Ruankao Embedded System Designer (Intermediate) flashcards

30 cards. Read the term and the definition here. The deck at the top of this page is optional.

CPU

The central processing unit is the main processor in a computer. It fetches instructions, decodes them, and carries out arithmetic, logic, and control operations while coordinating other components.

RAM

Temporary, volatile computer memory that holds programs and data currently in use. Its contents disappear when power is lost, but it allows fast reading and writing.

Cache memory

A small, fast memory area close to the processor that stores frequently used instructions and data. It reduces average access time by keeping likely-needed information nearby.

Addressable unit

The quantity identified by one memory address, such as a byte or a word. A bus-width calculation requires this convention before converting address counts into byte capacity.

DMA

Direct memory access by a configured peripheral or controller without the CPU explicitly copying every data item. Buffer ownership, completion, coherence and protection still need an explicit protocol.

Priority inversion

A scheduling situation in which a higher-priority task waits for a lower-priority holder of a resource, potentially while other work delays that holder. Appropriate protocols and assumptions determine bounds.

TLB miss

A failed lookup in the translation cache, requiring the processor’s defined translation procedure. It does not alone establish that the target virtual page is absent from physical memory.

Page fault

An architectural exception caused by an absent or disallowed page translation or access under the platform model. Handling depends on mapping, protection and residency rather than always reading a disk.

recursion

A technique in which a method calls itself on a smaller version of the same problem. It needs a base case that stops the calls and a recursive case that makes measurable progress toward that base case.

variable

A named storage location whose value can change during program execution. In most languages a variable has a declared or inferred data type, a scope, and a lifetime determined by where it is declared.

data type

A classification that specifies which values a variable may hold and which operations apply, such as integer, floating point, character, Boolean, or string. It determines memory representation and allowed behavior.

type casting

Converting a value from one data type to another. Whether a conversion is implicit and lossless depends on the language, source type, target type and actual value. Integer-to-floating conversions can lose precision when the integer exceeds the target’s exact range; narrowing conversions also require careful checks.

scope

The region of a program where a declared name is visible and usable. Block scope limits a variable to the braces where it is declared, while global scope makes it accessible throughout the file or program.

parameter passing

How arguments move into a method. Pass-by-value copies the argument, so changes stay local; pass-by-reference gives access to the caller's variable. Java passes object references by value.

object-oriented programming

A paradigm organizing code into classes that bundle data and behavior, with objects as instances. Its pillars are encapsulation, inheritance, polymorphism, and abstraction, which support reuse and manageable large programs.

encapsulation

Hiding an object's internal state behind private fields and exposing controlled access through public methods such as getters and setters. It protects invariants and lets implementation details change without breaking callers.

inheritance

A relationship in which a subclass acquires fields and methods from a superclass and may extend or override them. It models an is-a relationship and promotes code reuse, though deep hierarchies can become brittle.

polymorphism

The ability of one interface or method name to work with multiple types. Dynamically dispatched overridden methods, such as calls through virtual functions in C++, can resolve at run time. Ordinary overload resolution depends on the language’s compile-time rules; not every same-named method implies dynamic dispatch.

syntax error

A violation of a language's grammar, such as a missing semicolon or unbalanced parenthesis, detected during compilation or parsing. The program cannot run until the error is corrected.

runtime error

A failure arising during program execution rather than parsing, with behavior defined by the language and operation. Some languages raise exceptions for invalid indexing or division; in C++ certain invalid operations instead have undefined behavior and need not produce a predictable exception.

logic error

A mistake in the program's reasoning that produces incorrect output without crashing, such as using less-than instead of less-than-or-equal in a loop condition. It is found by testing and tracing, not by the compiler.

exception

An object signaling an abnormal condition during execution. Languages use try, catch, and finally blocks or similar constructs to handle it, allowing recovery instead of abrupt termination.

Quality scenario

A stated stimulus, operating condition and observable response with a measurement criterion, allowing alternative designs and tests to be compared against an actual requirement.

信息系统生命周期

信息系统从立项、开发、运行维护到退役的完整过程，通常划分为系统规划、系统分析、系统设计、系统实施、系统运行与维护五个阶段，各阶段有明确的输入、输出和评审要求。

系统集成

将硬件、网络、软件、数据等异构部件通过接口与协议连接，使其协同工作并满足用户整体业务目标的工程活动，强调技术整合与项目管理双重能力。

中间件

位于操作系统与应用软件之间、屏蔽底层异构性并提供通信、事务、消息等公共服务的软件层，常见类型包括消息中间件、应用服务器和数据库访问中间件。

面向服务架构

将业务功能封装为松耦合、可复用、通过标准协议调用的服务，并据此组合业务流程的架构风格，其实现常借助Web服务与注册中心完成服务发现与编排。

数据仓库

面向主题、集成、相对稳定且反映历史变化的数据集合，用于支持管理决策分析；其数据通常经抽取、转换、加载过程从业务系统汇入。

数据挖掘

从大量数据中自动发现潜在模式、关联规则与预测模型的分析技术，常用方法包括分类、聚类、关联分析和回归，其结果需结合业务判断使用。

数据治理

对数据资产的质量、标准、安全、权责和生命周期进行制度化管理的活动集合，目标是使数据可信、可用、可追溯，并明确数据所有者与 stewards 的职责。