CS3331 Concurrent Computing
Course Information Page
Fall 2020

Course Instructor		Dr. C.-K. Shene
Office		Rekhi 305
Class Meeting		Monday and Wednesday 16:05 - 17:20
Classroom		M&M U115
Office Hour		Monday and Wednesday 15:05 - 15:55
Phone		(906) 487-3392
e-mail		shene@mtu.edu

Most Recent Announcements is updated (December 4, 2020)

ThreadMentor FAQ (March 8, 2019)

Some Important Advises

This course will use C and C++. We will discuss C++ to some degree barely enough for our purpose. Thus, you should get a good reference of C/C++ in hand.

Always start doing your programming assignment EARLY. Waiting until the last few hours is definitely not a wise move, because there could be too many people running their programs at the same time making system response time extremely slow. Moreover, machines may crash causing extra delay. Normally, I do not grant any extension if your home machine, your phone line or the department machines crash in the last minutes. Always start early.

Concurrent programming requires a careful planning, which may need a mindset that is different from that of sequential programming you learned in previous courses. When you receive a programming assignment, do not sit in front of a workstation and start typing your program immediately. This trial-and-error approach, in general, DOES NOT work and may require longer time to finish your assignment. The best way of doing concurrent programming assignments is to think carefully with a pencil and a piece of paper before touching a computer keyboard.

Since the behavior of a concurrent program is dynamic, which means it acts differently from time to time, you should try to reason about the correctness of your program. In general, just running for one or two test cases is not sufficient. Our grader will read your program to find possible errors. When writing sequential programs, the same bug appears at the same place every time you run the same program. However, bugs of a concurrent program (e.g., race conditions and deadlocks) may not appear every time. As a result, you may have an incorrect program even though you may have had hundreds correct test runs.

Due to security reasons, starting with Fall of 2008, you will not receive submission confirmation e-mails. As a result, you should use the interactive version of submit and ensure all required files are submitted properly. If you choose to use the command line version of submit, you should use the recover command to verify your submission.
After submitting your assignment, DO NOT modify your files so that the system administrator can retrieve your files in case your submission did not go through successfully.

Always read the assigned materials listed in the weekly reading lists and do the problems. Weekly reading lists are available near the bottom of this page.

Catalog Description

This course will discuss concepts and techniques in concurrent computing. Major topics include: processes and threads, mutual exclusion, semaphores, monitors and condition synchronization, deadlock, safety and liveness, message passing, and concurrent architectures.

Course Credits, etc.

Credits: 3.0, Lec-Rec-Lab: (0-3-0), Semesters Offered: Fall, Spring

Course Prerequisites

CS 1142 or (CS1141 and CS1040) and CS2311 and CS2321

Course Objectives and Learning Outcomes

Know the basics of concurrent computing
Understand the importance of race conditions and the impact on synchronization and hence mutual exclusion
Learn to argue and prove the correctness of simple mutual exclusion based algorithms
Be able to use modern synchronization primitives in your programs
Learn to find race conditions and avoid deadlocks

Topics Covered

Basic systems concepts important to concurrent computing
Multi-process and multithreaded programming basics
Race conditions, critical sections, and synchronization
Pure software and hardware solutions and proof techniques for mutual exclusion
Synchronization primitives (i.e., semaphores, mutex locks, monitors of Hoare and Mesa types, basic message passing)
Deadlocks and livelocks
Survey of some languages (e.g., Java and Ada) and libraries (e.g., Pthreads)

Textbook

There is no suitable textbook for this course. All slides used in class and supporting materials are available:

in the common directory /classes/cs3331.ck/common or /local/classes/cs3331.ck/common or /mtu/classes/cs3331.ck/common
The class slides page is available if you prefer web access.
Software tools and example programs are only available in the common directory.

Check the /classes/cs3331.ck/common directory or the class slides page frequently for new slides and other available stuffs.

Other Useful Materials

Make your program
Multithreaded Programming with ThreadMentor
Solaris multithreaded programming
Unix Multiprocess Programming
Signals
Non-Local GOTO: setjmp() and longjmp()

Meeting Time and Office Hours

	Monday	Tuesday	Wednesday	Thursday	Friday
13:05 - 13:55		Dept. Meeting		TPR Meeting
14:05-15:05
15:05-15:30					Department Seminar
15:30-16:00	Office Hour Rekhi 305		Office Hour Rekhi 305		Department Seminar
16:00-16:30	Office Hour Rekhi 305		Office Hour Rekhi 305
16:35-17:50	CS3331 Fisher 325		CS3331 Fisher 325

After class and e-mail discussions are welcome

Course Syllabus

No. of Weeks	Course Topics
1	Basics
1	Hardware and OS Stuffs
3	Processes and Threads
1	Critical Sections and Mutual Exclusion
5	Synchronization	Mutex Locks
		Semaphores
		Monitors
		Message Passing
		Race Conditions and Livelock/Deadlock
2	Programming Language and System Supports
1	Other Topics

Important Dates

The following is a list of important dates. The exam dates are fixed and will not change; however, programming assignment due dates may be modified based on the course pace. Changes will be announced in class and on the info page.

Week	Date	Day	Time	Event
Week 4	February 3	Wednesday	10pm	Program 1 due
Week 6	February 17	Wednesday	Sync class	Exam 1 (50 min)
Week 7	February 26	Friday	11pm	Program 2 due
Week 9	March 19	Friday	11pm	Program 3 due
Week 10	March 24	Wednesday	Sync class	Exam 2 (50 min)
Week 11	April 2	Friday	11pm	Program 4 due
Week 13	April 16	Friday	11pm	Program 5 due
Week 14	April 23	Friday	11pm	Program 6 due
Final Week	April 26	Monday	5-7pm	Final (2 hours)

Date/Time changes will be announced in class.
Usually, I will follow the schedule shown above.

Weights Assigned to Programs

We will have maximum six programming assignments.

*Program 1*	*Program 2*	*Program 3*	*Program 4*	*Program 5*	*Program 6*
50	100	50	100	100	70

Weights Assigned to Exams

*Exam 1*	*Exam 2*	*Final*
100	100	200

Note that quizzes are not listed above. Pop quizzes will take place when I see they are needed. Pop quizzes are always short exams testing if you have acquired the most basic knowledge of each topic. As a result, they usually take place about one week after a major topic is completely discussed. However, it does not mean every topic will have a quiz.

Grading

The theory part has two midterm exams and some quizzes. The sum of the scores you will obtain has a 50% weight. The programming part has five to six assignments, depending on our progress. The sum of the scores you will obtain constitutes the other 50%. Grading will be based approximately on the scale shown in the table below.

Grade	A	AB	B	BC	C	CD	D	F
Scale	90-100%	85-89%	80-84%	75-79%	70-74%	65-69%	60-64%	0-59%

In addition, you will not receive a passing grade in the course unless you receive a passing grade on the exams alone (i.e., 60%). Likewise, you will not receive a passing grade in the course unless you receive a passing grade on the programming assignments alone (i.e., 60%).

Late Policy

No late programming assignments will be graded, except that you have a valid excuse.

Collaboration and Cheating

No assignment-specific assistance should be given nor received on any programming assignment. Any outside help concerning the use of the computer facilities is acceptable. You may discuss the meaning or intent of an assignment, but not ways to program it. All work turned in must be completely your own. You must not share code or even outlines! Any and all use of programs from other texts or references must be explicitly stated as part of program documentation. In general, any acts of plagiarism will result in a failing grade for the course along with the submission of a report to the Dean of Students. See the hand-out on Computer Science Policy on Cheating and Academic Integrity Resource Center for more details.

Exams and Final

You are responsible for avoiding exam time conflicts!!

ADA

``MTU complies with all federal and state laws and regulations regarding discrimination, including the Americans with Disability Act of 1990 (ADA). If you have a disability and need a reasonable accommodation for equal access to education or services at MTU, please call Dr. Bonnie B. Gorman, Dean of Students, (2212). For other concerns about discrimination, you may contact your advisor, department head, or the Affirmative Action Office (3310).''

The following two web sites may provide you with more information:

Other Useful Information

Downloadable Stuffs

Files are in HTML or PDF format.

Programming Information: How to Submit Programs (Thank You, Yuchen Wang); How to Install VM and ThreadMentor (Thank You, Yuchen Wang); ThreadMentor Visualization (Thank You, Yuchen Wang); Programming Assignment I; Programming Assignment II; Programming Assignment III; Programming Assignment IV; Programming Assignment V; Programming Assignment VI
Exam Information: Exam 1 Solutions (PDF file); Exam 2 Solutions (PDF file)
Reading Lists

Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7

Week 8 Week 9 Week 10 Week 11 Week 12 Week 13 Week 14

You are visitor since January 16, 2011
Last update: December 4, 2020

Please send comments and suggestions to shene@mtu.edu

Week 1	Week 2	Week 3	Week 4	Week 5	Week 6	Week 7
Week 8	Week 9	Week 10	Week 11	Week 12	Week 13	Week 14

CS3331 Concurrent Computing Course Information Page Fall 2020