ORACLE is a member of a large category of products known as database management systems (DBMS). The general purpose of a DBMS is to provide for the definition, storage, and management of data in a centralized area that can be shared by many users. ORACLE's database management system is patterned on the relational model described by Dr. E. F. Codd.

When talking about data in a database, it is common to look at the data in three different ways:

• External or user views: the view of the data that each person has for their functional area.
• Logical Data Model or Conceptual Model: the entire collection of user views for each functional area. In other words, it is the enterprise's overall view of the database.
• Physical or Internal view: the structure used to physically implement the logical data model on a physical medium.

The ORACLE database management system closely models this three layered structure. The physical or internal view will be elaborated upon in this tutorial.

SQL, as a relational data language supports certain basic functions to control, define and manipulate data. The three categories of SQL are as follows:

• The Data Manipulation Language (DML) is used to access and modify data with the SELECT, INSERT, DELETE and UPDATE verbs.
• The Data Definition Language (DDL) defines the physical database objects with the CREATE, DROP, and ALTER SQL verbs.
• The Data Control Language (DCL) provides the control statements that govern data security with the verbs GRANT and REVOKE.

These three parts of speech exist within SQL independently of the DBMS that is being used (ie: ORACLE, DB2 etc).

Some definitions of commonly used terms are necessary when working with DBMS's. A database is nothing more than a defined grouping of tables, indexes and storage areas. A database contains no data but acts as a high level identifier for tracking other objects. A tablespace is an object that defines the physical structure of the data sets used to house the table data. A table is an object, sometimes referred to as a relation. Each table consists of columns (vertically) and rows (horizontally) which define the physical characteristics of the data to be stored. A view is a virtual table. A view never stores data and does not physically exist. When you access a view the SQL statement that defined it is executed to return the desired data. Indexes are used to achieve more efficient access to tables. An index places table data in its logical sequence rather than its physical sequence. Lastly a column is a single nondecomposable data element in a table.

The tablespace owns all the tables contained there. The tables own the views, indexes, and columns defined for them.

For more information about tablespace in ORACLE, see About Tablespace in ORACLE.

When a query is issued with one or more tables as input, SQL will output a single table typically referred to as a results table. The results table can be composed of multiple rows and columns or it can contain a single value which was derived from one row and one column.

Extracting data from a database is the most common SQL operation. A database retrieval is called a query and to issue a query you use the SELECT command. The selection operation retrieves a specified subset of rows from a table. When selecting information from a table you will use some primary keywords. They are SELECT, FROM, WHERE and ORDER BY. SELECT and FROM are used in every query you do to select data from tables.

To get started with ORACLE SQL, each user requires an OS (operating system) account, an ORACLE database account and access to the SQL*Plus program. Go to the next section of this tutorial for a brief overview of the OS, what it is, how to get a system account, how to sign onto the system and onto ORACLE'S SQL*Plus.

Last revised 7/31/1998 by J. Warren, Kennesaw State University

School of Computer Science and Information Systems

Return to Oracle/SQL Tutorial Contents Page | Go to Next Section - Getting Started With Oracle on CSIS