Data storage: File-oriented system

As discussed earlier, the manual paper-based data storage possesses many disadvantages. This manual system was replaced by the computerized filling system. File oriented systems were the first attempt to computerize the data storage and various operations on data.

database system concepts

So, before the advent of DBMS, organizations were using the file systems supported by operating systems to store data. To perform various operations on these data, such as – insert, update, delete, sort, search – system had a number of application programs developed using some programming language. These application programs directly interact with the file system of the operating system, and operating systems in turn interact with the files on disk to perform various operations.

Let’s consider here a bank enterprise. Suppose a bank allows two different kinds of accounts – saving accounts and current accounts. So, to manage data about these two types of accounts, two different files are used

One file, to store details about all customers having saving account, and another, to store details about all customers having current account.

Also, system contains a number of programs to perform various tasks on these data. These programs are developed by application programmers using some programming language, such as c, c++, java, etc. some of the tasks are

Debit or Credit an account                                     

Find balance of a given account,

Find customer id, name, address, e-mail and contact number for all customers.

Khudiram Bose history

Such type of system looks simpler. But, it suffers from several serious drawbacks, particularly in comparison with DBMS. These drawbacks are described in due to these drawbacks, file-oriented systems are almost obsolete.

Why to learn File-oriented system?

With advent of DBMS, the File-Oriented systems are almost obsolete now. Though, learning of file-oriented systems provide following advantages.

1.       It provides useful historical information about how data are managed earlier.

2.       Characteristics of File-Oriented systems provide understanding complexities in designing a database.

3.       File-Oriented systems create many problems. Knowledge of those problems can help to stop problems of managing database.

4.       It provides useful knowledge for converting a file system to a database system.

Advantages of File-Oriented system

File oriented systems possess some advantages compared to paper-based system. These advantages are given below.

Back-up :

It is possible to require faster and automatic back-up of database stored in files of computer-based systems. Computer systems provide functionalities to serve this purpose. It is also possible to develop specific application programs for this purpose.

Holi essay in English 2020

Editing:

It is easy to edit any information stored in computers in sort of files.
Specific application programs or editing software are often used for this purpose.

Remote access:

In computer-based systems, it's possible to access data remotely. So, to access data, it's not necessary for a user to stay present at location where these data are kept.

Sharing:

Data stored in files of computer-based systems are often shared among multiple users at a same time.

In general, file-oriented systems overcome the problems suffered by the paper-bases systems. So, File-Oriented systems are far better than paper-based systems. But, there are drawbacks of file-oriented systems too as given in next sub-section.

Disadvantage of File-Oriented System

Data redundancy:

It is possible that an equivalent information could also be duplicated in several files. This leads to data redundancy. Data redundancy results in memory wastage.

For example, consider that some customers have both sorts of accounts-saving and current. during this case, data about customers – name, address, e mail, contact number – are going to be duplicated in both files, file for saving accounts and file for current accounts. This leads to requirement f higher storage space. In other words, same information will be stored in two different locations and, it wastes memory.

Limited data sharing:

Data are scattered in various files. Also, different files may have different formats and these files could also be stored in several folders (directories) could also be of various computers of various departments. So, due to this data isolation, it is difficult to share data among different application.

Concurrent access anomalies:

This is often for the sake of higher performance and faster response.
Consider an operation to debit and account. The program reads the old balance, calculates the new balance, and writes new balance back to database suppose an account features a balance of Rs. 5000. Now, a concurrent withdrawal of Rs. 1000 and Rs. 2000 may leave the balance Rs. 4000 or Rs. 3000 depending upon their completion time instead of the right value of Rs. 2000.But, thanks to lack of co-ordination among different application programs, this is often impossible in file processing systems.

Integrity problems:

Data integrity means the info contained within the database is both correct and consistent. For this purpose, the info stored in database must satisfy certain sorts of constraints (rules).
For example, a balance for any account must not be but zero. Such constraints are enforced within the system by adding appropriate code in application programs. But, when new constraints are added, like balance shouldn't be but Rs. 5000, application programs got to be changed. But, it's not and straightforward task to vary programs whenever required.

Difficulty in accessing data :

Accessing data isn't convenient and efficient in file processing system.
For example, suppose, there's a program to seek out information about all customers. But, what if there's a requirement to seek out out all customers from some particular city. during this case, there are two choices here: one, determine all customers using available program, then extract the needed customers manually. Second, develop new program to urge required information. Both options aren't satisfactory. for every and each different quite data access, separated programs are required. this is often neither convenient nor efficient.

worst natural disasters in the world last 10 years

Atomicity problems:

Any operation on database must be atomic. this suggests , it must happen in its entirely or not in the least .
For example, a fund transfer form one account to a different must happen in its entirely. But, computer systems are susceptible to failure, like system crash, virus attack. If a system failure occurs during the execution of fund transfer operation, it's going to possible that quantity to be transferred , say, Rs. 500, is debited from one account, but isn't credited to a different account.
This leaves database in inconsistent state. But, it's difficult to make sure atomicity during a file processing system.

Data inconsistency:

Due t data redundancy, it's possible that data might not be in consistent state. for instance , consider that an address of some customer changes. And, that customer has both sorts of accounts. Now, it's possible that this changed address is updated in just one file, leaving address in other file because it is. As a results of this, same customer will have two different addresses in two different files, making data inconsistent.

Security problems:

Database should be accessible to users during a limited way. For example, a customer can check balance just for his/her own account he/she shouldn't have access to information about other accounts.
But, in file processing system, application programs are added in a billboard hoc manner by different programmers. So, it's difficult to enforce such quite security constraints.
Metadata and system catalog

Metadata:

A metadata is that the data about data.

This means that data such as table name, column name, data type, size, constraints, authorized users, user access privileges for any table is called metadata for that table.

Metadata describes various objects, such as tables, in the database. Also, it makes it easier to access or manipulate those objects.

Metadata provides program-data independence. This means that data can be changed without requiring changes in application programs.

System catalog:

A system catalog is a repository for a metadata.

System catalog is a system-created database.

It contains description of all database objects, data dictionary information and user access information.

It also contains table-related description such as table names, owners, column names, data types, size, constraints, authorized users, user access privileges and so on.

System catalog is created by DBMS itself. It is managed in similar fashion as user-data are managed. The illustrates the differences between user-data and system-data.