
- DBMS - Home
- DBMS - Overview
- DBMS - Architecture
- DBMS - Data Models
- DBMS - Data Schemas
- DBMS - Data Independence
- DBMS - System Environment
- Centralized and Client/Server Architecture
- DBMS - Classification
- Relational Model
- DBMS - Codd's Rules
- DBMS - Relational Data Model
- DBMS - Relational Model Constraints
- DBMS - Relational Database Schemas
- DBMS - Handling Constraint Violations
- Entity Relationship Model
- DBMS - ER Model Basic Concepts
- DBMS - ER Diagram Representation
- Relationship Types and Relationship Sets
- DBMS - Weak Entity Types
- DBMS - Generalization, Aggregation
- DBMS - Drawing an ER Diagram
- DBMS - Enhanced ER Model
- Subclass, Superclass and Inheritance in EER
- Specialization and Generalization in Extended ER Model
- Data Abstraction and Knowledge Representation
- Relational Algebra
- DBMS - Relational Algebra
- Unary Relational Operation
- Set Theory Operations
- DBMS - Database Joins
- DBMS - Division Operation
- DBMS - ER to Relational Model
- Examples of Query in Relational Algebra
- Relational Calculus
- Tuple Relational Calculus
- Domain Relational Calculus
- Relational Database Design
- DBMS - Functional Dependency
- DBMS - Inference Rules
- DBMS - Minimal Cover
- Equivalence of Functional Dependency
- Finding Attribute Closure and Candidate Keys
- Relational Database Design
- DBMS - Keys
- Super keys and candidate keys
- DBMS - Foreign Key
- Finding Candidate Keys
- Normalization in Database Designing
- Database Normalization
- First Normal Form
- Second Normal Form
- Third Normal Form
- Boyce Codd Normal Form
- Difference Between 4NF and 5NF
- Structured Query Language
- Types of Languages in SQL
- Querying in SQL
- CRUD Operations in SQL
- Aggregation Function in SQL
- Join and Subquery in SQL
- Views in SQL
- Trigger and Schema Modification
- Storage and File Structure
- DBMS - Storage System
- DBMS - File Structure
- DBMS - Secondary Storage Devices
- DBMS - Buffer and Disk Blocks
- DBMS - Placing File Records on Disk
- DBMS - Ordered and Unordered Records
- Indexing and Hashing
- DBMS - Indexing
- DBMS - Single-Level Ordered Indexing
- DBMS - Multi-level Indexing
- Dynamic B- Tree and B+ Tree
- DBMS - Hashing
- Transaction and Concurrency
- DBMS - Transaction
- DBMS - Concurrency Control
- DBMS - Deadlock
- Backup and Recovery
- DBMS - Data Backup
- DBMS - Data Recovery
- DBMS Useful Resources
- DBMS - Quick Guide
- DBMS - Useful Resources
- DBMS - Discussion
DBMS - Secondary Storage Devices
Databases handle huge volumes of data that must be stored efficiently and accessed reliably. Database management systems use the primary memory (RAM) for generating speed while handling active operations, however RAMs are volatile and costly for large datasets. This is where the secondary storage devices come into the picture.
Secondary storage devices are non-volatile, provide higher capacities, and are quite cost-effective. These devices are essential for long-term data storage in DBMS. Read this chapter to get a good understanding of the types of secondary storage devices used in DBMS.
Secondary Storage Devices for Databases
For databases, we need stable storages. Secondary storage devices store data persistently. They are slower than primary memory but offer significantly larger storage capacities and are more affordable per byte. Data stored in secondary storage remains intact even when the system is turned off, ensuring data permanence.
Common secondary storage devices are mainly of two types â magnetic disks and magnetic tapes. Both these types of secondary storage devices have different use cases based on their performance and accessibility features.
Magnetic Disks
Magnetic Disks or Hard Disk Drives (HDD) are one of the most commonly used storage devices in DBMS. These devices are used for both personal and enterprise systems as they provide a perfect balance of performance, durability, and cost.
How Do Magnetic Disks Work?
Magnetic disks are circular plates coated with a magnetic material. Data is stored on these plates by magnetizing the areas on the disk that represent binary values of "0" or "1". Modern magnetic disks typically come in either single-sided or double-sided formats. Single-sided disks store data on one surface, while double-sided disks utilize both the surfaces for higher storage capacity.
Data Organization on Disks
The structure of magnetic disks is specially designed to maximize storage and facilitate fast access. Disks have three major parts that we need to consider −
- Tracks − Concentric circles on the surface of the disk where data is stored.
- Sectors − Divisions of a track that hold a fixed amount of data, usually 512 to 8192 bytes.
- Cylinders − Groups of tracks with the same diameter across multiple platters. Cylinders allow for faster data retrieval as the read/write head doesn't need to move between tracks.
A modern magnetic disk might implement Zone Bit Recording (ZBR). Here the tracks in different zones have varying numbers of sectors. This optimization techniques allows for higher storage density in the outer tracks without compromising performance.

Advantages of Magnetic Disks
Following are the advantages of using magnetic disks −
- Random Access − Magnetic disks, unlike sequential storage, allow direct access to specific data blocks, which makes them highly efficient for databases.
- High Capacity − Disks can store terabytes of data. They are suitable for modern applications with large datasets.
- Durability − Magnetic disks are designed to withstand repeated read and write operations.
Performance Example: Seagate Cheetah Disk
The Seagate Cheetah 15K.6 is a high-performance magnetic disk that illustrates the capabilities of modern storage. With a formatted capacity of 450 GB and rotational speeds of 15,000 rpm, it achieves an internal transfer rate of up to 2225 Mb/sec. Its average seek time is 3.4 ms for read operations, which makes it ideal for enterprise-level DBMS where speed is critical.
Magnetic Tape Storage: A Reliable Backup Solution
Magnetic tapes are another category of secondary storage devices. These devices are not so popular now, however they still remain indispensable for archival and backup purposes.
How Do Magnetic Tapes Work?
Magnetic tapes are sequential storage devices. Here the data is stored on long strips of magnetic material wound onto reels or cartridges. Accessing the data stored on magnetic tapes means scanning through previous blocks to reach the desired one. It is this property of sequential access that makes the tapes slower than disks for random data retrieval.
Characteristics of Magnetic Tapes
Given below are some of the important characteristics of magnetic tapes −
- High Storage Capacity − Modern magnetic tapes can store a very large volume of data, like hundreds of gigabytes per cartridge. It makes them suitable for large-scale backups.
- Cost-Effectiveness − Tapes are cheaper than disks, both in terms of initial cost and long-term storage.
- Sequential Access − Magnetic tapes are slower for specific data retrieval. They take longer time in reading or writing large continuous datasets.
Real-World Example: Sun Storage SL8500
The Sun Storage SL8500 is an example of such a magnetic tape. With a storage capacity of up to 70 petabytes and throughput rates reaching 193.2 TB/hour, this system is ideal for enterprises handling massive backups. Robotic arms and automatic labeling are used with them.
Applications of Magnetic Tapes
Given below are some of the important applications of magnetic tapes −
- Backup Storage − Tapes are essential tools for creating periodic backups of databases. They protect against disk failures and data corruption.
- Archiving − Magnetic tapes are ideal for storing and archiving historical or seldom-used data for future reference.
- Disaster Recovery − Tapes offer a reliable solution for recovering data in catastrophic scenarios.
Buffering in Secondary Storage
No doubt secondary storages have larger capacity, but their speed often lags behind that of primary memory devices. To bridge this gap, DBMS uses buffering techniques to optimize data transfer between disks and main memory.
- Double Buffering − Double buffering is a widely used technique where two buffers are alternated during data transfer. While one buffer is being filled with data from the disk, the other is processed by the CPU. This overlap ensures continuous data flow, reducing delays caused by waiting for disk operations.
- Impact on Performance − Double buffering improves the efficiency significantly when transferring multiple blocks of data. For example, when processing a database query that spans several disk blocks, the system can simultaneously read and process data, minimizing the idle time.
Access Times in Magnetic Disks
Consider the following three key steps while accessing data that is stored on magnetic disks −
- Seek Time − It's the time taken to position the read/write head over the correct track.
- Rotational Delay − The wait for the desired sector to rotate under the read/write head.
- Block Transfer Time − The time required to move the data block from disk to memory.
Example − A disk rotating at 15,000 rpm has an average rotational delay of 2 milliseconds. Now consider with seek time and block transfer time, the total delay can range from 9 to 60 milliseconds. While this is relatively fast, it will still be slower than the speeds achievable by the primary memory.
Applications of Secondary Storage in DBMS
Secondary storage devices are crucial for several DBMS operations −
- Online Storage − Magnetic disks handle active databases that require frequent read/write operations.
- Backup and Recovery − The data stored on magnetic tapes can be restored in case of failure or corruption.
- Archival Systems − Outdated but legally required records are stored on tapes to save on costs while meeting compliance requirements.
- Scalable Storage Solutions − Large-scale systems, like storage area networks (SANs), use secondary storage for managing enterprise-level databases.
Limitations of Secondary Storage
Despite their benefits, secondary storage devices face certain limitations −
- Latency − Access times are significantly slower compared to primary memory.
- Maintenance − Tapes require regular maintenance to ensure data integrity over time.
- Costs for High Performance − While basic storage is cheap, high-performance disks like the Seagate Cheetah can be costly.
Conclusion
In this chapter, we presented in detail the role of secondary storage devices in DBMS. Starting with an overview of the importance and features of secondary storage devices, we focussed on the organization of magnetic disks, including real-world examples like the Seagate Cheetah 15K.6.
In addition, we covered the use of magnetic tapes for backup and archival purpose, highlighting systems like the Sun Storage SL8500. Thereafter, we explored the buffering techniques and how they improve the performance of secondary storage devices. We finished the chapter by highlighting the applications and limitations of secondary storage devices in DBMS.