IB OCR Computer Science: High-Frequency Exam Topics Summary | IB OCR 计算机科学：高频考点总结

📚 IB OCR Computer Science: High-Frequency Exam Topics Summary | IB OCR 计算机科学：高频考点总结

This revision guide distils the most frequently examined concepts from both IB Diploma Computer Science (SL/HL) and OCR A-Level Computer Science (H046/H446). By focusing on the overlapping core – computational thinking, programming, data structures, algorithms, system architecture, networking, and societal impacts – it provides a streamlined review to boost your exam confidence. Each point is presented bilingually to support learners studying the dual curricula.

本复习指南提炼了IB文凭计算机科学（SL/HL）和OCR A-Level计算机科学（H046/H446）中最常考查的概念。通过聚焦重叠的核心内容——计算思维、编程、数据结构、算法、系统架构、网络及社会影响，它提供了一个精简的复习路径以增强你的考试信心。每个要点以双语呈现，支持双课程学习者的备考需求。

1. Computational Thinking and Problem Solving | 计算思维与问题解决

Computational thinking involves decomposition, pattern recognition, abstraction, and algorithm design. In IB Paper 1 and OCR Component 01, you are expected to break a complex problem into manageable parts, identify recurring patterns, and strip away unnecessary details to create a generalised solution.

计算思维包含分解、模式识别、抽象和算法设计。在IB试卷一和OCR单元01中，你需要将复杂问题拆分为可管理的部分，识别重复出现的模式，并去除不必要的细节以构建通用的解决方案。

Abstraction means representing complex reality by a simplified model. For example, a map uses symbols and colours instead of showing every physical detail. This concept is often tested through scenario-based questions in both specifications.

抽象意味着用简化的模型表示复杂的现实。例如，地图使用符号和颜色而不是显示每个物理细节。在两个大纲中这一概念常通过情景题来考查。

Decomposition: Splitting a system into smaller, self-contained modules, such as breaking a payroll program into input, calculation, and output stages.

分解：将系统分割为更小、自包含的模块，比如将工资程序划分为输入、计算和输出阶段。
Pattern recognition: Identifying similarities among problems, which can lead to efficient use of existing solutions like standard sorting algorithms.

模式识别：找出问题之间的相似性，这可以导向对现有解决方案（如标准排序算法）的有效利用。

2. Programming Fundamentals and Control Structures | 编程基础与控制结构

Both IB and OCR require mastery of sequence, selection, and iteration. You must be able to trace code, spot syntax and logic errors, and write snippets in pseudocode or a high-level language (often Java for IB and Python or C# for OCR).

IB和OCR都要求掌握顺序、选择和循环。你必须能够追溯代码、发现语法和逻辑错误，并用伪代码或高级语言（IB常用Java，OCR常用Python或C#）编写代码片段。

Variables, constants, data types (integer, real, boolean, char, string), and type casting form the bedrock. A typical exam question might ask you to convert an algorithm from a flowchart into pseudocode, or to identify why a loop doesn’t terminate.

变量、常量、数据类型（整形、实型、布尔、字符、字符串）和类型转换构成了基础。典型的考题可能要求将流程图算法转换为伪代码，或找出循环不终止的原因。

Control Structure	控制结构	Example / 示例
Selection – IF	选择 – IF	IF score >= 90 THEN grade = ‘A*’
Iteration – FOR	循环 – FOR	FOR i = 0 TO n-1
Iteration – WHILE	循环 – WHILE	WHILE NOT endOfFile DO read()

Sub-programs (functions, procedures) and parameter passing (by value vs by reference) are high-frequency topics. A procedure returns no value, while a function does. IB often asks about ‘scope’ and side effects.

子程序（函数、过程）和参数传递（值传递与引用传递）是高频主题。过程不返回值，而函数返回。IB常问及“作用域”和副作用。

3. Data Structures: Arrays, Lists, Stacks, and Queues | 数据结构：数组、列表、栈和队列

Arrays (static) and linked lists (dynamic) are examined for their memory allocation, access speed, and insertion/deletion efficiency. A 1D array is stored in contiguous memory; a linked list uses nodes with pointers.

数组（静态）和链表（动态）考查其内存分配、访问速度以及插入/删除效率。一维数组存储在连续内存中；链表使用带指针的节点。

Stacks follow LIFO (Last In First Out) and are used in recursion backtracking, undo features, and expression evaluation. Queues use FIFO (First In First Out), common in print spooling and breadth-first search.

栈遵循后进先出（LIFO），用于递归回溯、撤销功能和表达式求值。队列使用先进先出（FIFO），常见于打印缓冲和广度优先搜索。

Stack operations: push(), pop(), peek(). Overflow occurs when pushing to a full stack; underflow when popping from an empty stack.

栈操作：push()、pop()、peek()。当向满栈推入时发生溢出；从空栈弹出时发生下溢。
Queue operations: enqueue(), dequeue(). A linear queue may waste space; a circular queue reuses vacated slots by wrapping around.

队列操作：enqueue()、dequeue()。线性队列可能浪费空间；循环队列通过环绕方式重用空出的位置。

4. Algorithms: Searching and Sorting | 算法：查找与排序

Linear search scans each element sequentially until the target is found or the end is reached. Its average time complexity is O(n). Binary search requires a sorted array and halves the search space each step, giving O(log n).

线性搜索顺序扫描每个元素直到找到目标或到达末尾。平均时间复杂度为O(n)。二分搜索需要一个有序数组，每一步将搜索空间减半，时间复杂度为O(log n)。

Standard sorting algorithms like bubble sort (O(n²)), insertion sort (O(n²) but adaptive), and merge sort (O(n log n)) appear in both syllabi. You should be able to trace each step and compare their efficiency on nearly-sorted data.

标准排序算法如冒泡排序（O(n²)）、插入排序（O(n²)但自适应性）和归并排序（O(n log n)）都出现在两个大纲中。你应该能够追踪每一步，并比较它们在近似有序数据上的效率。

A merge sort divides the list recursively into sublists of size 1, then merges them in sorted order. The IB may ask you to write pseudocode for the merge step, while OCR often includes a dry-run question.

归并排序递归地将列表分割成大小为1的子列表，然后按序合并它们。IB可能要求编写合并步骤的伪代码，而OCR常包含手跑（dry-run）题目。

5. Recursion and Algorithm Efficiency (Big O) | 递归与算法效率（大O表示法）

Recursion is a method where a function calls itself to solve smaller instances of the same problem. You must identify the base case (to stop recursion) and the recursive case. Common examples: factorial, Fibonacci, and tree traversals.

递归是一种方法，函数调用自身以解决同一问题的更小实例。你必须识别基准情形（停止递归）和递归情形。常见例子：阶乘、斐波那契和树的遍历。

Big O notation describes the upper bound of time or space complexity as input size n grows. For example, O(1) is constant time, O(n) linear, O(n²) quadratic, and O(2ⁿ) exponential. Both IB and OCR expect you to evaluate the efficiency of an algorithm and choose the most suitable one for a given context.

大O表示法描述随着输入规模n增长的时间或空间复杂度的上界。例如O(1)是常数时间，O(n)线性，O(n²)二次方，O(2ⁿ)指数级。IB和OCR都要求你评估算法效率，并为给定场景选择最合适的算法。

Tracing a recursive call stack is a classic exam question. You may be asked how many recursive calls occur when computing factorial(5) or why a particular recursion leads to stack overflow.

追踪递归调用栈是经典的考题。你可能会被问到计算factorial(5)时发生了多少次递归调用，或者某个递归为何导致栈溢出。

6. Computer Architecture and the Fetch-Decode-Execute Cycle | 计算机架构与取指-译码-执行周期

The Von Neumann architecture stores both data and instructions in the same memory. The processor repeatedly fetches an instruction from memory, decodes it via the control unit, and executes it using the ALU, writing back the result if needed.

冯·诺依曼架构将数据和指令存储在同一内存中。处理器反复从内存取指令，通过控制单元译码，再使用ALU执行，必要时写回结果。

Key registers are Program Counter (PC), Memory Address Register (MAR), Memory Data Register (MDR), Current Instruction Register (CIR), and Accumulator (ACC). Their roles are tested via fill-in-the-blank or diagram-labelling questions.

关键寄存器包括程序计数器（PC）、内存地址寄存器（MAR）、内存数据寄存器（MDR）、当前指令寄存器（CIR）和累加器（ACC）。它们的角色会通过填空或图示标注题来考查。

Factors affecting CPU performance – clock speed, number of cores, cache size, and pipe-lining – are high-yield topics. Hyper-threading and multi-core parallelism enable simultaneous instruction execution, but not all tasks can be parallelised.

影响CPU性能的因素——时钟速度、核心数量、缓存大小和流水线——是高频考点。超线程和多核并行化能实现同时指令执行，但并非所有任务都可并行化。

7. Operating Systems and Memory Management | 操作系统与内存管理

The operating system manages hardware, provides a user interface, handles interrupts, and ensures security. Paging, segmentation, and virtual memory are core OS topics. Paging divides memory into fixed-size frames and processes into pages, mapping them via a page table.

操作系统管理硬件、提供用户界面、处理中断并确保安全。分页、分段和虚拟内存是操作系统的核心主题。分页将内存划分为固定大小的帧，将进程划分为页，并通过页表进行映射。

Virtual memory uses secondary storage to simulate extra RAM. When RAM is full, the OS swaps idle pages to disk (swap space). Excessive swapping causes disk thrashing, significantly slowing the system. Exam questions often ask about benefits and drawbacks.

虚拟内存使用辅存模拟额外的RAM。当RAM满时，OS将空闲页交换到磁盘（交换空间）。过多的交换导致磁盘抖动，显著降低系统速度。考题常问其优缺点。

Scheduling algorithms (round robin, FCFS, shortest job first, multi-level feedback queue) determine the order processes are executed by the CPU. You need to compare their fairness, throughput, and response time.

调度算法（轮转、先来先服务、最短作业优先、多级反馈队列）决定进程被CPU执行的顺序。你需要比较它们的公平性、吞吐量和响应时间。

8. Databases and SQL | 数据库与SQL

Relational databases use tables linked by foreign keys. Data integrity is enforced through entity integrity (primary key not null), referential integrity (foreign key values must exist in the parent table), and domain constraints.

关系型数据库使用通过外键链接的表。数据完整性通过实体完整性（主键非空）、参照完整性（外键值必须存在于父表中）和域约束来强制实施。

SQL commands are split into DDL (Data Definition Language: CREATE, ALTER, DROP) and DML (Data Manipulation Language: SELECT, INSERT, UPDATE, DELETE). Queries using JOIN, GROUP BY, HAVING, and nested SELECTs are frequently examined.

SQL命令分为DDL（数据定义语言：CREATE、ALTER、DROP）和DML（数据操纵语言：SELECT、INSERT、UPDATE、DELETE）。使用JOIN、GROUP BY、HAVING和巢状SELECT的查询经常被考查。

Example: SELECT Student.Name, Grade FROM Student INNER JOIN Enrolment ON Student.ID = Enrolment.StudentID WHERE Grade >= 7 FETCH FIRST 10 ROWS ONLY;

示例：SELECT Student.Name, Grade FROM Student INNER JOIN Enrolment ON Student.ID = Enrolment.StudentID WHERE Grade >= 7 FETCH FIRST 10 ROWS ONLY;

Normalization up to 3NF removes data redundancy and update anomalies. 1NF eliminates repeating groups; 2NF removes partial dependencies; 3NF eliminates transitive dependencies.

规范化至第三范式（3NF）消除数据冗余和更新异常。第一范式消除重复组；第二范式消除部分函数依赖；第三范式消除传递函数依赖。

9. Computer Networks and Protocols | 计算机网络与协议

The TCP/IP stack (Application, Transport, Internet, Link) is fundamental. Protocols like HTTP/HTTPS, FTP, SMTP, POP3/IMAP operate at the application layer; TCP and UDP at transport; IP at internet; Ethernet at link. IB and OCR both require you to match protocols to layers and explain their purpose.

TCP/IP协议栈（应用层、传输层、网际层、链路层）是基础。如HTTP/HTTPS、FTP、SMTP、POP3/IMAP等协议工作于应用层；TCP和UDP在传输层；IP在网际层；以太网在链路层。IB和OCR都要求你将协议与层配对并解释其用途。

Packet switching breaks messages into packets, each travelling independently and reassembled at the destination. Routers use the destination IP to forward packets. Circuit switching establishes a dedicated path for the entire communication but is less efficient for bursty data.

分组交换将消息拆分成包，每个包独立传输并在目的地重组。路由器使用目标IP转发包。电路交换为整个通信建立专用路径，但对突发性数据效率较低。

Client-server vs peer-to-peer models: the former centralises resources and management, while the latter distributes workload among equal peers. Hybrid models (like BitTorrent) combine both. Your exam may ask to recommend a model for a given organisation.

客户端-服务器与对等网络模型：前者集中资源和管理，后者在对等的节点之间分配工作负载。混合模型（如BitTorrent）结合了二者。考试可能要求为给定组织推荐一种模式。

10. Cyber Security and Encryption | 网络安全与加密

Symmetric encryption uses a single shared key (e.g., AES, DES). Asymmetric encryption employs a public/private key pair (RSA). Digital signatures use the sender’s private key to create a signature that the receiver verifies with the sender’s public key.

对称加密使用单一共享密钥（如AES、DES）。非对称加密使用公钥/私钥对（RSA）。数字签名使用发送方的私钥创建签名，接收方用发送方的公钥验证。

Common threats include phishing, malware, DoS attacks, SQL injection, and man-in-the-middle attacks. Mitigations: firewalls, anti-malware software, user training, penetration testing, and input sanitisation (e.g., parameterised queries).

常见威胁包括网络钓鱼、恶意软件、拒绝服务攻击、SQL注入和中间人攻击。缓解措施：防火墙、反恶意软件、用户培训、渗透测试和输入清理（如参数化查询）。

Data integrity and authentication are often combined using hash functions like SHA-256. A hash digest ensures that data has not been tampered with. Passwords should be hashed and salted before storage.

数据完整性和认证常通过SHA-256等哈希函数结合使用。哈希摘要确保数据未被篡改。密码在存储前应该经过哈希加盐处理。

11. Object-Oriented Programming Concepts | 面向对象编程概念

Classes are blueprints that encapsulate data (attributes) and behaviours (methods). Objects are instances of a class. Encapsulation hides internal state and forces interaction through public methods, promoting maintainability.

类是将数据（属性）和行为（方法）封装起来的蓝图。对象是类的实例。封装隐藏了内部状态并强制通过公共方法进行交互，从而提升可维护性。

Inheritance allows a subclass to derive attributes and methods from a superclass, supporting code reuse. Polymorphism lets objects of different classes be treated as objects of a common superclass, typically through overriding methods.

继承允许子类从超类派生属性和方法，支持代码重用。多态性使不同类的对象可被视为一个公共超类的对象，通常通过重写方法实现。

UML class diagrams depicting class names, attributes, methods, and relationships (association, aggregation, composition, dependency) are assessed in IB Paper 2 Option D and OCR’s Component 02. Composition (filled diamond) implies the part cannot exist without the whole.

显示类名、属性、方法和关系（关联、聚合、组合、依赖）的UML类图在IB试卷二选项D和OCR单元02中会进行评估。组合（实心菱形）意味着部分不能脱离整体而存在。

Association: a ‘has-a’ relationship between objects, e.g., a Library has Books.

关联：对象之间的“拥有”关系，例如图书馆拥有图书。
Aggregation: weak ownership, part can exist independently, e.g., a Department has Lecturers.

聚合：弱拥有关系，部分可以独立存在，例如系部拥有讲师。

12. Ethical and Social Implications | 伦理与社会影响

This cross-cutting topic demands critical evaluation of computing technologies. Issues include privacy, surveillance, digital divide, automation unemployment, intellectual property, and AI bias. Both IB and OCR require reasoned arguments using real-world examples.

这个跨领域主题要求对计算技术进行批判性评估。议题包括隐私、监控、数字鸿沟、自动化失业、知识产权和人工智能偏见。IB和OCR都要求用现实世界的例子进行有说服力的论证。

The General Data Protection Regulation (GDPR) sets principles for data collection: lawfulness, fairness, purpose limitation, data minimisation, accuracy, storage limitation, integrity, and accountability. A common exam question asks how a company should amend its data practices to comply.

《通用数据保护条例》（GDPR）规定了数据收集的原则：合法、公平、目的限制、数据最小化、准确、存储限制、完整性和问责制。常见的考题是问公司应如何修改其数据操作以合规。

Ethical algorithms must be transparent and avoid reinforcing biases. For instance, biased training data in a recruitment AI can perpetuate historical discrimination. The IB Ethical Scenarios paper requires applying ethical theories (consequentialism, deontology) to technology dilemmas.

合乎伦理的算法必须是透明的，并避免强化偏见。例如，招聘AI中带有偏见的训练数据可能延续历史上的歧视。IB伦理情景论文要求将伦理理论（后果主义、义务论）应用于技术难题。

Globalisation and cultural diversity: computing solutions must respect different cultural norms. Green computing addresses the environmental footprint of data centres and e-waste, emphasising energy-efficient hardware and recycling.

全球化与文化多样性：计算解决方案必须尊重不同的文化规范。绿色计算关注数据中心和电子垃圾的环境足迹，强调节能硬件和回收利用。

Published by TutorHao | Computer Science Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

IB OCR Computer Science: High-Frequency Exam Topics Summary | IB OCR 计算机科学：高频考点总结