Business data processing

Business data processing is the collection, storage, processing and reporting of an organisation's routine data – payroll, inventory, billing, accounts. Computers do this efficiently only when data is arranged logically. This note covers the data processing cycle, the data storage hierarchy, methods of organising data, file types, file organisation techniques and file utilities.

Key concepts

Introduction – data processing converts raw data into useful information. Cycle: data collection/input → preparation → processing (classifying, sorting, calculating, summarising) → storage → output/reporting. Methods: manual, mechanical, electronic (EDP). Processing modes: batch (transactions collected and processed together, e.g. payroll), online/real-time (processed immediately, e.g. ATM, ticket booking), time-sharing, distributed.

Data storage hierarchy (smallest to largest)

Level	Description	Example
Bit	0 or 1	1
Byte/Character	8 bits; one character	"A"
Field (data item)	Group of characters with one meaning	Name, Roll no.
Record	Related fields about one entity	One student's details
File	Collection of related records	Student file
Database	Integrated collection of related files	College database
A key field (primary key) uniquely identifies each record, e.g. employee ID.

Methods of organising data – file-oriented (traditional) approach: each application keeps its own files, causing redundancy, inconsistency and poor sharing. Database-oriented approach: data stored centrally and managed by a DBMS, reducing redundancy and improving integrity, security and sharing. Data can also be organised by subject, chronologically, alphabetically or by key value.

File types

• Master file – relatively permanent data (customer master, employee master); updated periodically.
• Transaction file – day-to-day changes used to update the master (sales, receipts).
• Output/report file – results for printing or display.
• Backup file – copy for recovery.
• Reference/table file – lookup data (tax rates, price lists).
• Work/scratch file – temporary during processing.
• Archive (history) file – old data kept for legal or historical purposes. Other classification: program files vs data files; text vs binary.

File organisation – how records are physically arranged on storage.

Organisation	How	Best for	Limitations
Sequential	Records stored one after another in key order	Batch processing with high activity (payroll); tape	Slow to find one record; updating needs whole file rewrite
Direct/Random (relative, hashed)	Key converted to address by a hashing function	Fast online access to individual records	Collisions; no ordered processing; needs direct-access device
Indexed sequential (ISAM)	Sequential file plus an index of keys and addresses	Both batch and random access	Index takes space; overflow areas slow it down
Indexed (multi-key)	Several indexes on different fields	Queries on many keys	Index maintenance
Hit rate / activity ratio = records accessed ÷ total records in a run – high activity favours sequential, low activity favours direct access. Volatility = frequency of additions/deletions.

Hashing example – address = key mod number of buckets; e.g. key 1234 with 100 buckets → 34. Two keys mapping to the same address cause a collision, handled by chaining or open addressing.

File utilities – routine programs supplied for file handling:

• Sorting – arrange records in ascending/descending key order.
• Merging – combine two or more sorted files into one sorted file.
• Copying – duplicate a file from one device to another (backup).
• Dumping – copy contents to printer/screen/another medium for inspection or recovery.
• Searching – locate records satisfying a condition.
• Maintenance – add, delete, modify records; reorganise files.

High-yield points

• Hierarchy: bit → byte → field → record → file → database.
• Primary key uniquely identifies a record.
• Master file = permanent data; transaction file = changes that update it.
• High hit rate favours sequential organisation; low hit rate favours direct access.
• ISAM = sequential file + index; hashing: address = key mod n.

Formulas

Hit rate $= \dfrac{\text{records accessed}}{\text{total records}}\times100\%$. Direct-address hashing: $\text{address} = \text{key} \bmod n$.

Worked example

A payroll file has 2000 records; in each monthly run 1900 are processed. Hit rate $= 1900/2000 = 95\%$ → sequential organisation is suitable. An enquiry system touching 20 records per session has hit rate 1 % → direct or indexed organisation suits better.

Exam tips

• Hierarchy order: bit → byte → field → record → file → database.
• Master file is permanent; transaction file holds changes that update it.
• Sequential organisation suits tape and high-hit-rate batch jobs.
• ISAM combines sequential storage with an index for random access.
• Merging requires the input files to be sorted first.

Exam-style questions

From Calmon's question bank, which has 25 questions on this topic.

1. In the database-oriented approach to organizing data, compared with the file-oriented approach:

• A. Data redundancy is always increased
• B. Programs must know the physical location of every record
• C. Data are integrated and shared by many applications under the control of a DBMS
• D. Each application keeps its own separate files

Answer: C. The database approach stores data centrally and shares it through a DBMS, reducing redundancy and inconsistency.

2. A copy of a file kept so that the data can be restored if the original is lost or damaged is called a:

• A. Master file
• B. Index file
• C. Backup file
• D. Transaction file

Answer: C. Backup files are duplicate copies, often stored on separate media, used for recovery.

Practise ASoE01 questions →