JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD B.Tech III ADVANCED OPERATING SYSTEMS Previous Question Paper (August - 2022 - R18)

A distributed operating system is a software system that manages a collection of independent, physically separate, and heterogeneous nodes connected via a network, presenting them to the user as a single, coherent computing system. It hides the underlying complexity of resource distribution and network communication.

Characteristics:

• Transparency: Hides the distributed nature of the system (location, migration, replication, and concurrency transparency).
• Resource Sharing: Allows users to access remote resources (printers, files, CPUs) as if they were local.
• Scalability: The ability to add more nodes to the system without significant performance degradation.
• Reliability and Availability: If one node fails, the system continues to function by rerouting tasks to other available nodes.

Advantages:

• Improved Performance: Tasks can be distributed across multiple processors to reduce execution time.
• Fault Tolerance: Redundancy ensures the system remains operational despite individual component failures.
• Incremental Growth: Systems can be expanded modularly by adding new nodes.

Communication primitives are the fundamental mechanisms used for inter-process communication (IPC) in distributed systems, typically implemented via message passing.

Blocking Communication (Synchronous): In this mode, the sender process is suspended until the message has been received by the receiver or acknowledged. Similarly, a blocking receive waits until a message arrives. This ensures strict synchronization between processes but can lead to performance bottlenecks and potential deadlocks if not managed carefully.

Non-blocking Communication (Asynchronous): In this mode, the sender process continues its execution immediately after dispatching the message without waiting for an acknowledgment. A non-blocking receive allows the receiver to check for a message and proceed if none is available. This increases concurrency and system throughput by decoupling the sender and receiver, though it requires more complex buffer management and status checking mechanisms to ensure data integrity.