I85PREP
⌘ K
Alibaba ACP Cloud Computing study materials
Alibaba ACP Cloud Computing study resources

Alibaba ACP Cloud Computing practice questions

One question at a time. Build your Alibaba ACP Cloud Computing understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which characteristic of cloud computing lets a customer automatically add ECS instances during a traffic spike and release them afterwards?

  1. Rapid elasticity Correct answer.
  2. Regional redundancy
  3. Pay-as-you-go billing
  4. Multi-tenant isolation

Explanation

Rapid elasticity is the property that resources can be provisioned and released automatically to match demand, scaling outward and inward as load changes. Isolation, billing model, and redundancy describe other aspects of cloud service delivery.

Question 2Cloud Computing Fundamentals and Alibaba Cloud Architecture

A company wants to run its own operating system and middleware on virtual machines in Alibaba Cloud while avoiding hardware procurement. Which service model fits?

  1. Software as a Service
  2. Function as a Service
  3. Infrastructure as a Service Correct answer.
  4. Platform as a Service

Explanation

Infrastructure as a Service provides virtualized compute, storage, and networking on which the customer installs and manages the OS and middleware. SaaS delivers finished applications, PaaS supplies a managed runtime, and FaaS runs event-driven code without server management.

Question 3Cloud Computing Fundamentals and Alibaba Cloud Architecture

What is the primary purpose of an Alibaba Cloud region?

  1. To host only managed database services
  2. To define a billing currency for resources
  3. To provide a shared global network backbone
  4. To group resources into a single fault-isolated geographic area Correct answer.

Explanation

A region is a geographic area containing one or more zones, and resources in different regions are isolated from each other for latency, compliance, and fault containment. Billing currency, global networking, and service scope are not what defines a region.

Question 4Cloud Computing Fundamentals and Alibaba Cloud Architecture

Within an Alibaba Cloud region, what does a zone primarily provide?

  1. A dedicated global DNS resolver
  2. A separate billing account
  3. A distinct regulatory jurisdiction
  4. An independent power and network fault domain Correct answer.

Explanation

A zone is an isolated facility with independent power, cooling, and network within a region, so deploying across zones improves availability. Zones do not change billing accounts, DNS scope, or legal jurisdiction.

Question 5Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which Alibaba Cloud component provides a private, logically isolated virtual network in which you define subnets and route tables?

  1. Alibaba Cloud DNS
  2. Cloud Enterprise Network
  3. Server Load Balancer
  4. Virtual Private Cloud Correct answer.

Explanation

A Virtual Private Cloud is the isolated virtual network construct where you define address ranges, subnets, and routing. Load balancing, enterprise interconnection, and DNS are separate services that operate on or across a VPC.

Question 6Cloud Computing Fundamentals and Alibaba Cloud Architecture

An application must remain available even if one data center fails. Which deployment approach best satisfies this within a single region?

  1. Deploy instances across multiple zones Correct answer.
  2. Deploy instances in a single region only
  3. Deploy all instances in one vSwitch
  4. Deploy all instances in one zone

Explanation

Spreading instances across multiple zones in a region means a single facility failure does not take down the whole application. Concentrating resources in one zone or one vSwitch creates a single point of failure.

Question 7Cloud Computing Fundamentals and Alibaba Cloud Architecture

In the Alibaba Cloud shared responsibility model, which task belongs to the customer?

  1. Maintaining the physical security of data centers
  2. Replacing failed physical disks in the storage cluster
  3. Patching the guest operating system of an ECS instance Correct answer.
  4. Managing the hypervisor layer

Explanation

The provider secures facilities and the virtualization layer, while the customer is responsible for what runs inside the instance, including guest OS patching. Physical security, disk replacement, and hypervisor management remain provider duties.

Question 8Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which statement best describes the relationship between Alibaba Cloud ECS and its underlying physical hosts?

  1. ECS instances require dedicated hardware in every case
  2. ECS instances are virtual machines scheduled on shared physical hosts Correct answer.
  3. ECS instances are containers without a guest OS
  4. Each ECS instance is bound permanently to one physical host

Explanation

ECS instances are virtual machines that the platform schedules onto physical hosts, which are typically shared among tenants unless a dedicated-host product is used. They are not permanently bound, container-only, or always on dedicated hardware.

Question 9Cloud Computing Fundamentals and Alibaba Cloud Architecture

A team needs block storage that can be attached to an ECS instance and persists independently of the instance lifecycle. Which service should they choose?

  1. Object Storage Service
  2. ApsaraDB for RDS
  3. Cloud Paralleled File System
  4. Elastic Block Storage Correct answer.

Explanation

Elastic Block Storage provides block-level volumes that attach to ECS instances and can outlive an instance. Object storage is for objects over HTTP, file storage is a shared file system, and RDS is a managed database.

Question 10Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which Alibaba Cloud service is designed to distribute incoming traffic across a group of ECS instances in one or more zones?

  1. Elastic IP Address
  2. Cloud Enterprise Network
  3. Server Load Balancer Correct answer.
  4. Resource Access Management

Explanation

Server Load Balancer accepts client traffic and forwards it to healthy backend instances according to a listener and health-check configuration. An EIP is a public address, CEN interconnects networks, and RAM controls permissions.

Question 11Cloud Computing Fundamentals and Alibaba Cloud Architecture

What is the main function of Alibaba Cloud Resource Access Management (RAM)?

  1. Controlling which identities can perform which actions on which resources Correct answer.
  2. Encrypting data at rest in OSS buckets
  3. Providing managed message queues between services
  4. Balancing traffic across availability zones

Explanation

RAM manages identities such as users, groups, and roles and authorizes their actions through policies scoped to resources. Encryption, traffic balancing, and messaging are handled by other services.

Question 12Cloud Computing Fundamentals and Alibaba Cloud Architecture

An application running on an ECS instance needs temporary credentials to read objects from OSS without storing long-term access keys on disk. Which approach is most appropriate?

  1. Attaching a RAM role to the ECS instance Correct answer.
  2. Sharing the primary account AccessKey with the team
  3. Creating a RAM user with administrator policy
  4. Embedding an AccessKey pair in the application source code

Explanation

A RAM role attached to an instance lets the instance obtain short-lived credentials from the metadata service, so no long-term secret is stored. Hard-coded keys, over-privileged users, and shared root keys all increase exposure.

Question 13Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which design most improves the fault tolerance of a stateless web tier in Alibaba Cloud?

  1. Run all instances on one ECS instance with a larger specification
  2. Use a single vSwitch for all instances
  3. Store session state only in local instance memory
  4. Place instances behind a load balancer across multiple zones Correct answer.

Explanation

Combining a load balancer with multi-zone instance distribution removes single points of failure and lets healthy backends absorb traffic. A single large instance, local-only session state, and one vSwitch all concentrate risk.

Question 14Cloud Computing Fundamentals and Alibaba Cloud Architecture

A workload requires very low, predictable latency to users in Frankfurt while data must remain in Germany. Which Alibaba Cloud placement decision is most appropriate?

  1. Deploy only with global CDN and no origin region
  2. Deploy in a region geographically close to Frankfurt Correct answer.
  3. Deploy in multiple regions far from Frankfurt
  4. Deploy in a single zone in another continent

Explanation

Choosing a nearby region keeps latency low while satisfying the data-residency requirement because resources stay within that geography. Distant zones, CDN-only designs without an origin, and far-flung multi-region layouts conflict with latency or residency goals.

Question 15Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which statement about a vSwitch in an Alibaba Cloud VPC is accurate?

  1. A vSwitch spans all zones in a region
  2. A vSwitch replaces the need for a route table
  3. A vSwitch provides public internet access by default
  4. A vSwitch is confined to a single zone and a subnet of the VPC CIDR Correct answer.

Explanation

A vSwitch is bound to one zone and carves out a subnet from the VPC address range, so multi-zone designs use multiple vSwitches. It does not span zones, grant internet access automatically, or eliminate routing.

Question 16Cloud Computing Fundamentals and Alibaba Cloud Architecture

Instances in a private subnet of a VPC need to download software updates from the internet without being directly reachable from outside. Which component enables this?

  1. An Elastic IP bound to every instance
  2. An inbound security group rule allowing all traffic
  3. A NAT gateway Correct answer.
  4. A public-facing Server Load Balancer

Explanation

A NAT gateway lets instances in private subnets initiate outbound internet connections while remaining unreachable from the internet. Opening inbound rules, exposing a public load balancer, or binding EIPs each makes instances externally reachable.

Question 17Cloud Computing Fundamentals and Alibaba Cloud Architecture

What is the role of a security group in Alibaba Cloud?

  1. To assign private IP addresses to vSwitches
  2. To replicate data across regions
  3. To encrypt traffic between zones
  4. To act as a stateful virtual firewall for instance network traffic Correct answer.

Explanation

A security group is a stateful virtual firewall whose rules allow or deny traffic to and from associated instances. Encryption, IP address assignment, and cross-region replication are handled by other mechanisms.

Question 18Cloud Computing Fundamentals and Alibaba Cloud Architecture

An architect wants to decouple a web front end from a slow backend processing service so that traffic bursts do not overwhelm the backend. Which pattern fits best?

  1. Storing requests in local instance memory
  2. A larger single backend instance
  3. A message queue between the front end and backend workers Correct answer.
  4. Synchronous direct calls with long timeouts

Explanation

A message queue buffers requests so producers and consumers can scale and fail independently, absorbing bursts. Synchronous calls propagate load, a bigger instance remains a bottleneck, and local memory is lost on failure.

Question 19Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which Alibaba Cloud storage service is best suited for storing millions of unstructured images served directly to web clients?

  1. ApsaraDB for RDS
  2. Object Storage Service Correct answer.
  3. Cloud Paralleled File System
  4. Elastic Block Storage

Explanation

Object Storage Service is designed for massive volumes of unstructured objects accessed over HTTP, making it suitable for image serving. Block storage attaches to instances, RDS is relational, and CPFS is a shared file system for high-performance compute.

Question 20Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which statement correctly distinguishes horizontal from vertical scaling?

  1. Horizontal scaling adds instances; vertical scaling increases the size of one instance Correct answer.
  2. Horizontal scaling increases instance size; vertical scaling adds instances
  3. Both terms describe adding storage capacity only
  4. Vertical scaling requires a load balancer; horizontal scaling never does

Explanation

Horizontal scaling, or scaling out, adds more instances to share load, while vertical scaling, or scaling up, enlarges a single instance. The definitions are not about storage only, and load balancers are typically used with horizontal scaling rather than being forbidden for vertical.

Question 21Cloud Computing Fundamentals and Alibaba Cloud Architecture

An e-commerce site expects unpredictable spikes and must keep costs proportional to usage while scaling automatically. Which combination best meets these goals?

  1. Manual provisioning of instances after each spike is reported
  2. Reserved instances sized for peak load in one zone
  3. A single large dedicated host running all services
  4. Auto Scaling across zones with pay-as-you-go instances behind a load balancer Correct answer.

Explanation

Auto Scaling adjusts capacity to demand, multi-zone placement adds resilience, pay-as-you-go billing matches cost to usage, and the load balancer distributes traffic. Peak-sized reservations, single hosts, and manual reaction all waste cost or fail under spikes.

Question 22Cloud Computing Fundamentals and Alibaba Cloud Architecture

A company must keep its database in Alibaba Cloud but connect it privately to an on-premises data center with predictable bandwidth and no traversal of the public internet. Which service is most appropriate?

  1. Alibaba Cloud DNS private zone
  2. A public Elastic IP on the database
  3. Express Connect dedicated connection Correct answer.
  4. An internet-facing Server Load Balancer

Explanation

Express Connect provides a dedicated private network link between on-premises environments and Alibaba Cloud, giving predictable bandwidth without public internet traversal. EIPs and internet-facing load balancers use public paths, and private DNS only resolves names.

Question 23Cloud Computing Fundamentals and Alibaba Cloud Architecture

A team deploys a stateful application across two zones but stores session data only on local disks of each instance. What is the most likely consequence during a zone failure?

  1. Users on the failed zone lose session state and may be logged out Correct answer.
  2. Sessions survive because two zones are used
  3. The load balancer automatically restores the lost session data
  4. The database replicates the local session files

Explanation

Multi-zone placement protects availability only for stateless components; local session data dies with the instance. A load balancer does not recreate application state, and a database does not automatically capture files stored on instance disks.

Question 24Cloud Computing Fundamentals and Alibaba Cloud Architecture

An application writes to OSS and requires that data remain readable even if an entire zone becomes unavailable. Which OSS configuration characteristic should be verified?

  1. That the bucket is in a single vSwitch
  2. That the bucket uses a multi-zone resilient storage class Correct answer.
  3. That the bucket name is globally unique
  4. That the bucket has versioning disabled

Explanation

Zone-level resilience for object data depends on the bucket's storage class and redundancy configuration, which determines how data is replicated across zones. Name uniqueness, versioning, and vSwitch association do not by themselves provide zone resilience.

Question 25Cloud Computing Fundamentals and Alibaba Cloud Architecture

A solutions architect must design for both cost efficiency and resilience. Which reasoning is most sound when choosing between a single-zone and multi-zone deployment for a non-critical batch tier?

  1. Use single-zone for the non-critical tier and reserve multi-zone for critical tiers Correct answer.
  2. Use multi-zone only when the region has one zone
  3. Avoid load balancers to reduce cross-zone charges in all tiers
  4. Always use multi-zone for every tier regardless of criticality

Explanation

Resilience spending should track business criticality, so a non-critical batch tier can accept single-zone risk while critical tiers pay for multi-zone protection. Blanket multi-zone, zone-count conditions, and avoiding load balancers everywhere misapply the trade-off.

Question 26Cloud Computing Fundamentals and Alibaba Cloud Architecture

Which Alibaba Cloud service provides a fully managed, distributed message queue that decouples producers and consumers in a cloud-native application?

  1. Alibaba Cloud Resource Access Management (RAM)
  2. Alibaba Cloud Object Storage Service (OSS)
  3. Alibaba Cloud Message Queue (MQ) Correct answer.
  4. Alibaba Cloud Server Load Balancer (SLB)

Explanation

Message Queue (MQ) is a managed messaging service that enables asynchronous communication and decoupling between application components. It supports reliable message delivery and is designed for distributed systems.

Question 27Cloud Computing Fundamentals and Alibaba Cloud Architecture

A company wants to run a containerized application on Alibaba Cloud with minimal management of the underlying cluster infrastructure. Which service should they use?

  1. Alibaba Cloud Elastic Compute Service (ECS) with self-managed Docker
  2. Alibaba Cloud Server Load Balancer (SLB)
  3. Alibaba Cloud Function Compute
  4. Alibaba Cloud Container Service for Kubernetes (ACK) Correct answer.

Explanation

ACK is a managed Kubernetes service that handles the provisioning, scaling, and maintenance of the Kubernetes control plane and worker nodes, reducing operational overhead. It allows teams to focus on deploying containerized applications rather than managing the cluster infrastructure.

Question 28Cloud Computing Fundamentals and Alibaba Cloud Architecture

A solutions architect is designing a system that must handle sudden spikes in read traffic to a product catalog stored in a relational database. The catalog data changes infrequently. Which approach best improves read scalability while reducing database load?

  1. Configure an Alibaba Cloud Server Load Balancer to distribute read queries across multiple database replicas
  2. Migrate the database to an ECS instance with a larger CPU and memory specification
  3. Store the catalog in Alibaba Cloud Object Storage Service (OSS) and serve it directly to users
  4. Enable Alibaba Cloud Redis as a read-through cache for catalog queries Correct answer.

Explanation

Using Redis as a cache for frequently read, infrequently changing data reduces the number of queries hitting the relational database, improving read scalability and lowering load. This is a common pattern for read-heavy workloads with static or slowly changing data.

Question 29Cloud Computing Fundamentals and Alibaba Cloud Architecture

A company is deploying a new application on Alibaba Cloud and needs to choose between a VPC and the classic network. Which statement accurately describes a key difference between these network types?

  1. A VPC allows instances to communicate over the public internet by default, while the classic network requires manual public IP assignment.
  2. A VPC provides logical isolation with user-defined IP ranges and routing, while the classic network shares a common address pool managed by Alibaba Cloud. Correct answer.
  3. A VPC does not support security groups, whereas the classic network does.
  4. A VPC is limited to a single availability zone, while the classic network spans multiple zones.

Explanation

A VPC (Virtual Private Cloud) gives users full control over their virtual network topology, including IP address ranges, subnets, and route tables, and provides logical isolation. The classic network is a shared network environment managed by Alibaba Cloud with less customization and isolation.

Question 30Cloud Computing Fundamentals and Alibaba Cloud Architecture

An application running on Alibaba Cloud uses Auto Scaling to manage a fleet of ECS instances behind a Server Load Balancer. The scaling group is configured with a minimum of 2 and maximum of 10 instances, and the scaling policy is based on average CPU utilization. During a traffic surge, the average CPU utilization exceeds the threshold, but the scaling group does not add instances. Which is the most likely cause?

  1. The scaling group has reached its maximum instance limit. Correct answer.
  2. The scaling policy is set to scale in rather than scale out.
  3. The ECS instances are deployed in a single availability zone.
  4. The Server Load Balancer health check is failing for all instances.

Explanation

Auto Scaling will not add instances beyond the maximum size configured for the scaling group. If the group is already at its maximum of 10 instances, it cannot scale out further even if the scaling policy triggers.

Question 31Elastic Compute Service (ECS)

An ECS instance is created in a VPC. Which component assigns the private IP address that the instance uses to communicate inside that VPC?

  1. The vSwitch to which the instance's ENI is attached Correct answer.
  2. The Security Group bound to the instance
  3. The route table of the VPC
  4. The ECS instance's own operating system DHCP client pool

Explanation

A vSwitch is the subnet-level construct inside a VPC, and its CIDR block is the pool from which private IP addresses are drawn for ENIs attached to it. Security groups filter traffic but do not allocate addresses, and route tables only direct traffic.

Question 32Elastic Compute Service (ECS)

An application running on an ECS instance needs to call an Alibaba Cloud API without storing an AccessKey on disk. Which mechanism should be used?

  1. A NAT Gateway with a shared bandwidth package
  2. An SSH key pair associated with the instance
  3. A user data script that embeds the AccessKey
  4. An instance RAM role attached to the instance Correct answer.

Explanation

An instance RAM role lets the platform deliver temporary STS credentials to the instance, so no long-lived AccessKey is stored on disk. The credentials are rotated automatically and scoped by the role's policy.

Question 33Elastic Compute Service (ECS)

Which ECS disk category is a shared block storage service whose performance scales with capacity and is suited to highly concurrent read/write workloads?

  1. Ultra disk Correct answer.
  2. Basic disk
  3. ESSD (Enhanced SSD)
  4. Local SSD disk

Explanation

Ultra disks are distributed block storage whose IOPS and throughput scale with disk size, making them suitable for high-concurrency workloads. ESSD offers predictable single-disk latency tiers, while local disks are tied to a single host.

Question 34Elastic Compute Service (ECS)

What is the primary purpose of an ECS security group?

  1. To provide a private connection between a VPC and an on-premises data center
  2. To distribute incoming traffic across multiple instances
  3. To encrypt data at rest on attached cloud disks
  4. To act as a stateful virtual firewall controlling inbound and outbound traffic for associated ENIs Correct answer.

Explanation

A security group is a stateful virtual firewall: allowed inbound traffic automatically permits its return traffic. Rules are applied at the ENI level, and an instance can belong to multiple groups.

Question 35Elastic Compute Service (ECS)

Which statement describes the relationship between an ECS instance type family and its underlying host?

  1. Instance families differ in CPU-to-memory ratio and are scheduled onto compatible hosts Correct answer.
  2. All instances of a family always run on the same physical host
  3. An instance type can be changed only by recreating the VPC
  4. The instance family determines the public IP address range

Explanation

Instance families group specifications with a characteristic CPU-to-memory ratio (compute-optimized, memory-optimized, general purpose). The platform schedules an instance onto a host that can satisfy the chosen specification.

Question 36Elastic Compute Service (ECS)

An operator wants an ECS instance to run a bootstrap script automatically the first time it starts. Which feature should be used?

  1. A snapshot policy
  2. User data Correct answer.
  3. A custom image
  4. Cloud Assistant command history

Explanation

User data supplies a script that the platform passes to the instance at first launch, typically consumed by cloud-init. It is the standard way to automate initial configuration.

Question 37Elastic Compute Service (ECS)

Which ECS purchase model is best suited to a steady, long-running production workload where the lowest per-hour cost matters most?

  1. Pay-as-you-go
  2. Subscription Correct answer.
  3. Savings plan with no commitment
  4. Spot instance

Explanation

Subscription (prepaid) billing commits to a term and yields a lower unit price than pay-as-you-go for predictable, always-on workloads. Pay-as-you-go suits variable demand and spot suits interruptible work.

Question 38Elastic Compute Service (ECS)

Which action is required before you can attach an existing data disk to an additional ECS instance?

  1. Detach it from its current instance first Correct answer.
  2. Change its disk category to basic
  3. Delete its automatic snapshots
  4. Convert it to a system disk

Explanation

A data disk can be attached to only one instance at a time, so it must be detached from its current instance before being attached elsewhere. The data on the disk is retained through detach and reattach.

Question 39Elastic Compute Service (ECS)

An instance in a VPC must reach the public internet for software updates, but it must not be directly reachable from the internet. Which configuration achieves this?

  1. Enable a public bandwidth package on the instance's ENI
  2. Bind the instance to a public-facing Server Load Balancer
  3. Assign an EIP and open all inbound ports
  4. Place the instance in a private vSwitch and route outbound traffic through a NAT Gateway Correct answer.

Explanation

A NAT Gateway performs source NAT for instances in private vSwitches, giving outbound-only internet access. Because no public address is bound to the instance, inbound connections from the internet cannot be initiated.

Question 40Elastic Compute Service (ECS)

A workload must survive the failure of a single Availability Zone. What is the minimum correct design on ECS?

  1. Deploy all instances in one zone but on different hosts
  2. Enable automatic snapshot policies for all disks
  3. Deploy instances across at least two zones in the same region behind a load balancer Correct answer.
  4. Attach multiple ENIs to each instance

Explanation

Availability Zones within a region have independent power and network, so spreading instances across zones removes a single-zone single point of failure. A load balancer then health-checks and routes around failed instances.

Question 41Elastic Compute Service (ECS)

A security group rule allows inbound TCP port 443 from 0.0.0.0/0. A client connects successfully and receives responses. Which statement is true about the return traffic?

  1. The response is allowed automatically because security groups are stateful Correct answer.
  2. The response is dropped unless the ENI has a public IP
  3. A separate outbound rule is required for the response
  4. The response is allowed only if the client is in the same vSwitch

Explanation

Security groups are stateful: once a connection is permitted inbound, the corresponding return traffic is automatically allowed regardless of outbound rules. This is unlike stateless ACLs, which require mirrored rules.

Question 42Elastic Compute Service (ECS)

An administrator wants to replace the operating system on a running instance with a different image while keeping the instance's existing data disk. Which approach is appropriate?

  1. Detach the data disk, release the instance, and create a new one from the image
  2. Use the Change Operating System feature on the instance and keep the data disk attached Correct answer.
  3. Stop the instance and convert the data disk into a system disk
  4. Take a snapshot and restore it to the system disk of a new instance

Explanation

ECS provides an operating system replacement action that reinstalls the system disk from a chosen image while the data disk is preserved. This avoids releasing the instance and losing its identity and network configuration.

Question 43Elastic Compute Service (ECS)

A batch job runs for about 40 minutes and can tolerate interruption. Cost must be minimized. Which purchase option fits best?

  1. Spot instance Correct answer.
  2. Subscription for one year
  3. Dedicated host
  4. Pay-as-you-go with a reserved instance

Explanation

Spot instances use spare capacity at a large discount but may be reclaimed, which is acceptable for interruptible batch work. The job should checkpoint progress to handle reclamation.

Question 44Elastic Compute Service (ECS)

An image is shared from Account A to Account B. Account B wants to launch instances from it in a different region. What must Account B do?

  1. Launch directly across regions from the shared image
  2. Convert the shared image into a snapshot in Account A
  3. Copy the shared image to the target region, then launch from the copy Correct answer.
  4. Re-share the image back to Account A first

Explanation

Images are regional resources, so an image available in one region must be copied to the target region before instances can be created there. Sharing grants visibility but does not replicate the image across regions.

Question 45Elastic Compute Service (ECS)

Which statement correctly describes the behavior of a pay-as-you-go instance that is stopped (not released) while it has a public IP assigned by the system?

  1. All attached data disks are automatically deleted
  2. The public IP is retained permanently and never changes
  3. The system-assigned public IP is released and may change on next start Correct answer.
  4. The instance is billed for vCPU and memory while stopped

Explanation

A system-assigned public IP is not a persistent resource; stopping the instance releases it, and a new address may be assigned on the next start. An EIP, by contrast, is a separately held resource that persists.

Question 46Elastic Compute Service (ECS)

A team needs a consistent point-in-time recovery copy of a data disk before a risky upgrade. Which action should be taken?

  1. Create a new ENI and attach it
  2. Enable a security group rule for backup traffic
  3. Detach the disk and store it offline
  4. Create a snapshot of the disk Correct answer.

Explanation

A snapshot captures the disk state at a point in time and can be used to roll back or create a new disk. It is the standard ECS mechanism for disk-level backup.

Question 47Elastic Compute Service (ECS)

An instance's system disk is nearly full and the application cannot be paused. What is a valid way to add storage without reinstalling the OS?

  1. Convert the system disk to a shared block device
  2. Resize the system disk downward to free space
  3. Attach a new data disk, format it, and mount it at the application's data path Correct answer.
  4. Detach the system disk while the instance runs

Explanation

Adding a data disk and mounting it at the path the application writes to expands usable storage without touching the running system disk. The new disk must be partitioned, formatted, and mounted before use.

Question 48Elastic Compute Service (ECS)

Which ECS feature allows commands or scripts to be run on many instances at once without logging in over SSH?

  1. User data
  2. Instance metadata service
  3. Cloud Assistant Correct answer.
  4. Serial console

Explanation

Cloud Assistant is an agent-based service that executes commands or scripts on selected instances and returns output, enabling fleet-wide automation. It works without inbound SSH access.

Question 49Elastic Compute Service (ECS)

An instance in a private vSwitch cannot reach the internet even though a NAT Gateway exists in the VPC. Which item is most likely misconfigured?

  1. The route table entry directing 0.0.0.0/0 to the NAT Gateway Correct answer.
  2. The instance's SSH key pair
  3. The instance's hostname
  4. The disk category of the system disk

Explanation

Outbound internet access through a NAT Gateway requires a route in the vSwitch's route table sending default traffic to the gateway. Without that route, packets have no path to the NAT device.

Question 50Elastic Compute Service (ECS)

What does the instance metadata service provide to code running inside an ECS instance?

  1. Direct access to other tenants' instances
  2. Persistent storage for application data
  3. A writable copy of the VPC route table
  4. Information such as instance ID, region, and the credentials of an attached RAM role Correct answer.

Explanation

The metadata service is reachable from inside the instance and returns instance attributes plus temporary credentials when a RAM role is attached. Applications use it to discover their environment without hard-coded configuration.