Storage Methods in DBMS

Introduction to storage methods in DBMS

  • Magnetic disks and magnetic tapes are used to store data in RDBMS.
  • The disk space analyzer maintains records for available space and used space in the disk.
Before learning the storage methods, lets learn about the basics of computer memory system.

Memory Hierarchy

The computer system handles various types of memory to achieve faster execution of process.

memory hierarchy

The memory hierarchy for computer system can be elaborated as:

1. Cache Memory and Main memory
Cache memory and main memory are at the top level in the memory hierarchy which are responsible for  fast execution.
Example: RAM, ROM etc.

2. Secondary memory
Secondary memory or storage is used to store data in computer system. The secondary storage is relatively slower than cache or main memory.
Example: Magnetic tape, hard disk, CD, DVD etc.

Access methods in DBMS

The main goal of DBMS is to return data which is requested by the user. In RDBMS it may be a record or set of records. In an object- oriented database it may be object or set of objects.

Indexes – Access Method
  • Index is the small table having two columns. The first columns consist of primary key of the table and second column consists of a set of pointers holding the address of the disk, where the particular key (value) can found.
  • The indexes are very useful to improve the search operation in the DBMS system.
Type of indexes are discussed as follows:

1. Function-based indexes
  • A function-based index computes the values of expression which are present in one or more column and stored in the table.
  • The expression can be an arithmetic expression or SQL function.
  • A function-based index can not contain null value.
CREATE INDEX Sample _idx on table_(a+c*(b+d));

2. Bitmap indexes
  • Bitmap index is used to work with well for low-cardinal  (refers to columns few unique values) columns in tables.
  • For example: boolean data which has only two values true or false.
  • Bitmap indexes are very useful in data ware house applications for joining the large fact tables.
3. Domain indexes
Domain index is used to create index type schema object and an application specific index. It is used for indexing data in application specific domain.

4. Clusters
  • Clustering in DBMS is design for high availability of data. Clustering is applied on tables which are repeatedly used by the user.
  • For example: When there are many employees in the department, we can create  index of non-unique key, such a that Dept-id. With this,  all employees belonging to the same department are consider to be within a same cluster.
  • Clustering can improve the performance of the system.
5. Indexed sequential access method (ISAM)
  • ISAM was developed by IBM for mainframe computers but the term is used in several concepts.
  • In DBMS, ISAM is used to access data in sequentially (sequence in which data is entered) or randomly (with an index).