INFO2004

INFO2004 · Cloud PlatformsINFO2004 · 云平台（AWS 云计算）

Design, build and migrate real cloud infrastructure on AWS — assessed by hands-on labs and a staged Cloud Migration Plan (with an oral defence), no final exam.在 AWS 上设计、搭建并迁移真实的云基础设施 —— 由动手实验室作业 + 分阶段的「云迁移方案」（含口头答辩）评分，没有期末考。

INFO2004 (Cloud Platforms) is a Year-2 course in Adelaide University's School of Computer Science & IT. Aligned with the AWS Academy Learner Lab, it teaches cloud computing models (IaaS/PaaS/SaaS), core AWS services (EC2, S3, RDS, Lambda), VPC networking and security, the AWS Well-Architected Framework, auto scaling and the six 'R' migration strategies — culminating in a cost-optimised, scalable architecture for a real SME case. Built from 85 real INFO2004 course materials in the AskSia Library.

INFO2004（Cloud Platforms）是阿德莱德大学计算机科学与信息技术学院的本科二年级课程。课程对接 AWS Academy Learner Lab，讲授云计算模型（IaaS/PaaS/SaaS）、AWS 核心服务（EC2、S3、RDS、Lambda）、VPC 网络与安全、AWS Well-Architected 架构框架、自动扩缩容以及云迁移的「6 个 R」策略，最终为一个真实的中小企业（SME）场景设计一套成本优化、可扩展的云架构。本页内容基于AskSia Library 中 85 份真实 INFO2004 课程材料构建。

Built from 85 real INFO2004 course materials in the AskSia Library.

基于AskSia Library 中 85 份真实 INFO2004 课程材料构建。

Faculty院系School of Computer Science & ITLevel层级Undergraduate · Year 2Credit学分6 ptsSemester学期2026 Semester 1Prereq先修INFO1012 Information Technology SystemsCampus校区Adelaide, SA — Adelaide University (formed in 2026 from the University of Adelaide + UniSA merger). Confirm the specific teaching campus on the official course page.
📚 AskSia Library data·85 AskSia Library resources·12 topics·100% continuous coursework, no final exam — two lab assignments (1.1 Networking & Compute, 1.2 Database) plus a staged Cloud Migration Plan (2.1 Recommendations, 2.2 Prototype, 2.3 Viva/oral defence).Built from 85 real INFO2004 (Cloud Platforms) course materials — the course roadmap, lecture pages and AWS lab guides — in the AskSia Library. No past-paper or exam-question bank exists for this course (it has no exam), so the self-test below is practice on the real topics and the exam-style items are authored on the highest-weight concepts, not mined exam stems.
📚 AskSia Library 数据·85 份 AskSia Library 资料·12 个主题·100% 平时作业、无期末考 —— 两次实验作业（1.1 网络与计算、1.2 数据库）外加一份分阶段云迁移方案（2.1 建议书、2.2 原型、2.3 口头答辩 Viva）。基于AskSia Library 中 85 份真实 INFO2004（Cloud Platforms）课程材料构建 —— 课程 roadmap、讲义页面与 AWS 实验指南。本课程没有期末考，也没有任何 past-paper / 真题库，因此下方自测是基于真实知识点的练习题，「考试风格」题目是围绕权重最高的概念自编的，而非挖到的真题。
Overview课程概览

What INFO2004 is aboutINFO2004 讲什么

INFO2004 (Cloud Platforms) is a 6-credit-point Year-2 course at Adelaide University, taught in the School of Computer Science & IT and delivered through online learning plus weekly on-campus seminars and AWS practical sessions. It gives students the knowledge and skills to design, implement and manage cloud-based IT infrastructures for small-to-medium enterprises, focused on Amazon Web Services (AWS) and aligned with the AWS Academy Learner Lab. Across a 10-week structure (three modules: Foundations of Cloud Computing; Cloud Security and Networking; Architecture and Migration), students explore the five essential characteristics of cloud computing, the service and deployment models, cloud economics and cost optimisation, core AWS services (EC2, ECS, Lambda, S3, EBS, EFS, RDS, DynamoDB), Virtual Private Cloud (VPC) design with security groups and network ACLs, the shared responsibility model, the AWS Well-Architected Framework, auto scaling and CloudWatch monitoring, and the six 'R' migration strategies — culminating in a staged Cloud Migration Plan that students implement and defend orally. The prerequisite is INFO1012 Information Technology Systems.

INFO2004（Cloud Platforms / 云平台）是阿德莱德大学计算机科学与信息技术学院开设的 6 学分本科二年级课程，采用线上学习配合每周线下研讨课（seminar）与 AWS 实操课（practical）的方式授课。课程教授为中小企业设计、实现并管理云端 IT 基础设施的知识与技能，重点放在 Amazon Web Services（AWS），并对接 AWS Academy Learner Lab。课程为 10 周结构、三个模块（云计算基础；云安全与网络；架构与迁移），学生会学习云计算的五大基本特征、服务与部署模型、云经济学与成本优化、AWS 核心服务（EC2、ECS、Lambda、S3、EBS、EFS、RDS、DynamoDB）、虚拟私有云（VPC）设计及其安全组与网络 ACL、责任共担模型、AWS Well-Architected 架构框架、自动扩缩容与 CloudWatch 监控，以及云迁移的「6 个 R」策略，最终落地为一份需要动手实现并进行口头答辩的分阶段云迁移方案。先修要求为 INFO1012 Information Technology Systems。

Topic map知识地图

The INFO2004 syllabus, topic by topicINFO2004 大纲 · 逐个主题

1

Foundations: cloud models & on-prem vs cloud基础：云模型与本地 vs 云

What 'the Cloud' means (Gartner/NIST), the five essential characteristics (on-demand self-service, broad network access, resource pooling, rapid elasticity, measured service), and how cloud compares with on-premises infrastructure.

「云」到底指什么（Gartner / NIST 定义）、云的五大基本特征（按需自助、广泛网络访问、资源池化、快速弹性、可计量服务），以及云与本地（on-premises）基础设施的对比。

2

Service & deployment models (IaaS / PaaS / SaaS)服务与部署模型（IaaS / PaaS / SaaS）

The three service models and how responsibility shifts between customer and provider, plus the four deployment models (private, community, public, hybrid) and why public cloud dominates.

三种服务模型，以及客户与云服务商之间责任如何随模型变化；外加四种部署模型（私有云、社区云、公有云、混合云），以及为何公有云占主导。

3

Cloud economics & cost optimisation云经济学与成本优化

The cost of IT systems, pricing/TCO and tools for managing cloud spend; EC2 purchasing options (On-Demand, Reserved, Spot, Savings Plans) and how to justify a cost-optimised architecture rather than over-provisioning.

IT 系统的成本、定价 / 总拥有成本（TCO）与云支出管理工具；EC2 的购买选项（按需、预留、Spot、Savings Plans），以及如何论证一套成本优化的架构、而非资源过度配置。

4

Cloud architecture & the Well-Architected Framework云架构与 Well-Architected 框架

The AWS Well-Architected Framework's five pillars (CORPS: Cost optimisation, Operational excellence, Reliability, Performance efficiency, Security) and reviewing architectures as lightweight, blame-free conversations.

AWS Well-Architected 框架的五大支柱（CORPS：成本优化、卓越运营、可靠性、性能效率、安全），以及把架构评审当作轻量、不追责的「对话」而非审计。

5

Cloud security & the shared responsibility model云安全与责任共担模型

Security 'of' the cloud vs 'in' the cloud; inherited, shared and customer-specific controls; IAM fundamentals (users, groups, policies, MFA) and aspects of cloud security and compliance.

云的安全 vs 云中的安全（security of vs in the cloud）；继承型、共享型与客户专属型控制；IAM 基础（用户、用户组、策略、MFA），以及云安全与合规的各个方面。

6

Cloud networking & Virtual Private Cloud (VPC)云网络与虚拟私有云（VPC）

Designing a VPC: subnets, route tables, internet gateways, VPC endpoints and CIDR blocks; securing it with security groups (stateful, instance-level) vs network ACLs (stateless, subnet-level).

设计 VPC：子网、路由表、互联网网关、VPC 端点与 CIDR 块；用安全组（有状态、实例级）vs 网络 ACL（无状态、子网级）为其加固。

7

Cloud compute options (EC2, ECS, Lambda)云计算服务（EC2、ECS、Lambda）

Virtualization, bare-metal, VMs, containers and microservices; the three AWS compute services — EC2 (with AMIs and instance types), ECS/Fargate containers and serverless Lambda — plus Elastic Beanstalk, and how to choose between them.

虚拟化、裸金属、虚拟机、容器与微服务；AWS 的三大计算服务 —— EC2（含 AMI 与实例类型）、ECS / Fargate 容器、无服务器 Lambda —— 以及 Elastic Beanstalk，并学会如何在它们之间取舍。

8

Cloud storage options (object, file, block)云存储选项（对象 / 文件 / 块）

Local vs network storage and the three storage types on AWS — object (S3), file (EFS) and block (EBS) — with data protection: versioning, lifecycle policies and encryption; CDNs for content delivery.

本地存储 vs 网络存储，以及 AWS 上的三种存储类型 —— 对象（S3）、文件（EFS）、块（EBS）—— 并配合数据保护：版本控制、生命周期策略与加密；以及用于内容分发的 CDN。

9

Cloud database options (RDS & DynamoDB)云数据库选项（RDS 与 DynamoDB）

The purpose of a database and the main database types (relational vs non-relational); managed options on AWS — RDS for relational workloads and DynamoDB for NoSQL — and how to choose for a given workload.

数据库的作用与主要类型（关系型 vs 非关系型）；AWS 上的托管选项 —— 面向关系型工作负载的 RDS 与面向 NoSQL 的 DynamoDB —— 以及如何针对特定工作负载选型。

10

Auto scaling & monitoring自动扩缩容与监控

Dealing with unexpected demand via load balancing then auto scaling; health checks; collecting metrics and resolution; CloudWatch alarms to run 'just enough' servers while absorbing spikes.

用负载均衡再到自动扩缩容来应对突发需求；健康检查；指标采集与分辨率；用 CloudWatch 告警实现「刚好够用」的服务器数量，同时扛住流量峰值。

11

Cloud migration strategies (the six 'R's)云迁移策略（6 个 R）

The six application migration strategies — Rehost (lift & shift), Replatform (lift, tinker & shift), Refactor/Re-architect, Repurchase, Retire, Retain — and choosing by effort vs opportunity to optimise, with a supporting business case.

六种应用迁移策略 —— Rehost（直接搬迁）、Replatform（小改后搬迁）、Refactor / Re-architect、Repurchase、Retire、Retain —— 并按「投入 vs 优化空间」结合业务论证来选型。

12

What's next: careers & certification, future cloud下一步：职业与认证、云的未来

Careers and certification pathways (AWS, Azure, Google Cloud) and emerging directions such as sky computing (the move beyond a single cloud provider).

云计算的职业与认证路径（AWS、Azure、Google Cloud），以及新兴方向，例如 sky computing（超越单一云服务商的趋势）。

Assessment考核方式

How INFO2004 is assessedINFO2004 怎么考核

Final exam: No期末考试：无
Component考核项 Weight占比 Note说明
Assignment 1.1 — Networking & Compute Lab (VPC, EC2, EBS)作业 1.1 —— 网络与计算实验（VPC、EC2、EBS） see course outline Hands-on AWS Academy lab: build and secure a VPC, launch EC2 instances and attach EBS storage. Due Week 8.AWS Academy 动手实验：搭建并加固一个 VPC、启动 EC2 实例并挂载 EBS 存储。第 8 周截止。
Assignment 1.2 — Database Lab (RDS & App Integration)作业 1.2 —— 数据库实验（RDS 与应用集成） see course outline Hands-on lab: build a database server with AWS RDS and integrate it with an application. Due Week 10.动手实验：用 AWS RDS 搭建数据库服务器并与应用集成。第 10 周截止。
Assignment 2.1 — Cloud Migration Plan: Recommendations作业 2.1 —— 云迁移方案：建议书 see course outline Recommend a migration approach for an SME case, justified with the 6 R's and cost optimisation. Due Week 4.为一个中小企业（SME）案例推荐迁移方案，用「6 个 R」与成本优化加以论证。第 4 周截止。
Assignment 2.2 — Cloud Migration Plan: Prototype Implementation作业 2.2 —— 云迁移方案：原型实现 see course outline Implement a working prototype of the proposed cost-optimised, scalable architecture on AWS. Due Week 12.在 AWS 上实现所提出的成本优化、可扩展架构的可运行原型。第 12 周截止。
Assignment 2.3 — Cloud Migration Plan: Viva / Oral Defence作业 2.3 —— 云迁移方案：口头答辩（Viva） see course outline Oral defence of your migration plan and prototype — explain and justify your design decisions. Held Week 13.对你的迁移方案与原型进行口头答辩 —— 解释并论证你的设计决策。第 13 周进行。

100% continuous coursework — no final exam. Assessment runs as two strands: Assignment 1 (hands-on AWS Academy labs — 1.1 Networking & Compute with VPC/EC2/EBS, and 1.2 a Database lab with RDS) and Assignment 2 (a staged Cloud Migration Plan for an SME case — 2.1 Recommendations, 2.2 Prototype Implementation, 2.3 a Viva / oral defence). Exact component weights are set in your course outline each semester — the course roadmap lists the structure and due weeks but not fixed percentages, so confirm weights against your own outline.

100% 平时作业，无期末考。考核分两条主线：作业 1（AWS Academy 动手实验 —— 1.1 网络与计算，含 VPC/EC2/EBS；1.2 数据库实验，含 RDS）与作业 2（针对中小企业案例的分阶段云迁移方案 —— 2.1 建议书、2.2 原型实现、2.3 口头答辩 Viva）。各部分的具体权重由每学期的 course outline 规定 —— 课程 roadmap 给出了结构与截止周次但未给固定百分比，请以你自己的 outline 为准核对权重。

Assessment timeline考核时间线

When each INFO2004 task is dueINFO2004 各项考核时间

Assignment 2.1 — Cloud Migration Plan: Recommendations作业 2.1 —— 云迁移方案：建议书
Week 4 (due 20 Mar 2026)第 4 周（2026-03-20 截止）
see course outline
Assignment 1.1 — Networking & Compute Lab (VPC, EC2, EBS)作业 1.1 —— 网络与计算实验（VPC、EC2、EBS）
Week 8 (due 17 Apr 2026)第 8 周（2026-04-17 截止）
see course outline
Assignment 1.2 — Database Lab (RDS & App Integration)作业 1.2 —— 数据库实验（RDS 与应用集成）
Week 10 (due 8 May 2026)第 10 周（2026-05-08 截止）
see course outline
Assignment 2.2 — Cloud Migration Plan: Prototype Implementation作业 2.2 —— 云迁移方案：原型实现
Week 12 (due 22 May 2026)第 12 周（2026-05-22 截止）
see course outline
Assignment 2.3 — Cloud Migration Plan: Viva / Oral Defence作业 2.3 —— 云迁移方案：口头答辩（Viva）
Week 13 (course conclusion)第 13 周（课程收尾）
see course outline
Self-test自测练习

Test yourself: INFO2004 practice questions自测一下：INFO2004 练习题

Question 1第 1 题
Under the AWS shared responsibility model, which of the following is the customer's responsibility (security 'in' the cloud) rather than AWS's (security 'of' the cloud)?在 AWS 责任共担模型下，以下哪一项属于客户的责任（云「中」的安全），而不属于 AWS（云「的」安全）？
  1. Physical security of the data-centre facilities
  2. Patching the host hypervisor and virtualization layer
  3. Encrypting your own data and configuring your guest OS, applications and IAM access
  4. Maintaining the underlying networking hardware AWS operates
  1. 数据中心机房的物理安全
  2. 修补宿主机 hypervisor 与虚拟化层
  3. 加密你自己的数据，并配置你的客户机操作系统、应用与 IAM 访问
  4. 维护 AWS 运营的底层网络硬件
Show answer查看答案
Answer: C. Encrypting your own data and configuring your guest OS, applications and IAM accessAWS handles security 'of' the cloud — physical facilities, hardware, the hypervisor and core infrastructure (options A, B, D). The customer handles security 'in' the cloud: their data, encryption, guest OS, applications and access (IAM). Some controls are inherited, some shared (e.g. patch management — AWS patches the infrastructure, you patch your guest OS), and some are customer-specific.
答案：C. 加密你自己的数据，并配置你的客户机操作系统、应用与 IAM 访问AWS 负责云「的」安全 —— 物理设施、硬件、hypervisor 与核心基础设施（A、B、D）。客户负责云「中」的安全：自己的数据、加密、客户机操作系统、应用与访问（IAM）。部分控制是继承型，部分是共享型（如补丁管理 —— AWS 修补基础设施，你修补客户机 OS），部分是客户专属型。
Question 2第 2 题
An SME's data-centre lease ends in 12 months and will not be renewed; management wants to be off-premises before then at low risk, while keeping the door open to modernise later. Per the six 'R' migration strategies, which approach best fits the immediate move?某中小企业的数据中心租约 12 个月后到期且不再续约；管理层希望在到期前低风险地完成下云迁移，同时为日后现代化改造留余地。按「6 个 R」迁移策略，眼下这一步最适合哪种方式？
  1. Refactor / Re-architect — rebuild the app cloud-native immediately
  2. Rehost ('lift and shift') the app onto EC2 now, and Replatform/Refactor later once stable
  3. Repurchase — replace the bespoke app with an off-the-shelf SaaS product
  4. Retire — decommission the application entirely
  1. Refactor / Re-architect —— 立即把应用重建为云原生
  2. 现在先 Rehost（直接搬迁）到 EC2，待稳定后再 Replatform / Refactor
  3. Repurchase —— 用现成的 SaaS 产品替换这套定制应用
  4. Retire —— 直接下线该应用
Show answer查看答案
Answer: B. Rehost ('lift and shift') the app onto EC2 now, and Replatform/Refactor later once stableThe hard constraint (off-premises within 12 months, low risk) points to Rehost — the lowest-effort 'lift and shift' that meets the deadline fastest. A full Refactor is highest effort/risk and not justified now; Repurchase doesn't fit a bespoke app; Retire would kill a live system. The course's own example sequences Rehost first to clear the deadline, then Replatform/Refactor later to capture optimisation — exactly the effort-vs-opportunity trade-off the 6 R's diagram captures.
答案：B. 现在先 Rehost（直接搬迁）到 EC2，待稳定后再 Replatform / Refactor硬性约束（12 个月内下云、风险要低）指向 Rehost —— 投入最低的「直接搬迁」，最快赶上截止期。完整 Refactor 投入与风险最高、现在并不划算；Repurchase 不适合定制应用；Retire 会直接关停在用系统。课程自带的示例就是先 Rehost 赶截止期、待稳定后再 Replatform / Refactor 拿优化收益 —— 正是 6 个 R 图所刻画的「投入 vs 优化空间」权衡。
Question 3第 3 题
Within an AWS VPC, what is the key difference between a security group and a network ACL?在 AWS VPC 内，安全组（security group）与网络 ACL（network ACL）的关键区别是什么？
  1. A security group is stateless and subnet-level; a network ACL is stateful and instance-level
  2. A security group is stateful and acts at the instance level (allow rules only); a network ACL is stateless and acts at the subnet level (allow and deny rules)
  3. Both are stateful instance-level firewalls; they differ only in price
  4. A security group routes traffic between subnets; a network ACL assigns CIDR blocks
  1. 安全组是无状态、子网级的；网络 ACL 是有状态、实例级的
  2. 安全组是有状态的、作用于实例级（只支持允许规则）；网络 ACL 是无状态的、作用于子网级（支持允许与拒绝规则）
  3. 两者都是有状态的实例级防火墙，只是价格不同
  4. 安全组在子网间路由流量；网络 ACL 分配 CIDR 块
Show answer查看答案
Answer: B. A security group is stateful and acts at the instance level (allow rules only); a network ACL is stateless and acts at the subnet level (allow and deny rules)A security group is stateful (return traffic is automatically allowed) and operates at the instance/ENI level with allow rules only. A network ACL is stateless (you must allow both directions) and operates at the subnet level, supporting both allow and deny rules. Routing between subnets is the job of route tables, not either firewall — so the other options confuse the two layers.
答案：B. 安全组是有状态的、作用于实例级（只支持允许规则）；网络 ACL 是无状态的、作用于子网级（支持允许与拒绝规则）安全组是有状态的（回程流量自动放行），作用于实例 / ENI 级，且只支持允许规则。网络 ACL 是无状态的（两个方向都要显式放行），作用于子网级，支持允许与拒绝规则。子网间路由是路由表的职责，与这两种防火墙无关 —— 其他选项把两层混淆了。
Assessment-style questions考核题型

Key assessment-style questions in INFO2004INFO2004 核心考核风格题

Shared responsibility model责任共担模型
Given a list of security tasks (physical security, hypervisor patching, data encryption, guest-OS patching, IAM config), classify each as AWS-responsible ('of' the cloud) vs customer-responsible ('in' the cloud), and name which are inherited, shared or customer-specific controls.给定一组安全任务（物理安全、hypervisor 补丁、数据加密、客户机 OS 补丁、IAM 配置），把每一项归类为 AWS 负责（云「的」安全）还是客户负责（云「中」的安全），并指出哪些是继承型、共享型或客户专属型控制。
Exam-style item on a core, recurring course topic (Module 2). Authored from the shared-responsibility lecture and its two self-test activities — not a mined exam stem.
Six 'R' migration strategies迁移的「6 个 R」策略
Given an organisation's drivers and constraints (e.g. a lease deadline, cost avoidance, a bespoke vs commodity app), recommend a migration strategy (or sequence of strategies) from the six R's and justify it by effort vs opportunity to optimise.给定某组织的驱动与约束（如租约截止期、规避成本、定制 vs 通用应用），从 6 个 R 中推荐一种迁移策略（或策略序列），并按「投入 vs 优化空间」加以论证。
Exam-style item mirroring the course's own worked example and learning activity 3.3 — author-written, no real exam stem exists for this course.
Well-Architected Framework (CORPS)Well-Architected 框架（CORPS）
Name the five pillars of the AWS Well-Architected Framework (Cost optimisation, Operational excellence, Reliability, Performance efficiency, Security) and explain which pillar a given design decision (e.g. adding an Auto Scaling group, choosing Spot instances) most improves.说出 AWS Well-Architected 框架的五大支柱（成本优化、卓越运营、可靠性、性能效率、安全），并解释某个设计决策（如加一个 Auto Scaling 组、选用 Spot 实例）最能改善哪个支柱。
Exam-style item on a high-weight framework taught in Module 1. Authored from the Well-Architected lecture, not mined.
VPC design & security (security group vs network ACL)VPC 设计与安全（安全组 vs 网络 ACL）
Describe the components of a VPC (subnets, route tables, internet gateway, VPC endpoint, CIDR block) and contrast a security group (stateful, instance-level) with a network ACL (stateless, subnet-level), choosing the right control for a stated scenario.描述 VPC 的组成（子网、路由表、互联网网关、VPC 端点、CIDR 块），对比安全组（有状态、实例级）与网络 ACL（无状态、子网级），并为给定场景选择合适的控制。
Exam-style item on a core Module 2 topic that also underpins Assignment 1.1. Author-written from the VPC lecture.
Cloud compute selection (EC2 / ECS / Lambda)云计算选型（EC2 / ECS / Lambda）
For a given workload, choose between EC2 (VMs), ECS/Fargate (containers) and Lambda (serverless) and justify the choice by abstraction level, scaling behaviour and cost — and recommend an EC2 purchasing option (On-Demand, Reserved, Spot, Savings Plans).针对给定工作负载，在 EC2（虚拟机）、ECS / Fargate（容器）与 Lambda（无服务器）之间选型，并按抽象层级、扩缩容行为与成本加以论证 —— 同时推荐一种 EC2 购买选项（按需、预留、Spot、Savings Plans）。
Exam-style item on Module 2's compute section. Authored from the AWS Compute lecture and the purchasing-options activity, not mined.
Cloud storage selection (object / file / block)云存储选型（对象 / 文件 / 块）
Match a use case (database volume, shared enterprise file drive, static-website assets / data lake) to the right AWS storage type — EBS (block), EFS (file) or S3 (object) — using latency, cost and access-pattern trade-offs, and name an S3 data-protection feature (versioning, lifecycle, encryption).把使用场景（数据库卷、企业共享文件盘、静态网站资源 / 数据湖）匹配到合适的 AWS 存储类型 —— EBS（块）、EFS（文件）或 S3（对象）—— 依据延迟、成本与访问模式权衡，并说出一项 S3 数据保护特性（版本控制、生命周期、加密）。
Exam-style item on Module 3's storage section, aligned with the storage-comparison lecture. Author-written.
Cloud database selection (RDS vs DynamoDB)云数据库选型（RDS vs DynamoDB）
Distinguish relational from non-relational databases (and the ACID properties of relational DBs), then choose between managed RDS (relational) and DynamoDB (NoSQL) for a stated workload's data model and scale.区分关系型与非关系型数据库（以及关系型数据库的 ACID 特性），再针对给定工作负载的数据模型与规模，在托管 RDS（关系型）与 DynamoDB（NoSQL）之间选型。
Exam-style item on Module 3's database section, aligned with Assignment 1.2. Authored from the database-types lecture, not mined.
Auto scaling & monitoring (load balancing, CloudWatch)自动扩缩容与监控（负载均衡、CloudWatch）
For a spiky-demand scenario (e.g. a web server sized for 50 users hit by 100), explain how load balancing then auto scaling absorb the spike, what role health checks and CloudWatch alarms play, and how this runs 'just enough' servers cost-effectively.针对突发需求场景（如按 50 用户配置的 Web 服务器突然涌入 100 人），解释负载均衡再到自动扩缩容如何扛住峰值，健康检查与 CloudWatch 告警各起什么作用，以及如何以「刚好够用」的服务器数量做到成本可控。
Exam-style item mirroring the course's own 'unexpected demand' scenario in the auto-scaling lecture. Author-written.
Cloud economics & cost optimisation云经济学与成本优化
Compare the cost of an on-premises setup with a cloud TCO, and justify a cost-optimised architecture (right-sizing, appropriate EC2 purchasing options, avoiding over-provisioning) rather than lift-and-shift over-provisioning.对比本地部署成本与云的总拥有成本（TCO），并论证一套成本优化的架构（合理规格、合适的 EC2 购买选项、避免过度配置），而非直接搬迁式的过度配置。
Exam-style item on Module 1's economics section and the cost-of-IT lecture. Author-written, no mined exam stem.
Cloud service & deployment models (IaaS/PaaS/SaaS)云服务与部署模型（IaaS/PaaS/SaaS）
Given a scenario, identify the service model (IaaS, PaaS or SaaS) and state how responsibility shifts between customer and provider across the stack; then pick the deployment model (private, community, public, hybrid) that fits the requirements.给定场景，判断其服务模型（IaaS、PaaS 或 SaaS），并说明客户与服务商之间的责任如何随技术栈层级转移；再选出符合需求的部署模型（私有云、社区云、公有云、混合云）。
Exam-style item on a foundational Module 1 topic (the five characteristics and service/deployment models). Author-written from the cloud-basics and types-of-cloud lectures.
Worked example例题精解

A worked INFO2004 problemINFO2004 例题

Choosing a migration strategy with the six 'R's用「6 个 R」选择迁移策略

Problem题目

An SME runs a legacy line-of-business web app on two ageing on-premises servers; its data-centre lease ends in 12 months and it will not renew. Management wants to be off-premises before the lease expires, at low risk, while keeping the door open to modernise later. Which of the six 'R' migration strategies fits, and how would you sequence the move?

某中小企业（SME）在两台老旧的本地服务器上运行一套遗留的业务 Web 应用；其数据中心租约将在 12 个月后到期且不再续约。管理层希望在租约到期前完成下云迁移、风险要低，同时为日后的现代化改造留出余地。在「6 个 R」中应选哪种迁移策略？又该如何安排迁移步骤？

Approach解题思路

Map the drivers to the 6 R's by effort vs opportunity to optimise. The hard constraint (off-premises within 12 months, low risk) points to Rehost — a 'lift and shift' that moves the app as-is onto EC2 with the least effort and change, so it is the fastest way to meet the deadline. The 'modernise later' goal does not justify a full Refactor/Re-architect now (highest effort/risk), nor Repurchase (swap to a SaaS product) since the app is bespoke. The clean two-step plan: (1) Rehost first to clear the data-centre deadline — provision a VPC, launch EC2 instances from an AMI, attach EBS, move the database to RDS; then (2) Replatform or Refactor later once stable — e.g. add an Auto Scaling group behind a load balancer and offload static assets to S3 to improve the Well-Architected pillars (cost optimisation, reliability, performance). This sequencing keeps near-term risk low while leaving optimisation opportunity open, exactly the trade-off the 6 R's diagram captures.

把业务驱动按「投入 vs 优化空间」映射到 6 个 R 上。硬性约束（12 个月内下云、风险要低）指向 Rehost —— 即「直接搬迁（lift and shift）」，把应用原样搬到 EC2，改动与投入最小，因而是赶上截止期最快的方式。「日后再现代化」的目标并不足以现在就做完整的 Refactor / Re-architect（投入与风险最高），也不适合 Repurchase（换成某款 SaaS 产品），因为这是一套定制应用。清晰的两步走方案：（1）先 Rehost 以赶上数据中心截止期 —— 开通 VPC、用 AMI 启动 EC2 实例、挂载 EBS、把数据库迁到 RDS；（2）待系统稳定后再 Replatform 或 Refactor —— 例如在负载均衡器后加一个 Auto Scaling 组、把静态资源卸载到 S3，以改善 Well-Architected 各支柱（成本优化、可靠性、性能）。这种排序在近期把风险压低，同时保留优化空间，正是 6 个 R 复杂度图所刻画的权衡。

Key terms核心术语

INFO2004 glossaryINFO2004 术语表

AWS (Amazon Web Services)亚马逊云服务（AWS）
Amazon's cloud platform; the primary provider this course is built around, accessed via the AWS Academy Learner Lab.
亚马逊的云平台，是本课程的核心实操平台，通过 AWS Academy Learner Lab 访问。
Five essential characteristics云的五大基本特征
NIST's defining cloud traits: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service.
NIST 定义云的五个特征：按需自助、广泛网络访问、资源池化、快速弹性、可计量服务。
IaaS / PaaS / SaaS基础设施/平台/软件即服务
The three core service models, differing by how many layers (from hardware up to the app) the provider manages for you.
三种核心服务模型，区别在于云服务商替你管理的层级（从硬件到应用）有多少。
On-premises本地部署（on-premises）
Traditional infrastructure run on hardware you own and operate end-to-end, contrasted with cloud.
在自有硬件上端到端自行运维的传统基础设施，与云相对。
EC2 / AMIEC2 / AMI（机器映像）
Elastic Compute Cloud — on-demand virtual servers, launched from an Amazon Machine Image (AMI) template that fixes the instance's configuration.
弹性计算云 —— 按需虚拟服务器，由 Amazon Machine Image（AMI）模板启动，模板固定了实例的配置。
VPC (Virtual Private Cloud)虚拟私有云（VPC）
A logically isolated virtual network dedicated to your AWS account, made of subnets, route tables, internet gateways and CIDR blocks.
专属于你 AWS 账户、逻辑隔离的虚拟网络，由子网、路由表、互联网网关与 CIDR 块构成。
Security group vs network ACL安全组 vs 网络 ACL
Two VPC firewalls: a security group is stateful and acts at the instance level (allow rules only); a network ACL is stateless and acts at the subnet level (allow and deny rules).
VPC 的两种防火墙：安全组是有状态的、作用于实例级（只支持允许规则）；网络 ACL 是无状态的、作用于子网级（支持允许与拒绝规则）。
Shared responsibility model责任共担模型
AWS handles security 'of' the cloud (infrastructure); the customer handles security 'in' the cloud (data, OS, apps, access). Controls can be inherited, shared or customer-specific.
AWS 负责「云的安全」（基础设施）；客户负责「云中的安全」（数据、操作系统、应用、访问）。控制分为继承型、共享型与客户专属型。
Well-Architected Framework (CORPS)Well-Architected 框架（CORPS）
AWS's five pillars of architecture excellence: Cost optimisation, Operational excellence, Reliability, Performance efficiency and Security.
AWS 架构卓越的五大支柱：成本优化、卓越运营、可靠性、性能效率、安全。
Lambda (serverless)Lambda（无服务器）
An event-triggered compute service that runs code without provisioning servers; consumes no resources except while monitoring its event source.
事件触发的计算服务，无需预置服务器即可运行代码；除监听事件源外不消耗资源。
Object / file / block storage对象 / 文件 / 块存储
The three AWS storage types — S3 (object), EFS (file) and EBS (block) — chosen by access pattern, with versioning, lifecycle and encryption for data protection.
AWS 的三种存储类型 —— S3（对象）、EFS（文件）、EBS（块）—— 按访问模式选型，并用版本控制、生命周期与加密做数据保护。
RDS / DynamoDBRDS / DynamoDB
Managed AWS databases: RDS for relational workloads, DynamoDB for NoSQL — chosen to fit the workload's data model and scale.
AWS 托管数据库：RDS 面向关系型工作负载，DynamoDB 面向 NoSQL —— 按工作负载的数据模型与规模选型。
Auto scaling & CloudWatch自动扩缩容与 CloudWatch
Auto scaling dynamically adds/removes servers based on load (e.g. active users); CloudWatch collects metrics and fires alarms, with health checks deciding what 'healthy' means.
自动扩缩容根据负载（如活跃用户数）动态增减服务器；CloudWatch 采集指标并触发告警，由健康检查界定何为「健康」。
The six 'R' migration strategies迁移的「6 个 R」策略
Rehost, Replatform, Refactor/Re-architect, Repurchase, Retire and Retain — ordered by effort vs opportunity to optimise when moving an app to the cloud.
Rehost、Replatform、Refactor/Re-architect、Repurchase、Retire、Retain —— 把应用迁上云时，按「投入 vs 优化空间」排序的六种策略。
Cost optimisation成本优化
Running systems to deliver business value at the lowest sustainable cost — e.g. choosing EC2 purchasing options and avoiding over-provisioning.
在交付业务价值的前提下，以可持续的最低成本运行系统 —— 例如选择合适的 EC2 购买选项、避免资源过度配置。
FAQ

INFO2004 — common questionsINFO2004 常见问题

How is INFO2004 assessed?INFO2004 怎么考核？
It's 100% continuous coursework with no final exam. There are two strands: Assignment 1 is hands-on AWS Academy labs (1.1 Networking & Compute with VPC/EC2/EBS, due Week 8; 1.2 a Database lab with RDS, due Week 10), and Assignment 2 is a staged Cloud Migration Plan for an SME case — 2.1 Recommendations (Week 4), 2.2 Prototype Implementation (Week 12), and 2.3 a Viva / oral defence (Week 13). The course roadmap lists this structure and the due weeks but not fixed component percentages, so confirm exact weights against your own course outline once enrolled.
100% 平时作业，没有期末考。分两条主线：作业 1 是 AWS Academy 动手实验（1.1 网络与计算，含 VPC/EC2/EBS，第 8 周截止；1.2 数据库实验，含 RDS，第 10 周截止）；作业 2 是针对中小企业（SME）案例的分阶段云迁移方案 —— 2.1 建议书（第 4 周）、2.2 原型实现（第 12 周）、2.3 口头答辩 Viva（第 13 周）。课程 roadmap 给出了这一结构与截止周次，但未给出固定的各部分百分比，入学后请以你自己的 course outline 为准核对具体权重。
What do I actually learn in INFO2004?INFO2004 到底学什么？
Cloud computing models (the five characteristics; IaaS/PaaS/SaaS; public/private/hybrid), cloud economics and cost optimisation, the AWS Well-Architected Framework (CORPS), cloud security and the shared responsibility model, VPC networking with security groups and network ACLs, AWS compute (EC2, ECS, Lambda), storage (S3, EBS, EFS) and databases (RDS, DynamoDB), auto scaling with CloudWatch, and the six 'R' migration strategies — all hands-on in the AWS Academy Learner Lab.
云计算模型（五大特征；IaaS/PaaS/SaaS；公有/私有/混合云）、云经济学与成本优化、AWS Well-Architected 框架（CORPS）、云安全与责任共担模型、VPC 网络及其安全组与网络 ACL、AWS 计算（EC2、ECS、Lambda）、存储（S3、EBS、EFS）与数据库（RDS、DynamoDB）、配合 CloudWatch 的自动扩缩容，以及云迁移的「6 个 R」策略 —— 全部在 AWS Academy Learner Lab 中动手实践。
What's the prerequisite for INFO2004?INFO2004 的先修课是什么？
You must have completed INFO1012 Information Technology Systems before enrolling. The course also expects about 15 hours of study per week (roughly 8–10 hours of content plus seminars, practicals and assessment prep).
选课前必须已完成 INFO1012 Information Technology Systems（信息技术系统）。课程还预期每周约 15 小时的学习量（约 8–10 小时内容，外加研讨课、实操课与作业准备）。
Is INFO2004 the same course as INFO2028 / INFO2041?INFO2004 和 INFO2028 / INFO2041 是同一门课吗？
They're closely related Cloud Platforms offerings at Adelaide University. INFO2004 is the in-person/standard offering; INFO2028 is the online ('UO') version and INFO2041 is a UniSA-aligned version. The content overlaps heavily, but always check your own course code's outline for the exact assessment and schedule.
它们是阿德莱德大学密切相关的 Cloud Platforms 课程版本。INFO2004 是线下/标准版；INFO2028 是在线（UO）版，INFO2041 是 UniSA 对应版。内容高度重叠，但请始终以你自己课程代码的 outline 为准核对具体考核与日程。
Can I use AskSia for INFO2004?INFO2004 能用 AskSia 吗？
Yes — Sia helps you understand cloud concepts and work through AWS design and migration problems step by step. Submitting Sia-generated work as your own is academic misconduct under Adelaide University policy; use it as a tutor, not a substitute.
可以 —— Sia 帮你理解云概念、逐步推演 AWS 设计与迁移问题。把 Sia 生成的内容当作自己的作业提交，在阿德莱德大学政策下属于学术不端；把它当作 tutor，而不是替代。
More at Adelaide更多 阿大 课程

Other Adelaide course guides阿大 其他课程指南

ACCT2007 at Adelaide· 阿大 ACCT2013 at Adelaide· 阿大 BIOL1004 at Adelaide· 阿大
All Adelaide courses全部 阿大 课程 Browse all AU courses浏览全部澳洲课程

AskSia is an independent study aid and is not affiliated with, endorsed by, or sponsored by Adelaide University. Course details may change — always confirm against the official handbook. Read about how this guide is built. AskSia 是独立的学习辅助工具，与阿德莱德大学没有任何隶属、背书或赞助关系。课程信息可能变动，请始终以官方 handbook 为准。了解本指南的编写方法。