NFS – NOTES FOR STUDY

Important Exam Questions for Parallel and Distributed Systems

Unit I: Fundamentals of Parallel and Distributed Systems

2-Mark Questions

• Define parallel computing and give one example of a parallel application.

• Define distributed computing and describe a typical use-case.

• State two key differences between parallel and distributed computing.

• What is data parallelism? Give a brief example.

• What is task parallelism? Give a brief example.

• State Flynn’s Taxonomy and name its four classes.

• Define SIMD and MIMD architectures in brief.

• What is the difference between shared memory and distributed memory architectures?

• What is a process and how does it differ from a thread?

• Define the CAP theorem in the context of distributed systems.

• What does pipelining mean in computer architecture?

• Give two examples of applications that benefit from parallel processing.

• Name two challenges of designing distributed systems (e.g. fault tolerance,

synchronization).

5-Mark Questions

• Explain the differences between parallel and distributed systems. Provide examples to
illustrate each.

• Describe the main characteristics and goals of parallel and distributed computing.

• Outline Flynn’s taxonomy (SISD, SIMD, MISD, MIMD) with examples for each category.

• Compare shared memory and distributed memory models. What are the advantages
and disadvantages of each?

• Explain the need for parallelism in modern computing. Provide examples of

performance gains.
 NFS – NOTES FOR STUDY

• Define multithreading. How does it improve performance compared to single-threaded

execution?

• Explain the CAP theorem. Why is it important for distributed databases?

• Discuss the main security concerns in distributed systems (e.g. confidentiality, integrity,
availability).

• Describe the concept of pipelining in CPUs. Provide a simple diagram of an instruction

pipeline and explain its stages.

• Explain data parallelism vs task parallelism with examples of each.

• List and explain two types of parallel computing models (e.g. SIMD vs MIMD).

10-Mark Questions

• Compare and contrast parallel computing and distributed computing in detail. Discuss
their architectures, typical applications, and challenges.

• Draw a diagram of a client-server architecture and a peer-to-peer network. Explain how

each model works and list its advantages and disadvantages.

• Explain the architecture of a distributed system, including its components (nodes,

network, middleware). Discuss how communication is handled in distributed systems.

• Discuss the goals and applications of parallel and distributed computing. Why are they
important in areas like scientific computing, big data, or real-time systems?

• Explain Flynn’s taxonomy in depth. For each of SISD, SIMD, MISD, and MIMD, give a real-
world example architecture or processor.

• Describe the concept of synchronization in parallel computing. Why is synchronization

needed, and what problems (e.g. deadlock, race condition) can occur without it?

• Discuss the issues and challenges in parallel and distributed systems (such as
heterogeneity, latency, fault tolerance, security). How are these challenges typically
addressed?
 NFS – NOTES FOR STUDY

Unit II: Parallelism Models and Algorithms

2-Mark Questions

• Define data parallelism in parallel algorithms.

• Define control parallelism (instruction-level parallelism) with an example.

• What is Amdahl’s Law? Provide the formula.

• Define speedup and efficiency in the context of parallel computing.

• What is utilization in parallel system performance?

• What is a benchmark in parallel computing? Give an example of a parallel benchmark.

• Name two interconnection network topologies used in parallel processors (e.g. mesh,
hypercube).

• What is a shared memory multiprocessor?

• What is a distributed memory network?

• Define leader election in distributed systems. Why is it needed?

• What is mutual exclusion in a distributed system?

• What is the divide-and-conquer strategy? Give an example of a parallel algorithm that

uses it.

• Define load balancing in parallel computing.

5-Mark Questions

• Explain performance metrics in parallel computing: speedup, efficiency, and utilization.

Include formulas in your explanation.

• State and explain Amdahl’s Law. What does it imply about the limits of parallelization?

• Compare static and dynamic interconnection networks in parallel processors. Give

examples of each.

• Describe one interconnection network topology (e.g. hypercube or mesh) with a

diagram, and discuss its advantages and disadvantages.

• Outline the architecture of a shared memory multiprocessor system. How do processors

communicate?

• Outline the architecture of a distributed memory system. How is memory accessed?

 NFS – NOTES FOR STUDY

• Explain the basic approach to parallel matrix multiplication. How can the work be
divided among multiple processors?

• Describe the parallel merge sort algorithm. Why does it run faster on multiple
processors?

• Explain a simple leader election algorithm in a distributed system (for example, the Bully
algorithm or Ring algorithm).

• Explain a basic mutual exclusion algorithm for distributed systems (for example, using a
token ring).

• Discuss load balancing strategies: static vs dynamic scheduling. Provide an example

scenario for each.

10-Mark Questions

• Discuss laws of parallel performance (Amdahl’s Law and Gustafson’s Law) and their
significance. Compare and contrast them.

• Explain various interconnection network topologies used in parallel computers (bus,

ring, star, fully connected, mesh, hypercube, etc.) with diagrams. Discuss the trade-offs
of each topology.

• Describe in detail the architecture of a parallel processor system, including processor

organization and network of interconnections.

• Discuss parallel algorithms for matrix multiplication. Explain at least one algorithm
(such as block multiplication) in detail, including how tasks are assigned and results
combined.

• Describe different parallel sorting algorithms (e.g., parallel merge sort, parallel
quicksort) and compare their time complexities and use cases.

• Explain in detail the leader election problem in distributed systems. Compare two
algorithms (such as Bully vs Ring) and discuss their message complexity and drawbacks.

• Discuss mutual exclusion in distributed systems. Explain two algorithms (such as token-
based and permission-based) with diagrams and compare their performance.

• Describe load balancing and task scheduling in parallel systems. Compare static
scheduling (e.g. round-robin) vs dynamic scheduling (e.g. work-stealing), including
advantages and disadvantages.
 NFS – NOTES FOR STUDY

• Case Study: Explain how a parallel task (e.g. matrix multiplication or parallel search) is
implemented using MPI. Outline the steps and message exchanges involved.

Unit III: Advanced Topics and Applications

2-Mark Questions

• Define cloud computing.

• List two deployment models of cloud computing (e.g. Public vs Private).

• What is Infrastructure as a Service (IaaS)?

• What is Platform as a Service (PaaS)?

• What is Software as a Service (SaaS)?

• What is Function as a Service (FaaS)?

• What is Unified Communications as a Service (UcaaS)?

• Name two benefits of cloud computing.

• Name two limitations of cloud computing.

• Define confidentiality in the context of cloud security.

• Define integrity in the context of cloud security.

• Define availability in the context of cloud security.

• What is cloud analytics?

• What is Big Data?

• Define Hadoop.

• What is MapReduce?

• What is Apache Spark?

• What is Edge Computing?

• What is Fog Computing?

 NFS – NOTES FOR STUDY

• What is GPU (CUDA) computing?

• Give an example of a cloud service provider (e.g. Amazon Web Services).

5-Mark Questions

• Explain the differences between public, private, and hybrid clouds. Give an example
scenario for each.

• Discuss the main benefits and limitations of cloud computing.

• Describe the cloud service models: IaaS, PaaS, SaaS, FaaS. Give an example of each.

• Explain the roles of confidentiality, integrity, and availability (CIA) in cloud security.
Why is each important?

• Describe common data security challenges in the cloud (such as data breaches, insider
threats) and how cloud providers mitigate these risks.

• Explain the concept of secure cloud software development. What practices help ensure
security (e.g., encryption, access control, audits)?

• Outline the MapReduce programming model with a simple example (such as counting
word frequency in documents).

• Compare Hadoop MapReduce and Apache Spark: how they process big data differently
(mention batch vs in-memory processing).

• Explain what edge computing and fog computing are, and how they complement
traditional cloud computing.

• Discuss how GPUs accelerate parallel computation (briefly mention CUDA or a similar
platform).

10-Mark Questions

• Compare and contrast cloud deployment models (public, private, hybrid, community) in
detail, including real-world examples of each.

• Explain in depth the cloud service models (IaaS, PaaS, SaaS, UcaaS, FaaS). For each,
describe how it works and give a commercial example (e.g., AWS EC2 for IaaS, Google
App Engine for PaaS, Salesforce for SaaS).

• Discuss the advantages and limitations of cloud computing. How does cloud computing
impact scalability, cost, and security for businesses?
 NFS – NOTES FOR STUDY

• Explain cloud security principles in detail. Discuss how the CIA triad (confidentiality,
integrity, availability) is maintained in cloud environments (e.g., through encryption,
redundancy, access controls).

• Describe a cloud case study (e.g., Amazon Web Services or Google Cloud Platform).
Discuss its key services (compute, storage, networking) and how it handles reliability and
scalability.

• Explain the MapReduce and HDFS architecture (as used in Hadoop). Describe how a
large dataset is stored and processed in parallel across a cluster (include fault tolerance).

• Discuss Apache Spark in detail: its architecture, how it uses in-memory processing, and
why it can be faster than Hadoop MapReduce for certain tasks.

• Explain edge and fog computing thoroughly. Describe a scenario (like IoT data
processing) where fog computing is beneficial.

• Discuss GPU parallel computing (CUDA) in detail. How does CUDA enable massive
parallelism on graphics cards? Include an example of a computation accelerated by a
GPU.

• Case Study: Describe how parallel matrix multiplication can be implemented using
OpenMP (shared-memory) and compare it to an MPI (message-passing)
implementation. Discuss the differences in programming model and performance.

Join Now:- https://chat.whatsapp.com/CX7qycrvE2QIpfeBPBrbxe