ACP-Cloud1 ACP Cloud Computing Certification Questions and Answers

For Layer-4 SLB using TCP, the health check determines ECS instance availability based on the successful completion of a TCP 3-way handshake. This ensures that the backend server is reachable and can accept connections, which is critical for reliable TCP-based load balancing.

=================

When the removal policy is set to “the earliest instance with the scaling configuration,” Auto Scaling will prioritize removing the oldest ECS instances that were created by Auto Scaling. This policy helps manage and rotate instances that may have been running the longest, ensuring newer instances remain available to handle the current load. This approach is beneficial for maintaining cost-effectiveness and operational efficiency.

Before data communication is set up, the security groups match the security group rules one by one to query whether to allow access requests. If the protocol type, port range, authorization type, and authorization object of two security group rules are the same, the following rules apply:

If the two rules have the same priority, the rule of the denial policy takes effect first, and the rule of the allowed policy does not take effect. This is because the security group rules follow the principle of minimum permission, which means that the most restrictive rule is applied when there is a conflict.

If the two rules have different priorities, the rule with a smaller priority number takes effect. This is because the security group rules follow the principle of priority, which means that the rule with a higher priority (lower number) is applied when there is a conflict. References: ECS Security Groups - Alibaba Fundamentals - Cloud Academy, Security group rules - Elastic Compute Service - Alibaba Cloud Documentation Center

Alibaba Cloud Auto Scaling ensures transaction completeness during scaling activities. If the intended number of ECS instances cannot be added to the scaling group, Auto Scaling rolls back the entire activity, releasing any partially added instances. This behavior prevents incomplete scaling activities, ensuring the scaling action either fully succeeds or is reverted entirely.

CloudMonitor allows users to create custom monitoring metrics and report data, which can then be analyzed and used for alerts. Customized monitoring is flexible, with no set limits on the number of metrics, and users can report data from Alibaba Cloud servers or other devices.

=================

A route table is a set of rules that determine how network traffic is routed within a VPC or between different network instances1. A VRouter is a hub that connects all VSwitches in the same VPC and enables communication between them2. When you create a VPC, a VRouter and a default route table are automatically created for the VPC1. The default route table contains system routes that are automatically generated by the system based on the network topology1. The system routes cannot be modified or deleted1.

However, you can also create custom route tables for your VPC and add custom routes to them1. Custom routes are manually configured by users to route traffic to specific destinations, such as other VPCs, VBRs, VPN Gateways, or NAT Gateways1. Custom routes can be modified or deleted as needed1. You can also associate VSwitches with different route tables to implement fine-grained routing control1.

Therefore, the statement that the routing entries of the route table can not be modified manually is not correct, as it only applies to system routes, not custom routes. The other statements are correct, as each VRouter may have multiple route tables, one default and several custom1; when a VPC is deleted, the corresponding route table is also deleted3; and the default route table cannot be deleted1. References: Route table overview - Virtual Private Cloud - Alibaba Cloud Documentation Center, VRouter overview - Virtual Private Cloud - Alibaba Cloud Documentation Center, Delete a VPC - Virtual Private Cloud - Alibaba Cloud Documentation Center

OSS does not use a hierarchical structure for objects, but instead uses a flat structure. All elements are stored as objects in buckets. To use OSS in the same manner in which local file systems are used, you can configure Cloud Storage Gateway (CSG).

https://www.alibabacloud.c om/help/doc-detail/31817.htm

 The statement C is correct, because in an OSS bucket, all elements are stored as objects. An object is the basic unit of OSS storage. Each object consists of a unique key, data, and metadata. The key is the name of the object, which is used to identify the object in a bucket. The data is the content of the object, which can be any type of file, such as text, image, video, or audio. The metadata is the information about the object, such as its size, type, creation time, and custom attributes. References: = Object Storage Service:Overview - Alibaba Cloud

To isolate ECS instances within different VSwitches in a VPC while allowing them to access the internet and other intranet resources, each VSwitch's instances should be in separate Security Groups. By configuring each group to allow all VPC CIDR blocks first and then restricting access to other VSwitch CIDR blocks with a higher priority rule, instances in each VSwitch are isolated without affecting the VPC's external connections. This approach follows Alibaba Cloud's Security Group best practices and CIDR-based access control within VPC environments.

=================

With a measurement cycle of 5 minutes and a requirement for five consecutive cycles where disk usage exceeds 80%, it would take 5×5=255 \times 5 = 255×5=25 minutes for CloudMonitor to confirm the condition. After confirmation, CloudMonitor typically processes and sends the alert, which may take an additional 5 minutes, totaling approximately 30 minutes before you receive the alert.

According to the Alibaba Cloud Auto Scaling documentation1, the Lifecycle Hook feature allows you to perform custom operations on instances that are added to or removed from a scaling group. You can define a lifecycle hook to specify a timeout period and a web hook URL. When an instance is about to be removed, Auto Scaling sends a notification to the web hook URL and waits for a response. During the timeout period, you can perform the necessary operations on the instance, such as gracefully shutting down the service, backing up the data, or sending a custom notification. After the operations are completed, you can send a response to the web hook URL to confirm the removal of the instance. This way, you can ensure that the instance is removed without affecting the user experience or causing data loss. Therefore, option D is the best solution to eliminate the customer’s concern. References: Lifecycle hooks and Alibaba Cloud Auto Scaling.

The maximum number of connections to an RDS instance primarily depends on the memory size of the instance. Higher memory capacity allows for more simultaneous connections, as memory is a key resource in managing and maintaining connections. Alibaba Cloud RDS scales the connection limit based on the instance's memory allocation to ensure stable database performance.

According to the Alibaba Cloud ECS documentation1, the recommended instance type for Tomcat is ecs.c5.xlarge, which has 4 vCPUs and 4 GiB of memory. This instance type is suitable for web applications that require high performance and low latency. The recommended instance type for Memcache is ecs.r5.large, which has 2 vCPUs and 16 GiB of memory. This instance type is optimized for memory-intensive applications that require high memory bandwidth and low latency. Therefore, option D is the most recommended configuration for deploying Tomcat and Memcache on two ECS instances. References: Instance type families and Alibaba Cloud Elastic Compute Service Product Introduction

According to the Alibaba Cloud ECS documentation1, the Pay-As-You-Go billing method is flexible and suitable for short-term or uncertain workloads. However, it is also the most expensive option compared to other billing methods. To reduce the cost of Pay-As-You-Go instances, you can purchase a Savings Plan, which is a discount plan that offers lower prices for a fixed amount of hourly usage. A Savings Plan can be applied to any Pay-As-You-Go instance in the same region, regardless of the instance type, operating system, or zone. Therefore, option A is the most cost-effective billing method for a 10-day test, as it allows you to enjoy the flexibility of Pay-As-You-Go and the savings of a fixed commitment. References: Billing methods and Savings Plan.

For a service focused on storing and providing fast access to a massive amount of images, the team should use:

Object Storage Service (OSS): For scalable and cost-effective storage of large image datasets.

Elastic Compute Service (ECS): For running the application’s backend and API logic.

Content Delivery Network (CDN): To cache and accelerate image delivery globally, improving access speed for users worldwide.While Auto Scaling is useful for applications with highly variable compute demands, it is not essential for this setup based on the described requirements.

 Alibaba Cloud CDN can record the real IP address of a visitor in the HTTP header field X-Forwarded-For, which can be obtained by the origin server or the user-defined log of Apache and Nginx. Alibaba Cloud CDN also provides a one-click function to enable the recording of the visitor’s real IP address in the console, which can be viewed in the access log. References:

Obtain the real IP address of a client from the origin server - Dynamic Content Delivery Network - Alibaba Cloud Documentation Center1

Welcome to Alibaba Cloud and find out more detail about cdn real ip2

Alibaba Cloud OSS automatically maintains three copies of each data object to ensure data reliability and redundancy. This setup allows OSS to recover data automatically without human intervention in case of hardware or system failures, ensuring high data durability.

According to the Alibaba Cloud Elastic Compute Service documentation, if the current public bandwidth does not meet your business requirements, you can upgrade or downgrade the public bandwidth configurations without the need to reboot the ECS instance. This can improve the network performance and response speed of the ECS instance. Upgrading the CPU, memory, or disk of the ECS instance may not solve the issue of bandwidth utilization, and may require rebooting the ECS instance. References: Public bandwidth - Elastic Compute Service

In Alibaba Cloud SLB, Layer-7 (HTTP/HTTPS) services require enabling the "X-Forwarded-For" header to retrieve the real IP address of clients, while Layer-4 (TCP/UDP) services can obtain the real IP address directly without additional settings.

=================

Modifying the MAC address of an ECS instance’s NIC can cause network disconnection issues because Alibaba Cloud's VPC network infrastructure relies on consistent NIC configuration. Changing the MAC address may prevent the instance from being identified correctly on the network.

=================

Elastic Computing on Alibaba Cloud ECS includes features that allow users to quickly start or release multiple instances, upgrade bandwidth without downtime, adjust CPU and memory configurations, and perform horizontal scaling in conjunction with Server Load Balancer (SLB). However, option E is incorrect as there is a limit to the number of disks that can be attached to an ECS instance.

Using lifecycle hooks in Auto Scaling allows for custom operations, such as updating the Redis whitelist when instances are added or removed. An OOS (Operation Orchestration Service) template can automate this process, and MNS (Message Notification Service) can trigger events to inform Redis when changes occur.

=================

Different Alibaba Cloud VPCs are completely isolated from each other by default, but they can communicate with each other over the intranet by using different methods, such as Cloud Enterprise Network (CEN), VPN gateways, VPC peering connections, and PrivateLink. These methods allow you to establish secure and reliable connections between VPCs without exposing your network traffic to the internet. References:

Overview of VPC connections - Alibaba Cloud

Cross-VPC peering - Alibaba Cloud

ALIYUN::VPC::VpcPeerConnection - Alibaba Cloud

According to the Alibaba Cloud Academy, a VPC is an isolated network environment that is completely isolated from each other. Each VPC has an independent tunnel ID, and a tunnel ID corresponds to a virtualized network. This means that different VPCs cannot communicate with each other through the tunnel ID. Subnets are not divided like the traditional network environment, but rather are logical divisions of the VPC. Different cloud servers within each subnet are connected by the same router, but different subnets are interconnected by route tables, not switches. ECS instances in a VPC use security group firewalls for Layer 4 network access control, not Layer 2. The internal networks of different VPCs are completely isolated and can be interconnected through IP addresses mapped to the outside, such as NAT Gateway or EIP. References: ACP Cloud Computing Certification Preparation Course - Internetworking with VPC, Alibaba Cloud VPC User Guide

When the black hole policy is triggered by excessive DDoS traffic, the affected server is temporarily isolated from the network to prevent further damage. During this period, no services are available, meaning the server cannot provide limited-speed services.

Alibaba Cloud OSS is designed for scalable and secure object storage, allowing users to upload and download files through multiple interfaces, including the web console and APIs, offering flexibility and ease of use for developers and administrators.

=================

RDS provides whitelist access policies. You can set permitted IP addresses and IP network segments to effectively prevent hackers from attacking the server by port scanning1. A whitelist is a list of IP addresses or CIDR blocks that are allowed to access an RDS instance. By default, the whitelist is empty, which means that no IP address can access the RDS instance. You can add IP addresses or CIDR blocks to the whitelist to allow access from specific sources. You can also modify or delete the whitelist as needed2. Whitelist access policies help improve the security of your RDS instance and protect your data from unauthorized access3. 

Alibaba Cloud OSS (Object Storage Service) is the ideal solution for storing large volumes of video data generated by a surveillance system. OSS provides scalable, secure, and cost-effective storage that can accommodate growing data needs without the space limitations of local ECS disks. Additionally, OSS enables easy access to historical data when required, making it well-suited for video storage in use cases like surveillance.

According to the Alibaba Cloud Academy, Auto Scaling supports three scaling modes: manual mode, scheduled task, and cloud monitor and event-triggered task. Manual mode requires manual intervention to adjust the number of ECS instances, which is not feasible and with minimum operation efforts. Scheduled task allows you to set the time and number of ECS instances to be added or removed in advance, which is suitable for predictable traffic changes. Cloud monitor and event-triggered task allows you to set the monitoring metrics and thresholds for scaling actions, which is suitable for dynamic traffic changes. Healthy check mode is not a scaling mode, but a feature of SLB to ensure the availability of ECS instances behind the load balancer. References: : ACP Cloud Computing Certification Preparation Course - Auto Scaling, : ACP Cloud Computing Certification Preparation Course - Server Load Balancer

The X-Forwarded-For field in HTTP header is used to identify the originating IP address of a client connecting to a web server through an HTTP proxy or a load balancer. When the “Obtain the Visitor’s Real IP Address” function is enabled in Alibaba Cloud SLB, the SLB instance adds the X-Forwarded-For field to the HTTP header of each request and forwards the request to the backend server. The backend server can then obtain the real IP address of the visitor from the X-Forwarded-For field1. The format of the X-Forwarded-For field is as follows:

X-Forwarded-For: client, proxy1, proxy2

where the value is a comma+space separated list of IP addresses, the left-most being the original client, and each successive proxy that passed the request adding the IP address where it received the request from. In this example, the request passed through proxy1, proxy2, and then the SLB instance (proxy3).2

References: Obtain client IP addresses - Server Load Balancer - Alibaba Cloud, X-Forwarded-For - MDN Web Docs

According to the Alibaba Cloud documentation1, Auto Scaling is a service that automatically adjusts the number of elastic computing resources based on your business demands and policies. When the demand for computing resources increases, Auto Scaling automatically adds ECS instances to ensure sufficient computing capabilities. When the demand decreases, Auto Scaling automatically removes ECS instances to reduce costs.

Auto Scaling supports adding an existing ECS instance into the scaling group, but the instance must meet some requirements1. One of the requirements is that the instance and the scaling group must be in the same region. However, they do not have to be in the same zone. A zone is a physical area within a region that has independent power grids and networks. A region is a geographic area where Alibaba Cloud deploys its resources. Therefore, the correct answer is B.

References:

1: Auto Scaling: Automatically Adjusts Computing Resources - Alibaba Cloud

2: Introduction to Auto Scaling - Alibaba Cloud Document Center

In Alibaba Cloud RDS, multiple databases can be created within a single instance, allowing users to manage different databases under one RDS instance efficiently. However, databases within the same instance must have unique names, and resources such as CPU and memory are shared across databases within the same instance, contrary to the options provided in answers B and C.

Weighted Round Robin (WRR) is a load balancing mode that assigns requests to backend ECS instances based on their weights. The higher the weight of the backend ECS instance, the higher the chance that the instance will receive requests. This mode can exploit the different service capacities of backend ECS instances by setting higher weights to ECS instances with higher capacities. For example, if there are two ECS instances in the backend server pool, one with a weight of 10 and the other with a weight of 20, the ECS instance with a weight of 20 will receive twice as many requests as the ECS instance with a weight of 10.

References:

1: SLB overview - Server Load Balancer - Alibaba Cloud Documentation Center

4: Server Load Balancer - Alibaba Cloud

5: Alibaba Server Load Balancer (SLB) Course - Cloud Academy

Alibaba Cloud ECS allows users to set an auto-release time for Pay-As-You-Go instances, giving flexibility in handling temporary workloads without manually releasing instances. Notably, the auto-release time can be adjusted multiple times even after the instance has been created, allowing for modifications as business needs change. This feature provides a convenient way to manage costs by automatically releasing instances when they are no longer needed. However, once the auto-release time is set, it must be adjusted or canceled if changes are required.

Alibaba Cloud provides comprehensive security protection covering networks, hosts, applications, databases, and content for on-cloud clients, but users still need to pay attention to security protection issues. Alibaba Cloud and its customers are jointly responsible for the security of customers’ applications built on Alibaba Cloud. With security responsibilities shared between Alibaba Cloud and its customers, Alibaba Cloud provides a secure infrastructure to decrease the enterprise security burden of customers. As such, customers can configure and use cloud products in a secure manner, thus relieving much of the underlying security burdens while allowing customers to focus more on their core business needs. However, customers still need to follow the best practices and guidelines for cloud security, such as using strong passwords, enabling multi-factor authentication, encrypting sensitive data, applying security patches, etc. Customers also need to monitor and manage the security risks and incidents of their cloud environment using Alibaba Cloud’s security products and services, such as Security Center, ActionTrail, Cloud Firewall, etc. References: Cloud Security- Alibaba Cloud, Security & Compliance Center, Shared Responsibility

In Alibaba Cloud RDS, read-only instances are synchronized from the master instance but do not allow actions such as creating or deleting databases, configuring backup policies, or changing account authorizations. However, you can configure the read-only instance’s whitelist to control access.

Alibaba Cloud Server Load Balancer (SLB) is designed to distribute incoming network traffic across multiple ECS instances within Alibaba Cloud, enhancing fault tolerance and application reliability. Backend servers for SLB must be Alibaba Cloud ECS instances, as SLB is specifically optimized to work within the Alibaba Cloud ecosystem. While it does not support instances from other cloud providers, it ensures integration and performance with Alibaba Cloud’s ECS instances.

In SLB, a Listener is configured with policies and forwarding rules that define how the load balancer will handle incoming requests, including protocol, port, and health check configurations. It does not represent an SLB instance itself or a collection of ECS instances, but it plays a crucial role in managing traffic routing and session persistence.

When forwarding layer 7 (HTTP) requests, Alibaba Cloud SLB preserves the client’s real IP address by including it in the X-Forwarded-For header. This allows the backend ECS instance to access the client’s actual IP rather than the SLB’s IP address. This practice is common for maintaining accurate client IP information in HTTP forwarding scenarios.

Alibaba Cloud Security Center is designed to enhance the security of ECS instances by providing proactive monitoring, vulnerability detection, and protection against threats such as brute-force attacks. Security Center includes features such as brute-force attack detection, which will alert you to and help mitigate potential threats to your ECS instance.

=================

When using Alibaba Cloud SLB to forward layer 7 (HTTP) service requests, SLB will replace the IP address in the HTTP header file to forward requests. Therefore, the source IP address that can be seen on the backend ECS instance is the IP address of SLB instead of the client’s real IP address. However, SLB also provides a feature called X-Forwarded-For (XFF) that can help you obtain the real IP address of the client. XFF is a standard HTTP header field that records the original IP address of the client and the proxy servers that the request passes through. You can enable XFF on the SLB console or by using the API, and then configure your backend ECS instance to parse the XFF header field and get the real IP address of the client. References: 1 - SLB overview - Alibaba Cloud Document Center - Layer 7 listeners - X-Forwarded-For

 According to the Alibaba Cloud SLB documentation1, SLB provides a health check feature that monitors the availability and performance of the backend servers. If a backend server fails the health check, SLB stops routing requests to it and redirects the traffic to other healthy servers. This way, SLB can improve the system’s availability by eliminating single points of failure (SPOFs). Therefore, option A is a correct statement about SLB.

According to the same documentation1, SLB instances are assigned with public IP addresses or private IP addresses based on the network type. If you delete an SLB instance and create a new one, a new IP address will be allocated to the new instance. You cannot specify or reserve an IP address for an SLB instance. Therefore, option B is also a correct statement about SLB.

Option C is incorrect, because SLB does not require you to install or configure any SLB agent on the backend servers. You only need to add the backend servers to the SLB instance and configure the listener and the server group. SLB will automatically distribute the traffic to the backend servers based on the load balancing algorithm and the health check status2.

Option D is also incorrect, because SLB does not provide a function to automatically synchronize data among backend servers. SLB only handles the network traffic distribution and does not interfere with the data on the backend servers. You need to use other methods or tools to synchronize data among backend servers, such as rsync, lsyncd, or Alibaba Cloud Data Transmission Service (DTS)3.

References: Server Load Balancer:SLB overview, Add backend servers, and Data Transmission Service.

 To execute a task at a specific time such as removing 1 ECS instance every night at 00:00, you need to perform two operations: create a scaling rule and create a scheduled task. A scaling rule is a set of instructions that defines how Auto Scaling scales computing resources in response to changes in business load. A scheduled task is a type of scaling task that can execute a specified scaling rule at a specified time. By creating a scaling rule that removes 1 ECS instance from the scaling group, and creating a scheduled task that executes this scaling rule every night at 00:00, you can achieve the desired result. You do not need to create a new scaling group, because you already have one scaling group created. You also do not need to create an event-triggered task, because this type of scaling task is triggered by events such as Cloud Monitor alarms or API calls, not by a specific time. References: 1, 2, and 3.

Alibaba Cloud SLB can distribute user requests to backend ECS instances, but these ECS instances do not have to have the same specification (CPU, Memory). SLB virtualizes the added group of ECS instances in the same region into an application pool, and you can manage backend servers through either the default server group, a primary server group, or VServer groups. You can add ECS instances of different specifications to the same server group or VServer group according to your needs. SLB supports weighted round-robin (WRR) and weighted least connections (WLC) algorithms to balance the load among backend servers with different weights. References:

SLB overview - Alibaba Cloud

Backend server overview - Alibaba Cloud

What is Server Load Balancer and How It Works - Alibaba Cloud

Server Load Balancer - Alibaba Cloud

Alibaba Server Load Balancer (SLB) Course - Cloud Academy

Default VPCs do count toward your overall VPC quota in Alibaba Cloud, contrary to option D. Each region has a single default VPC that is automatically created by Alibaba Cloud, and users can create additional VPCs, which are then labeled as non-default VPCs. Default VPCs also have different constraints and operational characteristics compared to user-created VPCs, such as specific configurations for ease of use in basic setups.

The DNS service provider used by the domain name (such as abc.com) does not need to support intelligent resolution, because the CNAME address allocated to the domain name will point to an IP address that is intelligently managed by Alibaba Cloud CDN system. Therefore, option C is not a correct description of modifying domain name resolution. References: What is Alibaba Cloud CDN? - CDN - Alibaba Cloud Documentation Center and Alibaba Cloud CDN

Alibaba Cloud provides two main options for securely connecting a VPC with a traditional IDC: VPN (Virtual Private Network) and a dedicated leased line connection. These methods enable stable and secure communication between the VPC and on-premises environments. HTTP agents and stepping stones (options B and C) are not standard or secure methods for IDC-VPC connectivity.

ApsaraDB for RDS is a stable and reliable online database service that supports MySQL, SQL Server, PostgreSQL, MariaDB, and PPAS. ApsaraDB for RDS provides a read/write splitting feature that allows you to create read-only RDS instances, which can offload read requests from the primary RDS instance of your database system1. Read/write splitting is implemented by using a dedicated proxy endpoint, which routes read and write requests to different RDS instances based on the SQL statements and the read weights of the read-only instances2. By using ApsaraDB for RDS to implement read/write splitting, you can achieve the following benefits21:

Read requests are distributed to multiple instances. This can improve the read performance and availability of your database system, as well as balance the load among the read-only instances. You can specify the read weights of the read-only instances to control the distribution of read requests.

Read/write requests to the database are separated. This can reduce the pressure on the primary RDS instance and ensure the stability and security of the write operations. You only need to connect to the proxy endpoint, and the proxy will automatically route the read and write requests to the appropriate RDS instances.

The other statements are not correct, as write requests are not distributed to multiple instances, but only sent to the primary RDS instance2; and both read and write requests are not distributed to multiple instances, but separated by the proxy endpoint2. References: [ApsaraDB for RDS - Alibaba Cloud], Read/write splitting - - Alibaba Cloud Documentation Center, Use the read/write splitting feature to increase the processing capability of an ApsaraDB RDS instance

Default and non-default VSwitches do not have the same constraints and operations. Default VSwitches have specific configurations and limitations set by Alibaba Cloud, designed to simplify initial setups. In contrast, non-default VSwitches, created by users, allow for more customized configurations and are typically used in more complex or specialized networking environments. Thus, option C is incorrect as the constraints and operations differ between default and non-default VSwitches.

To meet the company’s security requirements, the following solutions should be used together:

A. Enable the VPN on the bastion host (or directly use the VPN image on Alibaba Cloud Marketplace). The administrator uses VPN encrypted communication during O&M. This solution can support secure remote O&M, because VPN (Virtual Private Network) is a technology that creates a secure and encrypted connection over the Internet between the bastion host and the administrator’s device. VPN can protect the data transmitted between the bastion host and the administrator from being intercepted or tampered by malicious third parties1. Alibaba Cloud provides VPN Gateway service that allows users to create VPN connections between VPCs and on-premises data centers, or between VPCs in different regions2. Users can also use VPN images from Alibaba Cloud Marketplace, such as OpenVPN, to create VPN servers on ECS instances3.

B. Build an independent ECS instance as the bastion host or remote logon and O&M, and authorize the bastion host to access ECS instances running other subsystems. This solution can also support secure remote O&M, because a bastion host is a special-purpose ECS instance that acts as a proxy or a gateway for accessing other ECS instances in the VPC. A bastion host can enhance the security of the ECS instances by limiting the exposure of the ECS instances to the public network, and by implementing security policies and monitoring tools on the bastion host4. Alibaba Cloud provides Bastionhost service that allows users to centrally manage the access to cloud servers from external networks and provide secure connections to VPC resources5.

C. Use the security group function of the ECS instance, and respectively deploy ECS instances running different subsystems to independent security groups. This solution can isolate the networks between subsystems, because a security group is a virtual firewall that controls the inbound and outbound traffic of the ECS instances in the group. Users can configure security group rules to allow or deny access based on the network protocol, port, and source IP address. By deploying ECS instances running different subsystems to independent security groups, users can prevent unauthorized access or communication between the subsystems6.

The other solution is not suitable for the company’s scenario, for the following reason:

D. Create multiple ECS instances in the VPC to install subsystems of different departments- Allocate only Intranet IP addresses to all ECS instances, and deploy them in the same security groups. This solution cannot isolate the networks between subsystems, because ECS instances in the same security group can communicate with each other by default, regardless of whether they have intranet or internet IP addresses. Moreover, this solution may also prevent the ECS instances from accessing the internet or providing external services, which may affect the business operation of the company6.

References: What is a VPN? - Virtual Private Network - Cisco, VPN Gateway - Alibaba Cloud, OpenVPN - Alibaba Cloud Marketplace, Bastion Host - Alibaba Cloud Document Center, Bastionhost - Alibaba Cloud, Security groups - Elastic Compute Service - Alibaba Cloud