Distributed Operating Systems - Important Questions for Supply Exam (Easy Language)
Q2. Lamports Logical Clock
Lamport's Logical Clock is a method used to order events in a distributed system. Since there is no
global clock shared between systems, each process uses its own logical clock (a counter) to keep
track of events in order.
Rules:
1. Every event in a process increases the local clock by 1.
2. When a process sends a message, it includes its clock value.
3. The receiver updates its clock to the maximum of its own clock and the received clock, then adds
1.
Why Its Useful:
- Helps maintain the order of events in distributed systems.
- Ensures consistency during communication.
Diagram Required: Yes (to show message passing between processes with timestamp).
Q3. What is Distributed OS? Explain Advantages
A Distributed Operating System manages a group of independent computers and makes them
appear as a single unified system to users.
Advantages:
1. Resource Sharing - Printers, files, and devices can be shared across the network.
2. High Reliability - Even if one machine fails, the system continues.
3. Scalability - You can easily add more machines to improve performance.
4. Better Performance - Tasks can be split and run in parallel across systems.
5. Transparency - Users feel they are working on a single system.
Diagram Required: Optional (for system architecture).
Q4. Limitations of Distributed Systems
Despite their benefits, distributed systems also have limitations:
1. System Complexity - Difficult to design, program, and maintain.
2. Security Risks - Data travels over the network, so needs protection.
3. Network Dependency - Entire system may slow down if the network is weak.
4. Synchronization Issues - Difficult to maintain time and data consistency.
5. Cost - Setup and maintenance can be expensive.
Q5. Non-Token Based Algorithms
These algorithms do not use a physical token. Instead, they use message passing to control access
to the critical section.
Example: Ricart-Agrawala Algorithm
- When a process wants to enter the critical section (CS), it sends request messages to all other
processes.
- Other processes reply if they are not interested or if the request has a higher priority.
- The process enters CS when it receives replies from all others.
Use Case: Used where the number of messages is not a big concern.
Diagram Required: Yes (message exchange between processes).
Q6. Ricart-Agrawala Algorithm
This is a famous non-token based mutual exclusion algorithm.
Steps:
1. Process sends a REQUEST message with a timestamp to all other processes.
2. Other processes reply with OK if:
- They are not in CS, or
- Their request has a later timestamp.
3. When the requesting process gets replies from all, it enters the CS.
4. After using CS, it sends REPLY to waiting processes.
Important Points:
- Ensures mutual exclusion.
- Uses timestamp comparison for fairness.
Diagram Required: Yes (REQUEST and REPLY flow).
Q7. Token Based Algorithms
These algorithms use a special message called a token. Only the process holding the token can
enter the critical section.
Working:
- Initially, one process holds the token.
- After using the critical section, the token is passed to the next process that needs it.
Advantages:
- Fewer messages compared to non-token methods.
- No starvation if token is passed in order.
Diagram Required: Yes (show token passing between nodes).
Q9. Deadlock Handling Strategies
Deadlock happens when processes wait for each other forever. Distributed OS uses the following
strategies:
1. Deadlock Prevention - Design system so that deadlock never occurs (e.g., no circular wait).
2. Deadlock Avoidance - System uses algorithms like Bankers Algorithm to avoid unsafe states.
3. Deadlock Detection and Recovery - Allow deadlock but detect and fix it (kill or restart processes).
4. Ignore the Problem - In some systems, deadlocks are rare and are ignored.
Diagram Required: Optional (for resource allocation graph).
Q13. Multiprocessor OS Structure and Design Issues
Multiprocessor OS manages systems with multiple CPUs.
Design Issues:
1. Scheduling - How to assign tasks to processors.
2. Communication - How CPUs communicate and share data.
3. Synchronization - Control access to shared memory.
4. Resource Sharing - Printers, I/O, and devices shared among CPUs.
5. Load Balancing - Equal distribution of tasks among processors.
Diagram Required: Yes (multiprocessor structure).
Q16. Distributed File System (DFS)
DFS allows users to access and store files across a network as if they were local.
Features:
1. Transparency - Access files without knowing their physical location.
2. Reliability - Multiple copies of files are maintained.
3. Concurrency - Many users can access files at once.
4. Fault Tolerance - System works even if one server fails.
5. Scalability - Supports large number of files and users.
Diagram Required: Optional (DFS architecture).
Q20. Distributed Shared Memory (DSM)
DSM allows processes running on different machines to share memory over the network.
How It Works:
- Memory is divided into pages or blocks.
- When a process accesses a page, it is fetched from the node that owns it.
Algorithm Example: Page-based DSM
- Data is transferred between nodes on page faults.
- Maintains consistency using copy-on-write or invalidation techniques.
Issues:
1. Data Consistency
2. Synchronization
3. Communication Overhead
Diagram Required: Yes (shared memory layout).
End of Notes Enough for 40+ Marks in Exam
Tip: Write in your own words, draw diagrams neatly, and underline headings in the exam!
