CS3551 Distributed Computing Previous Year Question Papers - Anna University

UNIT I: Introduction

Part A (2 Marks)

• Define a Distributed System and its features.
• Differentiate between Message Passing and Shared Memory systems.
• What are the various types of Transparency in distributed systems?.
• Define Synchronous and Asynchronous execution.

Part B (13/16 Marks)

• Design Issues and Challenges: Detailed discussion of factors like heterogeneity, openness, security, and scalability.
• Model of Distributed Computations: Explain distributed programs, execution models, and communication network models.
• Communication Primitives: Explain blocking, non-blocking, synchronous, and asynchronous primitives in detail.

UNIT II: Logical Time and Global State

Part A (2 Marks)

• What is Clock Skew and Clock Drift?.
• Define the "Happen-Before" relation.
• What is the significance of Scalar Time and Vector Time?.

Part B (13/16 Marks)

• Clock Synchronization: Detailed explanation of the NTP framework and physical clock synchronization.
• Message Ordering: Discuss message ordering paradigms such as Causal Order and Total Order.
• Snapshot Recording: Explain the Chandy-Lamport Algorithm for global state detection in FIFO channels.

UNIT III: Distributed Mutex and Deadlock

Part A (2 Marks)

• Name the requirements of a Distributed Mutual Exclusion algorithm.
• What is a Phantom Deadlock and a False Deadlock?.
• Contrast the AND model and the OR model of deadlocks.

Part B (13/16 Marks)

• Mutual Exclusion Algorithms: Detailed study of Lamport's, Ricart-Agrawala's, and Suzuki-Kasami's Broadcast algorithms.
• Deadlock Detection: Explain the Chandy-Misra-Haas Algorithm for AND and OR models.
• Deadlock Handling: Strategies for prevention, avoidance, and resolution in distributed systems.

UNIT IV: Consensus and Recovery

Part A (2 Marks)

• Define the Consensus Problem and Byzantine Agreement.
• What is a Checkpoint and a Recovery Line?.
• Define the "Domino Effect" in failure recovery.

Part B (13/16 Marks)

• Consensus Algorithms: Discuss agreement in failure-free and failure-prone synchronous/asynchronous systems.
• Checkpoint-Based Recovery: Detailed explanation of Coordinated (e.g., Koo-Toueg) and Asynchronous checkpointing algorithms.
• Log-Based Recovery: Concepts and definitions related to failure recovery.

UNIT V: Cloud Computing

Part A (2 Marks)

• Define Cloud Computing and its core characteristics.
• Differentiate between Scalability and Elasticity in cloud contexts.
• What is a Hypervisor and its role in Virtualization?.

Part B (13/16 Marks)

• Service and Deployment Models: Comparative study of IaaS, PaaS, SaaS, and public/private/hybrid cloud models.
• Virtualization and Load Balancing: Role of virtualization in scalability and various load-balancing algorithms.
• Cloud Services: Detailed look at compute (e.g., EC2), storage (e.g., S3), and application services (e.g., Azure/AWS).

Most Repeated / High-Weight Questions

Mutual exclusion algorithms, deadlock detection and prevention, clock synchronization, Chandy-Lamport snapshot algorithm, checkpointing and recovery mechanisms, cloud service models.