Scribd
Upload
13 views2 pages

M2 - Modulebank

The document outlines various questions related to parallel and distributed computing, focusing on middleware, race conditions, GPU processing, data replication, cache coherence, system stability, thread management, fault tolerance in HDFS, e-commerce consistency, and the impact of node failures in distributed systems. Each question requires explanations, justifications, designs, and analyses of different mechanisms and techniques to enhance performance, reliability, and efficiency in distributed environments. The content emphasizes practical applications and theoretical concepts in the context of online banking, image processing, and big data workloads.

Uploaded by

Gnaneshwar Reddy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
13 views2 pages

M2 - Modulebank

The document outlines various questions related to parallel and distributed computing, focusing on middleware, race conditions, GPU processing, data replication, cache coherence, system stability, thread management, fault tolerance in HDFS, e-commerce consistency, and the impact of node failures in distributed systems. Each question requires explanations, justifications, designs, and analyses of different mechanisms and techniques to enhance performance, reliability, and efficiency in distributed environments. The content emphasizes practical applications and theoretical concepts in the context of online banking, image processing, and big data workloads.

Uploaded by

Gnaneshwar Reddy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

PARALLEL & DISTRIBUTED COMPUTING

PART-A

Q-1

A. Identify and explain key security concerns when implementing middleware in an online
banking system.
B. Justify how the role of middleware in maintaining ACID (Atomicity, Consistency, Isolation,
Durability) properties.
C. Design a Middleware Solution: Suppose you are developing a distributed system
for an online banking application. Apply middleware models to design a secure and
efficient communication mechanism between multiple bank branches.

Q-2
A. Describe the concept of race conditions and explain how they can lead to data
inconsistency in multi-user environments.
B. Justify how timestamp-based mechanisms manage concurrent transactions and
prevent anomalies like lost updates and dirty reads.
C. Process Synchronization Mechanism: Design a scenario where multiple users are
trying to access and update a shared database in a distributed system, apply different
synchronization techniques to prevent race conditions.

Q-3

A. Describe how GPUs handle parallelism differently from traditional CPUs and why
they are preferred for large-scale image processing.
B. Analyze the speedup achieved by parallel execution of image transformations
compared to sequential CPU-based methods.
C. Design and implement a parallel algorithm using CUDA for processing large-scale
image datasets. Explain how your solution improves performance compared to
traditional CPU-based processing

Q-4

A. Design a system that uses data replication techniques such as primary-backup or


quorum-based replication to ensure high availability despite node failures.
B. Implement a check pointing mechanism where system states are periodically
saved so that computations can resume from the last checkpoint after a failure.
C. Analyze the overhead introduced by replication, check pointing, and recovery
mechanisms and their effect on overall system latency and throughput.

Q-5
A. Implement a directory-based cache coherence mechanism that minimizes
unnecessary coherence traffic and reduces memory access delays.
B. Analyze how the proposed protocol reduces cache miss rates and improves CPU
performance.
C. Propose a novel hybrid approach that dynamically switches between snooping and
directory-based coherence depending on workload characteristics.

Q-6
A. Justify how techniques like rate limiting, priority queues, and load balancing help
maintain system stability.
B. Design a message-passing protocol that leverages message queues to balance load
and prevent bottlenecks in a high-speed distributed environment.
C. Designing a Distributed Messaging System: Create a new message-passing protocol
for high-speed communication in a distributed computing environment. Explain
how it ensures low latency and high reliability.
Q-7
A. Develop a thread pool mechanism where multiple worker threads pick up tasks from
a shared queue to balance the workload effectively.
B. Analyze how different task allocation strategies influence response time and
resource utilization.
C. Building a POSIX Thread-Based Application: Create a multithreaded application
using POSIX threads that demonstrates load balancing in a distributed system.
Provide a sample code structure.
Q-8
A. Explain how HDFS ensures fault tolerance and load distribution to handle massive
datasets efficiently.
B. Develop a scenario where YARN dynamically allocates resources to improve
parallel execution efficiency in a large-scale cluster.
C. Scalability Analysis: Assess the scalability of Apache Hadoop in handling big data
workloads. What are its strengths and limitations when compared to other
distributed frameworks?
Q-9
A. Describe why some e-commerce functionalities (e.g., real-time payments) require
strong consistency while others (e.g., product listings) can tolerate eventual
consistency.
B. Apply sharding techniques and evaluate how they affect consistency, latency, and
scalability in a global e-commerce database.
C. Evaluating Consistency Models: In a distributed database system, evaluate different
consistency models (strong consistency, eventual consistency, etc.) and justify
which model is most suitable for an e-commerce platform.
Q-10
A. A distributed system consists of multiple nodes communicating over a network. If
one node fails, how does this impact system performance and reliability?
B. Implement an efficient synchronization mechanism to ensure consistency
while minimizing delays in distributed communication.
C. Design a middleware model that enhances communication efficiency between
heterogeneous distributed systems.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy