Secondary Storage Management**

🧠 What is Secondary Storage?

Secondary Storage refers to non-volatile, long-term data storage devices like:
Hard Disk Drives (HDDs)
Solid State Drives (SSDs)
Magnetic Tapes
Optical Discs (CDs/DVDs)

📌 It stores data permanently, even after the system is turned off, unlike primary memory (RAM).

🔧 Need for Secondary Storage Management

To store large amounts of data not currently in use.
To allow efficient retrieval, updating, and deletion of data.
To manage the limited space efficiently.
To support file systems, backups, and virtual memory.

🔄 Responsibilities of the OS in Secondary Storage Management

Responsibility	Description
✅ Disk Scheduling	Efficient ordering of read/write requests to minimize disk head movement
📂 File Allocation	Managing where files are stored on disk (contiguously or scattered)
🧹 Free Space Management	Keeping track of used and unused disk blocks
🛡️ Protection	Preventing unauthorized access to stored data
🗃️ Directory Management	Maintaining directory structure for file organization
♻️ Storage Reclamation	Reusing disk space freed by deleted files

💽 Disk Structure

Disk = multiple platters coated with magnetic material
Each platter has:
Tracks (concentric circles)
Sectors (divisions of each track)
Data is read/written by a moving head

📊 Access Time Components

Term	Description
Seek Time	Time to move the read/write head to the correct track
Rotational Delay	Time for disk to rotate and position the desired sector
Transfer Time	Time to actually read or write data

📌 Disk Scheduling Algorithms

(Used to manage multiple read/write requests efficiently)

Algorithm	Description
FCFS (First Come First Serve)	Serve requests in the order they arrive
SSTF (Shortest Seek Time First)	Choose the request closest to current head
SCAN (Elevator Algorithm)	Move head in one direction, then reverse
C-SCAN (Circular SCAN)	Like SCAN, but returns directly to start
LOOK & C-LOOK	Similar to SCAN/C-SCAN but stops at last request, not disk end

🧩 Free Space Management Techniques

Method	Description
Bit Map	Use bits (0/1) to represent free/used blocks
Linked List	Keep a list of free blocks
Grouping	Keep a list of block groups
Counting	Track starting block and number of free blocks in sequence

📂 File Allocation Methods (how files are stored on disk)

Method	Description
Contiguous	File occupies a set of consecutive blocks
Linked	Each block points to the next block
Indexed	Use a separate index block to store all block addresses

✅ Advantages of Proper Secondary Storage Management

Better disk utilization
Faster read/write operations
Easier file retrieval and storage
Enhanced data security
Support for large-scale multitasking systems

✅ Summary

Key Area	Purpose
Disk Scheduling	Optimize performance of I/O operations
File Allocation	Efficiently manage storage space for files
Free Space Management	Track and reuse available disk blocks
Access Time Management	Reduce seek and latency delays
Protection & Security	Prevent unauthorized data access

🧩 Topic: Disk Structure

💽 What is Disk Structure?

The disk structure defines the physical and logical organization of data on a hard disk drive (HDD) or solid-state drive (SSD).

It includes:

How data is stored and accessed
How the disk is divided (tracks, sectors, cylinders)
How the OS interacts with disk components for I/O

🧱 Physical Structure of a Disk

A hard disk typically consists of the following:

1. Platters

Flat circular disks coated with magnetic material
Rotate around a spindle
Data is stored on both surfaces

2. Tracks

Concentric circles on the surface of each platter
Each track is divided into sectors

3. Sectors

The smallest unit of storage on a disk (commonly 512 bytes or 4 KB)
Each track contains multiple sectors

4. Cylinders

A set of tracks vertically aligned across platters (same track number on all platters)
Allows simultaneous access across platters

5. Read/Write Head

One head per surface
Mounted on a moving arm to access data on different tracks

6. Spindle

Rotates all platters together at high speed (e.g., 5400 or 7200 RPM)

🖼️ Diagram: Disk Structure Overview

           Top View of a Platter
         +------------------------+
         |   +----+----+----+    | ← Track 0, 1, 2...
         |   | T0 | T1 | T2 |    |
         |   +----+----+----+    |
         +------------------------+
                  ↑
              Read/Write Head

        Side View: Multiple Platters

     Platter 0:  T0   T1   T2   ...  ← Track
     Platter 1:  T0   T1   T2   ...
                 ↑
             Cylinder (all T1s)

⚙️ Disk Addressing: CHS vs LBA

Method	Description
CHS (Cylinder-Head-Sector)	Old method using physical geometry (rarely used now)
LBA (Logical Block Addressing)	Modern method – assigns unique number to each sector

⏱️ Disk Access Time

Component	Description
Seek Time	Time to move head to correct track (avg: 2–10 ms)
Rotational Delay	Time for sector to rotate under head (avg: 3–5 ms)
Transfer Time	Time to read/write data (depends on RPM and size)

🧮 Total Access Time = Seek Time + Rotational Delay + Transfer Time

📊 Summary Table: Disk Components

Component	Function
Platters	Store data magnetically
Tracks	Concentric circles storing data
Sectors	Smallest storage unit
Cylinders	All tracks aligned vertically across platters
R/W Head	Reads/writes data from/to disk
Arm Assembly	Moves heads across tracks
Spindle	Rotates the platters

✅ Key Points for Exams

Track: circular path on disk
Sector: subdivision of a track
Cylinder: same track number across all platters
Seek Time: time to move head
Rotational Delay: wait for sector under head
Transfer Time: time to read/write

🧩 Topic: Free Space Management

🧠 What is Free Space Management?

Free space management is the process of tracking all the unallocated (free) blocks or sectors on the disk so that:

New files can be allocated space
Deleted files can free up their space for reuse

📌 It helps the OS manage disk storage efficiently and avoid fragmentation.

🎯 Objectives

Keep track of which disk blocks are free
Allocate space for new files or growth
Deallocate space when files are deleted

🧱 Techniques for Free Space Management

🔹 1. Bit Map (Bit Vector)

A bit array is used to represent each block on the disk:
1 → block is allocated
0 → block is free

✅ Advantages:

Simple and compact
Easy to find contiguous free blocks

❌ Disadvantages:

Needs memory to store bit map (1 bit per block)
Slow for large disks (scanning required)

🔍 Example:

For 8 blocks: 0 1 1 0 0 1 0 0 ⇒ Blocks 0, 3, 4, 6, 7 are free

🔹 2. Linked List (Free List)

All free blocks are linked together using pointers
Each free block contains a pointer to the next free block

✅ Advantages:

Easy to manage
No extra memory needed (uses block itself)

❌ Disadvantages:

Cannot quickly find multiple free blocks
Pointer overhead reduces usable space

🔹 3. Grouping

First free block contains addresses of n free blocks
One of those n blocks contains the next group of n free blocks, and so on

✅ Advantages:

Fast allocation of multiple free blocks
Better than simple linked list

❌ Disadvantages:

Slightly more complex than simple linked list

🔹 4. Counting

Keeps track of:
Starting block address
Number of contiguous free blocks

✅ Advantages:

Good for managing large contiguous blocks
Requires less memory

❌ Disadvantages:

May not be helpful in highly fragmented disks

🔍 Example:

[Start: 12, Count: 5] → Blocks 12 to 16 are free

📊 Comparison Table

Method	Space Efficiency	Search Speed	Suitable For
Bit Map	High	Medium	Contiguous allocation
Linked List	Low (pointer overhead)	Low	Small files
Grouping	Medium	High	Bulk allocation
Counting	High	High (for contiguous)	Large free blocks

✅ Summary

Concept	Key Point
Free space management	Tracks unallocated disk blocks
Bit map	Uses bits (0 = free, 1 = used)
Linked list	Each free block links to the next
Grouping	Groups of free block addresses stored together
Counting	Stores start address and number of free blocks

🧩 Topic: File Allocation Methods

🧠 What is File Allocation?

File Allocation refers to the way the blocks on disk are assigned to files so that:

File data can be stored and accessed efficiently.
The operating system knows where a file’s contents are located.

🎯 Objectives of File Allocation

Maximize disk space utilization
Minimize file access time
Support for growing files
Handle fragmentation efficiently

📦 Main File Allocation Methods

Method	Key Idea	Suitable For
1. Contiguous	Store file in a single continuous block	Simple & fast access
2. Linked	Store blocks anywhere; link using pointers	Dynamic file size
3. Indexed	Use index block to store all addresses	Random access

🔹 1. Contiguous Allocation

Each file occupies consecutive disk blocks.

🧱 Structure:

File A → Blocks 5, 6, 7, 8

✅ Advantages:

Simple and fast (direct access)
Excellent read performance

❌ Disadvantages:

External fragmentation
Difficult to grow file if adjacent blocks are full
Wastes space if reserved blocks are unused

🔹 2. Linked Allocation

Each file is a linked list of disk blocks, which may be scattered on disk.
Each block stores data + pointer to next block.

🧱 Structure:

File B → Block 2 → Block 10 → Block 17 → null

✅ Advantages:

No external fragmentation
File size can grow easily

❌ Disadvantages:

Sequential access only
Pointer overhead in each block
Can be unreliable if pointer is lost

🔹 3. Indexed Allocation

Uses a special index block to store all block addresses of the file.
Supports direct access to any block.

🧱 Structure:

Index Block (File C) → [3, 7, 11, 14]
→ File data in: Block 3, Block 7, Block 11, Block 14

✅ Advantages:

Supports random access
No external fragmentation
Easy to grow file

❌ Disadvantages:

Overhead of index block
For large files, may need multi-level indexing

📊 Comparison Table

Feature	Contiguous	Linked	Indexed
Access Type	Direct & Sequential	Sequential only	Direct & Sequential
File Growth	Difficult	Easy	Easy
Fragmentation	External	None	None
Overhead	Low	Pointer per block	Index block(s)
Access Speed	Fast	Slow	Medium–Fast

✅ Summary

Contiguous: Simple and fast, but causes fragmentation.
Linked: Flexible, but slower and no direct access.
Indexed: Supports random access, but has indexing overhead.

🧩 Topic: Disk Scheduling

🧠 What is Disk Scheduling?

Disk Scheduling refers to the method of choosing the order in which pending disk I/O requests are served.

Since disk head movement is slow, the goal is to:

Minimize seek time (head movement)
Improve throughput
Reduce latency

⚙️ Disk Access Time = Seek Time + Rotational Delay + Transfer Time

Since seek time is the most time-consuming, disk scheduling mainly focuses on minimizing head movement.

📦 Types of Disk Scheduling Algorithms

Algorithm	Principle	Best For
FCFS	Serve requests in arrival order	Simple & fair
SSTF	Serve closest request (min seek time)	High efficiency
SCAN	Move head in one direction, then reverse	Heavy loads
C-SCAN	Only one-direction scan, jump to start	Uniform service
LOOK	Like SCAN but only go as far as needed	Less movement
C-LOOK	Like C-SCAN, optimized movement	Balanced performance

🔹 1. FCFS (First-Come, First-Served)

Serve requests in the order they arrive.
Simple, fair, but can be inefficient.

🔍 Example:

Requests: 98, 183, 37, 122 Head at: 100

Total Head Movement = |100–98| + |98–183| + |183–37| + |37–122| = 345 tracks

✅ Fair ❌ Long seek times

🔹 2. SSTF (Shortest Seek Time First)

Serve the nearest request to current head position.
Like a greedy algorithm.

🔍 Example:

Requests: 98, 183, 37, 122 Head at: 100

Closest to 100 → 98 → 122 → 183 → 37 Total movement = less than FCFS

✅ Efficient ❌ May cause starvation

🔹 3. SCAN (Elevator Algorithm)

Head moves in one direction, serves all requests, then reverses.
Like an elevator going up and down.

🔍 Head at 50, moving right

Requests: 34, 78, 90, 123, 10 → Serve: 78 → 90 → 123 → reverse → 34 → 10

✅ Reduces variance ❌ Longer wait at the ends

🔹 4. C-SCAN (Circular SCAN)

Moves in one direction (e.g., right), and jumps to beginning after reaching end.

🔍 Example:

→ 78 → 90 → 123 → jump → 10 → 34

✅ Uniform response time ❌ Unused movement during jump

🔹 5. LOOK

Like SCAN, but stops at the last request (not end of disk).

🔍 Example:

→ 78 → 90 → 123 → reverse → 34 → 10

✅ Saves time (no full scan)

🔹 6. C-LOOK

Like C-SCAN but only moves as far as last request, then jumps to the first.

🔍 Example:

→ 78 → 90 → 123 → jump → 10 → 34

✅ Efficient version of C-SCAN

📊 Comparison Table

Algorithm	Seek Time	Fairness	Starvation	Directional
FCFS	High	High	No	No
SSTF	Low	Low	Yes	No
SCAN	Medium	Medium	No	Yes
C-SCAN	Medium	High	No	Yes
LOOK	Low	Medium	No	Yes
C-LOOK	Low	High	No	Yes

✅ Summary

FCFS: Simple, but may cause long delays.
SSTF: Efficient but may starve far requests.
SCAN / LOOK: Balanced, like elevator.
C-SCAN / C-LOOK: Uniform service for all tracks.

🧩 Topic: Performance and Reliability Improvements in Disk Systems

🎯 Objective

Modern operating systems and storage hardware apply various techniques to improve:

📈 Performance (speed of data access and transfer)
🔒 Reliability (safety and consistency of stored data)

⚙️ 1. Performance Improvements

🔹 A. Disk Caching

Uses fast memory (RAM) as a buffer for frequently accessed data.
Reduces physical disk access.
Improves read/write speed dramatically.

✅ Example: OS caches file system metadata in RAM for faster access.

🔹 B. Read-Ahead and Write-Behind

Read-Ahead: OS predicts future data use and preloads blocks into cache.
Write-Behind: Delays writes to group multiple writes together.

✅ Reduces disk I/O ✅ Boosts throughput

🔹 C. Disk Scheduling Algorithms

Techniques like SSTF, SCAN, and LOOK reduce head movement.
Optimize the order of disk access for minimal seek time.

🔹 D. Disk Striping (RAID-0)

Splits data across multiple disks for parallel read/write operations.

✅ Increases speed, especially for large files ❌ No redundancy (not reliable)

🔹 E. Solid-State Drives (SSD)

Use flash memory, no moving parts.
Faster than HDDs in access time and data transfer.

✅ High-speed storage ✅ No mechanical delays

🛡️ 2. Reliability Improvements

🔹 A. Redundant Array of Independent Disks (RAID)

RAID = group of multiple disks working together to improve reliability & performance

Level	Feature	Benefit
RAID 1	Disk Mirroring	Full data redundancy
RAID 5	Striping with Parity	Balanced reliability and speed
RAID 6	Dual Parity	Can survive 2 disk failures

🔹 B. Disk Mirroring (RAID-1)

Data is written to two identical disks.
If one fails, the other has a complete copy.

✅ Ensures data safety ❌ Expensive (requires double storage)

🔹 C. Journaling File Systems

File systems like ext4, NTFS, and XFS maintain a log (journal) of changes.
In case of a crash, the system can recover data from the journal.

✅ Prevents data corruption ✅ Fast recovery after crashes

🔹 D. ECC (Error Correction Code)

Used to detect and correct errors in data stored on disk.
Ensures data integrity.

🔹 E. Backups and Snapshots

Backup: Copy of data stored elsewhere for recovery.
Snapshot: Point-in-time copy of disk state (used in cloud systems and databases)

✅ Ensures recoverability ✅ Minimizes data loss

📌 Summary Table

Category	Technique	Purpose
Performance	Disk Caching, SSDs	Fast data access
	Scheduling Algorithms	Reduce seek time
	RAID-0 (Striping)	Faster read/write
Reliability	RAID-1, RAID-5/6	Data redundancy
	Journaling File Systems	Crash recovery
	ECC, Backups	Error correction & data safety

✅ Key Takeaways

Performance is enhanced by reducing mechanical delays and caching.
Reliability is improved through redundancy, journaling, and backups.
Modern systems often combine performance and fault-tolerance techniques.

🧩 Topic: Storage Hierarchy

🧠 What is Storage Hierarchy?

The storage hierarchy refers to the organization of different types of memory and storage devices in a computer system, based on:

Speed
Cost
Volatility
Capacity
Access time

📌 It is designed to balance performance and cost by storing frequently used data in faster (but expensive) memory and infrequently used data in slower (but cheaper) memory.

🏛️ Structure of Storage Hierarchy

   ↑   Faster, More Expensive, Smaller
   |   
   |   CPU Registers
   |   Cache Memory (L1, L2, L3)
   |   Main Memory (RAM)
   |   Solid-State Drive (SSD)
   |   Hard Disk Drive (HDD)
   |   Optical Disks (CD/DVD)
   |   Magnetic Tapes
   ↓   Slower, Cheaper, Larger

📊 Hierarchy Table

Level	Storage Type	Access Time	Cost per Bit	Volatile	Capacity
1	CPU Registers	1 ns	Very High	Yes	Very Low
2	Cache (L1/L2/L3)	1–10 ns	High	Yes	Low
3	Main Memory (RAM)	10–100 ns	Medium	Yes	Moderate
4	SSD	0.1–1 ms	Low	No	High
5	HDD	1–10 ms	Very Low	No	Very High
6	Optical/Magnetic	Seconds	Very Low	No	Huge

🎯 Key Characteristics

Term	Description
Volatile	Data is lost when power is off (e.g., RAM)
Non-Volatile	Data is retained without power (e.g., HDD, SSD)
Access Time	Time taken to read/write data
Cost per Bit	Cost of storing a single bit (decreases down the hierarchy)
Capacity	Amount of data that can be stored (increases down the levels)

🔁 Why Use a Storage Hierarchy?

Fast memory (like cache) is expensive and small.
Slow memory (like HDD) is cheap and large.
Combining different types of memory:
Keeps hot data close to the CPU
Moves cold data to slower storage

✅ Achieves high performance at low cost

📌 Examples in Practice

CPU uses cache for recently used instructions
OS uses RAM for active programs
Virtual memory uses HDD/SSD as an extension of RAM
Backup data is stored in magnetic tapes or external drives

✅ Summary

Feature	Upper Levels (e.g., Cache, RAM)	Lower Levels (e.g., HDD, Tape)
Speed	Very Fast	Slow
Cost	High	Low
Volatility	Volatile	Non-Volatile
Capacity	Low	High