Structure of the Internet

Physical Structure

• Each continent uses backbone cables connected by trans-continental leased lines fed across the sea bed.
• National Internet Service Providers (ISPs) connect directly to this backbone and distribute the internet connections to smaller providers.

Mesh Network

Mesh Network 

A mesh network is a topology network where all nodes (computers) are connected, linked together, in a network.

Advantages

• No cabling cost
• More nodes that are installed, the faster and more reliable the network becomes, since one blocked or broken connection can easily be circumvented by another route.
• New nodes are automatically incorporated into the network.
• Faster communication since data packets do not need to travel via central switch.

Disadvantages

• It is costly

Data Stuff

Data Capturing

• Data has to be entered into a database:
  • Manual Methods – entered by a user, cheques scanned at the bank using magnetic ink character recognition (MICR). The amount has to be manually entered.
  • Lottery Tickets – exams can be read using optical mark recognition, or other types of forms can include optical character recognition.
  • Smart Card Readers – scanners used at airports to scan passports.

 

Selecting and Managing Data

• Data can be selected before its added to a database, if it matches a criteria.
  • Speed camera, only takes pictures of cars, that exceed the speed limit.
• Once in the database, SQL can be used to select data from multiple tables.
  • Ordering stock, updating or adding data.

 

Exchanging Data

• Method of transferring data between computers without the need for human intervention Electronic Data Interchange (EDI)
• Can be used in different applications – Exams board use this to send results to school.

Relational Database and Normalisation

Relational Database Design

• Data is held in tables – called relations and the tables are linked by means of common attributes.
• Relational database is a collection of tables in which relationships are modeled b shared attributes.
• One row of a table holds one record. Each column represents one attribute.
• A table holding data about an entity Book may have the following rows and columns.

• To describe the table shown above you would write Book (BookID, DeveyCode, Title, Author, DatePublished).
• The entity name is shown outside the brackets.
• The attributes are listed inside the brackets.
• The primary key is underlined. It is composed of one or more attributes that will uniquely identify a particular record in the table. (When describing an entity is called an entity identifier)

Indexing

• An index of primary keys will be automatically maintained by the database software.
• This allows for the given position of each record to be identified according to its primary key.
• One or more secondary indexes may be defined when the database is created, for any attribute often used as a search criterion.
• In the table earlier both Author and Title might be defined as secondary keys.
• This wills peed up searching on either of these fields, which would otherwise have to be searched sequentially.

Normalisation

• Normalisation is used to come up with the best possible design for a relational database.
• Tables need to be organised in a way that:
  • No data is unnecessarily duplicated.
  • Data is consistent throughout the database.
  • Consistency should be an automatic consequences of not holding any duplicated data.
  • This means that anomalies will not arise when data is inserted, amended or deleted.
  • The structure of each table is flexible enough to allow you to enter as many or as few items as required
  • The structure should enable a user to make all kinds of complex queries relating to data from different tables.
• 3 basic stages of normalisation known as first, second and third normal form

Database

 A simple database

• The simplest kind of database is a flat file, consisting of information about a single entity
• Definition: An entity is a category of object, person, event or thing of interest about which data needs to be recorded
• For example you might hold data about club members or concert venues

Entities

• You may have thought of these entities:
  • Customer
  • Guide or Product
  • Subscription
• Other entities that could be considered include customer order, subject, author (of a revision guide)
• We will keep it simple and just consider Customer, Product and Subscription.

Writing an entity description

• This will be a database system, called RevisionSubs
• Each entity in the database has attributes
• The entity descriptions can be written in this format:
  • Customer (custID, title, firstname, surname, email)
  • Product (productID, title, subject, level, price)
  • Subscription (subID, startDate, endDate)

Entity identifier (primary key)

• Each entity needs an identifier which uniquely identifies a particular record.
• In a relational database, the identifier is known as the primary key.
• It is underlined in the entity description:
  • Customer (custID, title, firstname, surname, email)
  • Product (productID, title, subject, level, price)
  • Subscription (subID, startDate, endDate)
• If there is no natural attribute for a primary key, one should be introduced.

Composite primary key

• Sometimes two or even more attributes are needed to uniquely define a record.
• For example, in a customer order consisting of many different order lines, each order line may be uniquely identified by the two attributes orderNumber and orderLine.
• OrderLine (OrderNumber, OrderLine, ProductID, …)
• OrderNumber, OrderLine is a composite primary key

Secondary key

• The primary key field is automatically indexed so that any particular record can be found very quickly.
• In some databases, searches may often need to be made on other fields.
• In the product table:
  • Product (productID, title, subject, level, price)

  if searches often need to be made on title or subject, either or both of these fields could be defined as a secondary key.

  • They would then be indexed for faster lookups.

Relationships between entities

• The three entities are linked, or related.
• There are three possible ways in which two entities may be related:
  • One-to-one e.g. Husband and Wife.
  • One-to-many e.g. Mother and Child, School and Pupil.
  • Many-to-many e.g. Actor and Film, Recipe and Ingredient.
• What is the relationship between Customer and Subscription?
• What is the relationship between Product and Subscription?

Entity relationship diagrams

• An entity relationship (E-R) diagram is a graphical way of representing the relationships between entities.
• We can say, for example, that one school has many pupils, or many pupils attend one school.

E-R diagram

• There is a one-to many relationship between Customer and Subscription
  • One customer may have several subscriptions, but a particular subscription belongs to only one customer.
• There is a one-to-many relationship between Product and Subscription
  • One product may appear on several subscriptions, but a subscription is for only one product.

Database structure

• Each entity is represented by a table.
• Tables in a relational database are commonly referred to as relations.
• A database contains one or more relations.
• A relation has rows, each row containing one record.
• The columns in the relation each contain one field (i.e. attribute) belonging to the records

Creating a relationship

• To create a relationship between Customer and Subscription, we need to include custID in the entity description of Subscription
• Subscription (subID, startDate, endDate, custID )
• ProductId also needs to be included in the entity description of Subscription

• Subscription (subID, startDate, endDate, custID, productID)
• custID and productID are foreign keys in Subscription, shown in italics • A foreign key always goes on the “many” side of a relationship

Compression

Lossy Compression

• Works by removing non essential information
• The heavily compressed images looks untidy, blocky compression.
• However, we can make out the subject of the image well, but the degree to which they are compressed comes at the cost of quality
• Sound compression works in a similar way
• MPS files use lossy compression
• Voice is transmitted over the internet or mobile telephone networks using lossy compression

Lossless Compression

• Works by recording patterns in data rather than the actual data
• Using these patterns and a set of instructions on how to use them, the computer can reverse the procedure and reassemble an image, sound or text file with exact accuracy and no data is lost.
• Results in a much larger file than a lossy file, but one that is still significantly smaller than the original