

    Ref QB-BQQCUVSource: AWS Lambda Developer Guide, What is AWS LambdaReport a problem with this question

  8. 8. An application needs a custom kernel module and specific operating system settings on its servers, and the team accepts responsibility for patching the guest operating system. Which compute option best fits?

    • A.Amazon EC2, where the customer controls and patches the guest OS✓ Answer
    • B.Amazon ECS, where the customer defines tasks and rebuilds the kernel
    • C.AWS Fargate, where the customer sets CPU and edits the task's host
    • D.AWS Lambda, where the customer uploads code and tunes the kernel

    Amazon EC2 gives the customer control of the guest operating system, and under the shared responsibility model the customer patches it. With AWS Lambda and AWS Fargate, AWS manages the underlying servers, so the customer cannot tune them, and Amazon ECS only orchestrates containers.

    Ref QB-Q7G7RRSource: AWS Shared Responsibility Model; Amazon EC2 User Guide, What is Amazon EC2Report a problem with this question

  9. 9. A small-business owner with little cloud experience wants a simple website on a virtual private server, bundled with storage and data transfer at a predictable price. Which service best fits?

    • A.Amazon Lightsail, offering simple server bundles at a fixed monthly price✓ Answer
    • B.AWS Outposts, offering AWS-managed infrastructure installed on the premises
    • C.Amazon EC2, offering many instance families for fine-grained custom setups
    • D.AWS Batch, offering scheduled compute for large queues of processing jobs

    Amazon Lightsail bundles a virtual private server with storage and data transfer at a predictable price, which suits simple websites and users new to the cloud. Amazon EC2 could host the site too, but it asks the owner to make far more choices.

    Ref QB-8QNCBSSource: Amazon Lightsail documentation, What is Amazon LightsailReport a problem with this question

  10. 10. A research group must run hundreds of thousands of computing jobs that can wait in a queue. It wants AWS to schedule them and provision the right amount of compute. Which service best fits?

    • A.Amazon SQS, which queues the jobs and provisions workers to run them
    • B.AWS Elastic Beanstalk, which deploys the jobs as a scaled web application
    • C.AWS Batch, which queues the jobs and provisions compute to run them✓ Answer
    • D.Amazon Lightsail, which runs the jobs on a bundle of fixed-price servers

    AWS Batch is built for batch computing at scale: it queues jobs, schedules them and dynamically provisions the compute they need. Amazon SQS holds messages in a queue but provisions no compute, so it would still need something else to do the work.

    Ref QB-KEWMURSource: AWS Batch User Guide, What is AWS BatchReport a problem with this question

  11. 11. Traffic to an online store doubles every evening and falls overnight. The company wants the number of instances to follow demand so that it does not pay for idle capacity. What provides this?

    • A.Elastic Load Balancing, giving elasticity by launching new instances
    • B.AWS Auto Scaling, giving elasticity by adding and removing capacity✓ Answer
    • C.Amazon CloudFront, giving elasticity by creating instances at the edge
    • D.AWS Compute Optimizer, giving elasticity by resizing instances nightly

    Auto scaling provides elasticity: scaling policies launch instances for peaks and terminate them when demand falls, within a set minimum and maximum. Elastic Load Balancing only distributes traffic among instances, and AWS Compute Optimizer only makes rightsizing recommendations.

    Ref QB-3XQF9RSource: Amazon EC2 Auto Scaling User Guide, What is Amazon EC2 Auto ScalingReport a problem with this question

  12. 12. A web application runs on several EC2 instances in two Availability Zones. The company needs a single entry point that spreads requests across the instances and stops sending requests to any instance that fails its health check. What provides this?

    • A.An internet gateway, which shares incoming requests among the instances
    • B.A load balancer, which routes requests only to healthy registered targets✓ Answer
    • C.A NAT gateway, which translates inbound requests for healthy private hosts
    • D.An Auto Scaling group, which forwards requests to its newest instances

    A load balancer accepts incoming traffic as a single entry point, routes it to registered targets across Availability Zones and monitors their health so that only healthy targets receive requests. Auto Scaling changes how many instances exist; it does not distribute the traffic.

    Ref QB-JEJAPQSource: Elastic Load Balancing User Guide, How Elastic Load Balancing worksReport a problem with this question

  13. 13. In an Amazon VPC, what makes a subnet a public subnet rather than a private one?

    • A.Its route table sends internet-bound traffic to a NAT gateway
    • B.Its Availability Zone sits at an edge location near the internet
    • C.Its route table sends internet-bound traffic to an internet gateway✓ Answer
    • D.Its security groups allow inbound web traffic from the internet

    A route table decides where traffic goes by destination and target, and a subnet is public when its route table has a route to an internet gateway. A subnet whose internet-bound traffic goes to a NAT gateway is private, because the internet cannot start connections to it.

    Ref QB-2XC54QSource: Amazon VPC User Guide, How Amazon VPC works and Subnets for your VPCReport a problem with this question

  14. 14. Application servers in a private subnet must download software updates from the internet but must not accept connections started from the internet. What should the team add?

    • A.A VPC peering connection, which reaches the internet through a partner
    • B.An internet gateway, which lets only outbound traffic out of the VPC
    • C.A virtual private gateway, which sends the servers' updates to the web
    • D.A NAT gateway, which lets the servers start outbound connections✓ Answer

    A NAT gateway lets instances in a private subnet initiate outbound connections while blocking unsolicited inbound ones. An internet gateway allows two-way communication, a virtual private gateway is the AWS side of a VPN, and peering connects two VPCs privately.

    Ref QB-HE7YGPSource: Amazon VPC User Guide, NAT gatewaysReport a problem with this question

  15. 15. Which statement correctly compares security groups and network ACLs in a VPC?

    • A.Security groups are stateful at the instance level, network ACLs stateless at the subnet level✓ Answer
    • B.Security groups are stateless at the instance level, network ACLs stateful at the subnet level
    • C.Security groups are stateful at the subnet level, network ACLs stateless at the instance level
    • D.Security groups are stateless at the subnet level, network ACLs stateful at the instance level

    A security group applies to an instance and is stateful, so return traffic is allowed automatically. A network ACL applies to a subnet and is stateless, so return traffic must be allowed explicitly by a rule.

    Ref QB-ZWZSWNSource: Amazon VPC User Guide, Infrastructure security: compare security groups and network ACLsReport a problem with this question

  16. 16. Logs show unwanted traffic from one specific IP address, and the team wants to block it for an entire subnet. Which control fits, and why?

    • A.Amazon Inspector, because its scans can automatically cut off a source
    • B.A network ACL, because its rules can explicitly deny a specific source✓ Answer
    • C.A security group, because its rules can explicitly deny a specific source
    • D.A route table, because its routes can explicitly drop a specific source

    Network ACLs work at the subnet level and support both allow and deny rules, so they can block a specific address. Security groups support allow rules only, route tables decide where traffic goes, and Amazon Inspector finds vulnerabilities rather than filtering traffic.

    Ref QB-GEUMBNSource: Amazon VPC User Guide, Control subnet traffic with network access control listsReport a problem with this question

  17. 17. A company wants to register a new domain name, translate that name into IP addresses and steer users away from an unhealthy endpoint at the DNS level. Which service provides all of this?

    • A.Elastic Load Balancing, which distributes requests to registered targets
    • B.Amazon Route 53, which provides a highly available managed DNS web service✓ Answer
    • C.Amazon CloudFront, which caches content at edge locations near viewers
    • D.AWS Global Accelerator, which gives static IPs on the AWS global network

    Amazon Route 53 is a DNS web service with three main functions: domain registration, DNS routing and health checking, and it can route traffic away from unhealthy resources. Amazon CloudFront caches content, and Elastic Load Balancing spreads traffic among targets, neither of which registers domains.

    Ref QB-YWNFRMSource: Amazon Route 53 Developer Guide, What is Amazon Route 53Report a problem with this question

  18. 18. A firm moves large data volumes between its data center and AWS every day. It needs consistent network performance and does not want the traffic to cross the public internet. Which option best fits?

    • A.AWS Direct Connect, a dedicated private connection from the data center✓ Answer
    • B.AWS Transit Gateway, a Regional hub that removes the need for any circuit
    • C.AWS Client VPN, a managed service that links user devices to the VPC
    • D.AWS Site-to-Site VPN, a pair of encrypted tunnels that bypass the internet

    AWS Direct Connect links a data center to AWS over a dedicated private fiber connection that bypasses internet service providers, which gives more consistent network performance. AWS Site-to-Site VPN is encrypted but runs over the internet, so its performance varies with internet conditions.

    Ref QB-KNCBH3Source: AWS Direct Connect User Guide, What is AWS Direct Connect; AWS Direct Connect FAQsReport a problem with this question

  19. 19. A company has chosen AWS Direct Connect. A manager assumes that because the connection is private, its traffic is already encrypted. Which statement is accurate?

    • A.Its link is private, and IPsec encryption applies automatically
    • B.Its link runs over the internet, with TLS added by the provider
    • C.Its link runs over the internet, with IPsec added on two tunnels
    • D.Its link is private, and encryption is an add-on chosen separately✓ Answer

    Private does not mean encrypted: AWS Direct Connect keeps traffic off the public internet, but encryption is something the customer adds, for example MACsec on select dedicated connections. The two-tunnel IPsec design describes AWS Site-to-Site VPN.

    Ref QB-3676X2Source: AWS Direct Connect FAQs, MACsec and security; AWS Direct Connect User GuideReport a problem with this question

  20. 20. A company with one office needs an encrypted connection from its office network to a VPC. It must be set up quickly and cheaply, and internet variability is acceptable. Which option best fits?

    • A.AWS Site-to-Site VPN, an IPsec connection built over the internet✓ Answer
    • B.AWS PrivateLink, a private endpoint that reaches the on-site network
    • C.AWS Direct Connect, a fiber link ordered through a physical location
    • D.AWS Client VPN, a TLS session each user starts from their own device

    AWS Site-to-Site VPN creates an encrypted IPsec connection over the internet between an on-premises network and a VPC, and it can be set up quickly. AWS Direct Connect takes longer to establish, and AWS Client VPN serves individual users rather than a whole office network.

    Ref QB-JQX5C2Source: AWS Site-to-Site VPN User Guide, What is AWS Site-to-Site VPNReport a problem with this question

  21. 21. Remote employees working from home and while traveling need secure access to resources in a VPC from their own laptops. Which option best fits?

    • A.AWS Transit Gateway, a Regional hub that each laptop attaches to directly
    • B.AWS Client VPN, a managed client-based VPN that each person connects to✓ Answer
    • C.AWS Direct Connect, a dedicated private line run to each employee's home
    • D.AWS Site-to-Site VPN, an IPsec tunnel that each laptop maintains itself

    AWS Client VPN is a managed, client-based VPN that lets individual users connect securely from any location. AWS Site-to-Site VPN joins a whole office or data center network to a VPC, and AWS Direct Connect is a dedicated link to a facility, not to a person's home laptop.

    Ref QB-385BX2Source: AWS Client VPN Administrator Guide, What is AWS Client VPNReport a problem with this question

  22. 22. A company has dozens of VPCs and several on-premises networks. It wants one central hub to connect them instead of building many separate one-to-one links. Which option best fits?

    • A.AWS PrivateLink, a private endpoint that joins whole networks to each other
    • B.Amazon Route 53, a DNS service that joins the networks through name lookups
    • C.AWS Transit Gateway, a Regional router that interconnects many networks✓ Answer
    • D.Amazon VPC peering, a single link that joins every network to all the rest

    AWS Transit Gateway acts as a Regional virtual router and hub for many VPCs and on-premises networks. VPC peering connects two VPCs one-to-one, so it would have to be repeated for each pair, while AWS PrivateLink exposes specific services rather than joining whole networks.

    Ref QB-NGJF7USource: Amazon VPC Transit Gateways guide, What is a transit gateway; Amazon VPC Peering GuideReport a problem with this question

  23. 23. Instances in a VPC must reach a supported AWS service privately, with no internet gateway, NAT device or public IP address in the path. Which option best fits?

    • A.AWS Transit Gateway, which exposes the service through a Regional hub
    • B.AWS Global Accelerator, which exposes the service through static IPs
    • C.AWS PrivateLink, which exposes the service through a VPC endpoint✓ Answer
    • D.AWS Direct Connect, which exposes the service through a dedicated line

    AWS PrivateLink provides private access to supported services through VPC endpoints, without an internet gateway, NAT device, public IP address, AWS Direct Connect or VPN. AWS Global Accelerator uses public static IP addresses, and AWS Transit Gateway connects networks rather than exposing services.

    Ref QB-6ZQMSUSource: AWS PrivateLink Guide, What is AWS PrivateLinkReport a problem with this question

  24. 24. A team wants a managed service to create, publish and manage a REST API that mobile apps will call, with AWS Lambda functions running behind it. Which service fits this need?

    • A.AWS PrivateLink, which publishes the API to apps on the internet
    • B.AWS Global Accelerator, which creates API routes and caches responses
    • C.Amazon Route 53, which resolves the API's name and manages versions
    • D.Amazon API Gateway, which creates and manages APIs in front of backends✓ Answer

    Amazon API Gateway is the service for creating, publishing and managing APIs, and it can front backends such as AWS Lambda functions. Amazon Route 53 only handles DNS, AWS PrivateLink provides private connectivity rather than public APIs, and AWS Global Accelerator does not cache.

    Ref QB-MQNTP4Source: Amazon API Gateway Developer Guide, What is Amazon API GatewayReport a problem with this question

Other topics in this exam

Practice questions built from the domains and task statements in the published AWS exam guide (CLF-C02). AWS and AWS Certified Cloud Practitioner are trademarks of Amazon.com, Inc. or its affiliates; this site is not affiliated with or endorsed by Amazon Web Services, and these are not real exam questions. Fees, scoring, exam languages, Support plans, service names and prices change over time — confirm the current details on the AWS website before you book. The real exam also has multiple-response questions, which this single-answer practice does not reproduce. AWS exam guide (CLF-C02) →