What is parallel processing?

Carrying out more than one computation at the same time, using multiple processors or cores.

What must a program do to make use of more than one core?

It must be written so that separate parts of the work can run independently, because a sequential program occupies one core however many are available.

WJEC A-Level Computer Science 601-5345-3

Hardware: Architecture, Memory and Storage

Topic overview Components of a computer system and how they connect Von Neumann architecture and the CPU The fetch-execute cycle Memory and caching Parallel processing Input and output devices and their uses Secondary storage, fragmentation and defragmentation

Networks and Data Transmission

Topic overview Serial and parallel transmission Simplex, half duplex and full duplex Multiplexing and switching Packets and the TCP/IP protocol stack Networks, standards and protocols Collisions and shared media Routing, addressing and the internet

Logic Gates and Boolean Algebra

Topic overview Logic gates and their truth tables Boolean notation and expressions Building and reading truth tables Boolean identities and rules Simplifying Boolean expressions De Morgan's laws Masking and clearing registers

Data Representation and Data Types

Topic overview Bits, bytes and words The binary number system Hexadecimal notation Character sets and storing characters Primitive data types and storage requirements Binary arithmetic and negative numbers Real numbers, truncation and rounding

Data Structures

Topic overview Data structures and data types One-dimensional arrays Two-dimensional arrays Manipulating arrays Records and fields Arrays of records Choosing a data structure

Organisation of Data: Files and Records

Topic overview Files, records and fields Fixed and variable length records Master and transaction files File access and organisation methods Sequential file access and update algorithms File security, backup and archiving Data validation and verification

Database Systems

Topic overview Flat files and the relational model Relational and other contemporary database systems Primary, foreign, composite and candidate keys Indexes and links between tables Entity relationship modelling Normalisation to third normal form The database management system, queries and user views

The Operating System

Topic overview The need for an operating system The kernel and process management Memory protection and the backing store The user interface between user and hardware Directory structures and file attributes Utility software Batch, real time control and transaction processing

Algorithms and Programs

Topic overview Defining and designing algorithms Variables, constants and identifiers Scope, subprograms and parameter passing Sequence, selection and repetition Mathematical and logical operations Standard algorithms: sorting, searching and compression Dry running, trace tables and test data

Principles of Programming

Topic overview Procedural programming Event-driven and visual programming Mark-up languages Object-oriented programming: objects, classes and methods High-level languages Low-level languages Choosing a language for a problem

Systems Analysis

Topic overview The systems life cycle and the feasibility study Methods of investigation Analysis, abstraction and decomposition Design and data flow diagrams Testing and changeover Maintenance Backup, recovery and documentation

Software Engineering and Program Construction

Topic overview Integrated development environments and their purpose Editors, syntax highlighting and code entry tools Debugging tools: breakpoints, stepping and watches Translation, source code and object code Lexical analysis and the symbol table Syntax and semantic analysis Code generation and optimisation

Types of Software System

Topic overview Off-the-shelf, bespoke and custom software Open source and proprietary software Computers in weather forecasting Computer aided design and manufacture Robotics in industry Computer generated graphics and animation Expert systems

Practical Programming

Topic overview Designing data structures: fields, records and data types Key fields and methods of access Validation of input data Designing routines with pseudocode Tracing code and test data Prototyping and evaluating a solution Annotated listings and readable code

Data Security and Integrity

Topic overview Dangers of holding files of personal data Levels of permitted access and access control Passwords and authentication Clerical procedures and write-protection Threats to computer systems Malicious and accidental damage to data Contingency planning and disaster recovery

Computing, Society and the Law

Topic overview Social and economic change from computing Computers and the nature of employment The digital divide and cultural change Environmental cost of computing Data protection and freedom of information law The Computer Misuse Act 1990 and computer crime Copyright, surveillance and professional codes

Advanced Data Structures and Logic

Topic overview Arrays in three dimensions Stacks and queues using pointers and arrays Linked lists using pointers and arrays Binary trees using pointers and arrays Hash tables and collision resolution NAND, NOR and Boolean simplification Masking, clearing registers and encryption

Recursion, Sorting and Searching

Topic overview Recursion Quicksort and merge sort Comparing sorting algorithms Big O notation and algorithm complexity Searching and shortest-path algorithms Traversal of data structures Data compression and its efficiency

Programming Paradigms and Translators

Topic overview Functional, declarative and imperative paradigms Classes, methods and object-oriented design Standardisation of computer languages Ambiguity in natural language and unambiguous notation Backus-Naur form and syntax diagrams Compilers, interpreters and assemblers Translation and execution errors

System Design and Human-Computer Interaction

Topic overview The waterfall approach Agile approaches Documentation and when it is produced Contemporary approaches to communicating with computers Natural language interfaces and spoken ambiguity Design review and validation Evaluating a computer based solution

Software Engineering Tools and Professional Practice

Topic overview Software tools for the software engineering process Software packages in systems analysis and specification Software packages in systems design Testing tools and testing strategies Program version management Program libraries and project management tools Codes of conduct and professional behaviour

Advanced Architecture and Assembly Language

Topic overview Components of contemporary architecture Von Neumann and Harvard architectures The fetch-execute cycle in detail Parallel processing and its limits Assembly language and instruction formats Voice input, command and recognition Mobile and wireless networking

Communication Networks and File Organisation

Topic overview Calculating data transfer rates Lowest-cost routing The internet as a global infrastructure Creating, organising and processing files Hashing algorithms Overflow and file reorganisation Choosing an organisation and a route

Number Representation and Floating Point

Topic overview Binary arithmetic Sign and magnitude representation Two's complement Floating point representation Converting between real numbers and floating point Truncation, rounding and accuracy Shifts, overflow and underflow

Databases and Distributed Systems

Topic overview Consistency, redundancy and data independence Relational organisation and normalisation Restructuring data into third normal form Entity relationship modelling Query languages and SQL Validation, verification and the DBMS Big Data, data warehousing and mining Distributed systems and distributed databases

Operating System Management

Topic overview Types of operating system Interrupts and interrupt handling Memory management and partitioning Buffering and data transfer Scheduling and process states Scheduling policies and priorities Evaluating operating system design

Software Systems and Applications

Topic overview The range of software system types Safety related and safety critical systems Computer control and automation The scope of automation in practice Expert systems The internet, intranets and search Selecting and evaluating a software system

Data Security, Cryptography and Attack

Topic overview Integrity problems in online updating The need for cryptography Symmetric and asymmetric techniques Classical ciphers and cryptographic algorithms Comparing cryptographic strength Biometric technologies Malicious software and its vectors Attacks, defence and penetration testing

Advanced Architecture and Assembly Language

Parallel processing and its limits

3 things to know
A2 Unit 4

Overview

Parallel processing, difference between concurrency and parallelism and name of the law describing that limit.

What you need to know

Answer each one before you reveal it.

Your recallNot started

Key term

What is the difference between concurrency and parallelism?

Process

Why can a task not always be made faster by adding processors?

Key term

State Amdahl's law in words.

Questions from this subtopic

Each one with the model answer underneath.

1
What is parallel processing?
Model answer
Carrying out more than one computation at the same time, using multiple processors or cores.
2
What must a program do to make use of more than one core?
Model answer
It must be written so that separate parts of the work can run independently, because a sequential program occupies one core however many are available.
3
What is the difference between concurrency and parallelism?
Model answer
Concurrency is managing several tasks whose execution overlaps in time; parallelism is actually executing them at the same instant on different hardware.
4
Why can a task not always be made faster by adding processors?
Model answer
Only the parallelisable part of a task benefits, so the sequential portion sets a floor on the total time.
5
What is the name of the law describing that limit?
Model answer
Amdahl's law.
6
State Amdahl's law in words.
Model answer
The speed-up from parallelisation is limited by the proportion of the task that must run sequentially.

See the rest and revise

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.

1Closed book

2Open book

3Closed book

The 3-step method, run for you

Answer cold, learn it, then prove it stuck. Levo tracks the jump between your first attempt and your last.

Hardware: Architecture, Memory and Storage

Networks and Data Transmission

Logic Gates and Boolean Algebra

Data Representation and Data Types

Cram a whole paper in one sitting

Sort every topic into what you know and what you do not, then drill only the weak ones.

Fetch, , execute.

decodecompilestore

Blank Blitz makes facts automatic

Rapid timed gap-fills on the terms and numbers you need to recall without thinking.

AEncryption is reversible.

BThe stored form could be decrypted.

CA hash cannot be reversed.

Build It gets the exam shape right

Assemble a full-mark answer from its marking points, in the order an examiner wants them.

A bit is a single binary digit.

TrueFalse

Test It catches misconceptions

Spot the fault in a statement. The traps are the mistakes students actually make.

721941163128510

Brain Breaks between sessions

Short focus games to reset attention, so the next set lands better than the last.

Every question in Advanced Architecture and Assembly Language, on a schedule.

Attempt a question from memory, write the correct version, then attempt it again. Levo records the jump between your first attempt and your last, and brings each question back on the day you are due to forget it.

Start revising, £5 a month See how the method works

Or £46 a year, about £3.83 a month. Cancel anytime.

Parallel processing and its limits

Overview

What you need to know

What is the difference between concurrency and parallelism?

Why can a task not always be made faster by adding processors?

State Amdahl's law in words.

Questions from this subtopic

What is parallel processing?

What must a program do to make use of more than one core?

What is the difference between concurrency and parallelism?

Why can a task not always be made faster by adding processors?

What is the name of the law describing that limit?

State Amdahl's law in words.

Other ways to work this

What is inside Levo

The 3-step method, run for you

Cram a whole paper in one sitting

Blank Blitz makes facts automatic

Build It gets the exam shape right

Test It catches misconceptions

Brain Breaks between sessions

Every question in Advanced Architecture and Assembly Language, on a schedule.