‭ /16‬

1
‭Stages in database design‬
‭-‬ ‭Analyze user environment‬
‭-‬ ‭Develop data model‬
‭-‬ ‭Choose a dbms‬
‭-‬ ‭Develop logical model by mapping conceptual model to dbms‬
‭-‬ ‭Develop a physical model‬
‭-‬ ‭Implement a physical model‬
‭Four Levels of data‬
‭-‬ ‭Real World‬
‭-‬ ‭Enterprise in its environment‬
‭-‬ ‭mini-world/universe of discourse: part of the world that is represented in the‬
‭database‬
‭-‬ ‭Conceptual model‬
‭-‬ ‭Entities, entity sets, attributes, relationships‬
‭-‬ ‭ER(Entity relationship), EER(Enhanced Entity relationship), UML (Unified‬
‭Modeling Language) diagram‬
‭-‬ ‭Logical Model‬
‭-‬ ‭Metadata, data about data‬
‭-‬ ‭Record types, data item types, data aggregates‬
‭-‬ ‭Schema - stored in data dictionary, catalog‬
‭-‬ ‭Data occurrences‬
‭-‬ ‭Databases‬
‭-‬ ‭Data instances‬
‭-‬ ‭Files‬

‭6 steps designing a data structure‬

‭1.‬ ‭Identify the data elements‬
‭a.‬ ‭Interviewing users, analyzing existing systems, evaluate comparable systems‬
‭2.‬ ‭Subdivide the data elements‬
‭a.‬ ‭If a data element contains two or more components, subdivide the element into‬
‭those components, won’t need to parse items this way‬
‭b.‬ ‭Extent to subdivision depends on how it will be used, difficult to predict all future‬
‭uses, most designers subdivide data as much as possible‬
‭c.‬ ‭Easy to rebuild by concatenating components‬
‭3.‬ ‭Identify the tables and assign columns‬
‭a.‬ ‭Divide the data elements into logic groups‬
‭b.‬ ‭Groups become tables of the database, elements and attributes will become‬
‭columns‬
‭4.‬ ‭Identify the primary and foreign keys‬
‭a.‬ ‭Each table should have a primary key that uniquely identifies each row‬
‭i.‬ ‭Values of primary keys should be seldom, short, and easy to enter‬
‭correctly‬
‭ii.‬ ‭If possible, use an existing column for primary key‬
 ‭iii.‬ ‭ composite primary key is a primary key made up by more than one‬
A
‭column in a table‬
‭b.‬ ‭If two tables have a one to many relationship‬
‭i.‬ ‭Add a foreign key column to the table on the many side‬
‭ii.‬ ‭Foreign key must have same data type as primary key column‬
‭1.‬ ‭Eg. Employee Table and Orders Table‬
‭c.‬ ‭If two tables have a many to many relationship‬
‭i.‬ ‭Define a linking table to relate them‬
‭ii.‬ ‭Each of the many to many tables will have a one to many relationship‬
‭with the linking table‬
‭d.‬ ‭If two tables have a one to one relationship‬
‭i.‬ ‭Relate by primary keys‬
‭ii.‬ ‭This relationship is typically used to improve performance‬
‭e.‬ ‭Foreign Keys‬
‭i.‬ ‭A column that references columns of another table‬
‭ii.‬ ‭Use to manage relationships/dependencies between tables‬
‭iii.‬ ‭Foreign key column must have the same data type and value as the‬
‭primary key column its related to‬
‭iv.‬ ‭Must have same data type and value as the primary key column its‬
‭related to‬
‭5.‬ ‭Normalize data structure‬
‭a.‬ ‭Normalization separates the data in a data structure into related tables‬
‭b.‬ ‭Normalization reduces data redundancy, which can cause storage, performance‬
‭and maintenance problems‬
‭c.‬ ‭Un-normalized data structure a table can contain 2 or more entities‬
‭i.‬ ‭Repeating columns with repeating values‬
‭d.‬ ‭In a normalized data structure each table contains info about a single entity‬
‭i.‬ ‭Each piece of info stored in one place‬
‭e.‬ ‭To normalize: apply the normal forms in sequence,‬
‭i.‬ ‭7 normal forms, but considered normalized if first 3 applied‬
‭ .‬ ‭Identify the indexes‬
6
‭a.‬ ‭To search and manipulate data faster‬
‭ /23‬
1
‭3 Normal Forms‬
‭1.‬ ‭First: a table must not contain any repeating columns‬
‭2.‬ ‭Second: Every non-key column must depend on the entire primary key‬
‭3.‬ ‭Third: Every non-key column must depend only on the primary key‬
‭-‬ ‭Usually between 2 and 3‬
‭-‬ ‭Benefits‬
‭-‬ ‭More tables->more clustered indexes = data retrieval more efficient‬
‭-‬ ‭Makes operations on data entries more efficient‬
‭-‬ ‭Data redundancy is minimized‬
‭First Normal form‬
‭-‬ ‭Each column must contain a single atomic value‬
‭Second Normal Form‬
‭-‬ ‭Move columns that don’t depend on entire primary key to another table and establish a‬
‭relationship‬
‭Third Normal Form‬
‭-‬ ‭All columns must come solely from primary key‬
‭-‬ ‭A column that is computed from other columns contains derived data‬