Operating Systems

Unit 8: File Management & Device Management

From file allocation strategies to disk scheduling algorithms — master storage management with full numerical problems and start earning through Linux administration.

⏱️ Time to Complete: 7 hrs theory + 5 hrs lab  |  💰 Earning Potential: ₹5,000–₹15,000/month  |  📝 30 MCQs (Bloom's Mapped)

💼 Jobs this unlocks: Linux Admin (₹4–8 LPA)  |  Storage Engineer (₹6–12 LPA)

Section A

Opening Hook — How IRCTC Handles 30 Million Ticket Records Daily

🚂 When Disk Scheduling Goes Wrong During Tatkal

IRCTC stores 30 million ticket records daily across distributed file systems. During the Tatkal rush window, up to 15 lakh tickets are booked within 30 minutes. That's 8,333 bookings per second, each requiring multiple disk reads and writes — checking seat availability, updating reservation charts, generating PNR numbers, and logging transactions.

When disk scheduling goes wrong during Tatkal, the entire booking system slows down. Queues pile up, users get timeout errors, and thousands of passengers miss their train bookings. The difference between a well-optimised SSTF scheduler and a naive FCFS approach? Up to 10× difference in response time.

Behind the scenes, IRCTC's storage engineers use optimised file allocation strategies and disk scheduling algorithms to ensure data is read and written at maximum speed. Every millisecond of disk head movement matters when 25,000 concurrent users are hitting the server simultaneously.

Could YOU optimise IRCTC's disk access? Could you calculate the exact seek time for different scheduling algorithms and choose the best one? That's exactly what this unit teaches you — with full worked numericals you'll see in your exams.

🇮🇳 IRCTC🇮🇳 TCS🇮🇳 Infosys🇮🇳 Google IndiaNetAppDell EMC
A single enterprise hard disk can have over 500 billion sectors. The disk scheduling algorithm decides which sector to read next — a wrong choice can make the system 10× slower. The Linux kernel's default I/O scheduler has changed 4 times in 20 years, from Linus Elevator → Deadline → CFQ → mq-deadline, each time optimising disk head movement patterns.
Section B

Learning Outcomes — Bloom's Taxonomy Mapped

Bloom's LevelLearning Outcome
🔵 RememberList file attributes, operations, and types; define contiguous, linked, and indexed allocation methods
🔵 UnderstandExplain how directory structures differ and why tree-structured is most common in modern operating systems
🟢 ApplyCalculate total seek time for FCFS, SSTF, SCAN, C-SCAN, LOOK, and C-LOOK disk scheduling algorithms
🟢 AnalyzeCompare allocation methods and determine which suits SSD vs HDD storage scenarios
🟠 EvaluateAssess disk scheduling algorithms for fairness, throughput, and starvation; recommend the best for given workloads
🟠 CreateDesign a Python-based disk scheduling simulator that visualises head movement and compares algorithms