BSc: Distributed And Network Programming

Distributed and Network Programming

• Course name: Distributed and Network Programming
• Code discipline: XYZ
• Subject area: xxx

Short Description

This course covers the following concepts: Network programming concepts: Layered architecture, TCP and UDP sockets, multithreaded servers; Distributed systems concepts: system architecture, inter-process communication, remote procedure calls, peer-to-peer systems, coordination, replication, and fault tolerance..

Prerequisites

Prerequisite subjects

• Networks: 1) Understanding Application, Transport, and Network layers, 2) Basic socket programming experience
• Operating systems

Course Topics

Course Sections and Topics

Section	Topics within the section
Introduction to subject, computer networks basics, transport layer protocols, and socket programming	General introduction to the course Computer networks basic Socket programming UDP socket programming TCP socket programming
Multithreaded socket programming, RPCs, and distributed system architecture	Multithreading and multithreaded socket programming Remote procedure calls (RPCs) Distributed system architectures
Coordination, consistency, and replication in distributed systems	Clock synchronization algorithms (NTP, Berkeley) Logical clock (Lamport clocks) Mutual exclusion algorithms: permission-based, token-based Election algorithms: Bully, Ring Consistency models Replica management Consistency protocols
Fault tolerance and security in distributed systems	Intro to fault tolerance: Failure models, Failure masking by redundancy Process resilience: process groups, process replication, consensus in faulty systems, failure detection Reliable group communication: atomic multicast, Distributed commit Recovery: checkpointing Intro to security: threats, design issues, cryptography Secure channels: authentication, message integrity and confidentiality, secure group communication Access control: general issues, firewalls, secure mobile code, denial of service Secure naming Security management: Key management, secure group management, authorization management

Intended Learning Outcomes (ILOs)

What is the main purpose of this course?

Distributed and networked systems have become an integral part of our life, we use various applications such as chatting, online transactions, or cloud storage apps. All these popular applications are supported by an infrastructure (of servers) that is organized based on some concepts of distributed systems. The purpose of this course is to provide the students with the necessary concepts, models, and real-world problem-solving techniques of network programming and distributed systems.

ILOs defined at three levels

Level 1: What concepts should a student know/remember/explain?

By the end of the course, the students should be able to ...

• Concepts of network programming
• Different distributed system architectures
• Various synchronization and coordination techniques
• Different consistency models and replication methods
• Approaches to achieve fault tolerance and security in distributed systems

Level 2: What basic practical skills should a student be able to perform?

By the end of the course, the students should be able to ...

• Difference between different transport protocols, when and why one is preferred over another
• Difference between different distributed system architectures (centralized, decentralized, and hybrid)
• How a mutual exclusion is achieved between concurrent servers (centralized, distributed, token-ring, and decentralized)
• How a new leader is elected in peer-to-peer systems (bully, ring)
• How to achieve a consistent replicas across distributed systems (consistency models and protocols, content replication and placement)
• Some methods to achieve the fault tolerance in distributed systems

Level 3: What complex comprehensive skills should a student be able to apply in real-life scenarios?

By the end of the course, the students should be able to ...

• Building a custom application protocols on top of the existing transport protocols
• Writing multithreaded server and client apps with sockets
• Using RPC for inter-process communication: command execution, file transfer
• Building peer-to-peer systems with distributed protocol such as Chord
• Building fault-tolerant systems with failure detection and leader election

Grading

Course grading range

Grade	Range	Description of performance
A. Excellent	90-100	-
B. Good	75-89	-
C. Satisfactory	60-74	-
D. Poor	0-59	-

Course activities and grading breakdown

Activity Type	Percentage of the overall course grade
Laboratory assignments	55%
Final exam	35%
Attendance	10%

Resources, literature and reference materials

Open access resources

• Textbook: Maarten Van Steen, and Andrew S. Tanenbaum. Distributed systems (3rd Edition) Leiden, The Netherlands: Maarten van Steen, 2017. Available online: https://www.distributed-systems.net/
• Reference: George F. Coulouris, Jean Dollimore, and Tim Kindberg. Distributed systems: concepts and design (5th Edition) Addision Wesley, 2012. Available online: https://www.cdk5.net/wp/
• Reference: Sukumar Ghosh. Distributed systems: an algorithmic approach (2nd Edition) Chapman&Hall /CRC, Author’s own course material, Spring 2015. Available online: http://homepage.divms.uiowa.edu/~ghosh/16615.html

Teaching Methodology: Methods, techniques, & activities

Activities and Teaching Methods

Activities within each section

Learning Activities	Section 1	Section 2	Section 3	Section 4
Development of individual parts of software product code	1	1	1	1
Homework and group projects	1	1	1	1
Testing (written or computer based)	1	1	1	1
Oral polls	1	1	1	1
Discussions	1	1	1	1

Formative Assessment and Course Activities

Ongoing performance assessment

Section 1

Activity Type	Content	Is Graded?
Question	What are the distributed systems?	1
Question	Give an example of distributed systems.	1
Question	What are the advantages of layered architecture?	1
Question	What are the roles of transport protocols?	1
Question	How the TCP and UDP differ from each other? When one is preferred over the other?	1
Question	What is socket programming?	1
Question	How socket programming is different for UDP and TCP?	1
Question	Write a simple UDP/TCP client-server echo program	0
Question	Write a simple chatting program using UDP/TCP sockets	0
Question	Given the simple echo server program, apply socket timeouts and catch timeout exceptions	0
Question	Write a UDP-based reliable file transfer protocol	0
Question	Write a program that parallelly executes the CPU-bound tasks using multiple processes	0

Section 2

Activity Type	Content	Is Graded?
Question	How the threads differ from processes?	1
Question	What are the I/O and CPU-bound tasks?	1
Question	For what kind of tasks, using threads is preferred than using processes?	1
Question	What is a remote procedure call?	1
Question	What are some well-known distributed system architectures?	1
Question	Discuss the structured and unstructured decentralized architectures.	1
Question	You have a list of large numbers, and you need to find if they are prime or not. Would you use multithreading, multiprocessing, or sequential programming in order to complete the task asap? Prove it in practice.	0
Question	You need to send multiple requests to a server and receive responses. Assume there is a few msecs of delay before you receive the response from the server. Would you use multithreading, multiprocessing, or sequential programming in order to complete the task asap? Prove it in practice. (Order of the requests/responses doesn't matter)	0
Question	Discuss two ways of creating the threads using threading module in Python: 1) passing the worker function as a target, 2) subclassing the Thread class	0
Question	Given the function implemented locally, make it available to be called through RPC from remote process? Use xmlRPC.	0

Section 3

Activity Type	Content	Is Graded?
Question	How NTP protocol works?	1
Question	How Berkeley protocol works?	1
Question	Discuss the mutual exclusion algorithms.	1
Question	Discuss the permanent and server-initiated replicas and their difference	1
Question	Explain the Primary-backup protocol.	1
Question	Given three machines with drifting clocks, adjust their clocks using Berkeley algorithm.	0
Question	Explain how Bully algorithm for election works	0
Question	Explain how Ring algorithm for election works	0
Question	Explain the centralized (permission-based) method of mutual exclusion	0

Section 4

Activity Type	Content	Is Graded?
Question	Discuss the failure models	1
Question	Discuss different failure masking techniques by redundancy	1
Question	What is k-fault tolerant group?	1
Question	What is general model of failure detection?	1
Question	Explain basic reliable multicasting	1
Question	Explain what is authentication	1
Question	Explain what are message confidentiality and integrity	1
Question	Same as above	0

Final assessment

Section 1

1. Describe an advantage of layered architecture?
2. Describe the differences between TCP and UDP protocols?
3. Provide examples when using UDP can be more reasonable than TCP?
4. Describe how UDP and TCP socket programming differ from each other?

Section 2

1. Discuss the differences between the threads and processes.
2. What is the Race condition?
3. Discuss the ways to protect the shared data from the race condition
4. You're given the worker function that just sleeps for a second and quits, implement the same behavior in a subclass of the Thread.
5. Discuss the RPC and its advantages over using the low-level socket programming?
6. Discuss the decentralized architecture: structured and unstructured p2p systems.

Section 3

1. Discuss NTP and Berkeley protocols for synchronization and explain their key difference
2. Discuss permission-based and token-based solution for mutual exclusion.
3. Discuss content replication: permanent, server-initiated, and client-initiated replicas.
4. Explain the Primary-backup protocol, its advantages and disadvantages.

Section 4

1. Same as above

The retake exam

Section 1

Section 2

Section 3

Section 4