3 semester hours
Wednesday Evenings 6:30 p.m. – 9:30 p.m. in Doolan 222
Instructor: Ray Toal, Doolan 110, rtoal@lmu.edu, +1 310.338.2773

Learning Outcomes

To master the principles of concurrent and distributed programming and become proficient in applying these principles through programming exercises in multithreading, networking and simulation. Students will develop code using distributed and concurrent facilities that are provided by the operating system (case study: WinAPI with C API and C++ wrappers), language intrinsic (case studies: Java, Perl, and Ada 95), and threading libraries (case study: pthreads)

Prerequisites

Mastery of a high-level programming language such as C++ or Java; expert knowledge of data structure and algorithm design; familiarity with object-oriented programming, computer hardware, and operating systems; some mathematical maturity (this is a graduate class).

Readings

There really isn't any required textbook for this course, but I very strongly recommend you get the book by Ben-Ari and read it; you can also get one of the books from Garg and skim it. Also, as there will be many programming assignments throughout the course, those of you that learn best from books will want to pick a few of the language-specific texts mentioned below as well. In general, read as much as you can from the list below: it might improve your grade!

Theory: You'll want some solid, academic, treatment of distributed algorithms, models for concurrency and distribution, and distributed programming languages. Classic textbooks and journal papers are your best bet here. Examples:

• M. Ben-Ari, Principles of Concurrent and Distributed Programming, Prentice-Hall, 1990. {A fabulous 225-page paperback; fifteen years old but completely relevant today; an oldie but a goodie.}
• Vijay K. Garg, Elements of Distributed Programming, Wiley-IEEE Press, 2002. {Looks like a pretty good text; seems rigorous and complete, but I haven't read it yet.}
• Vijay K. Garg, Concurrent and Distributed Computing in Java, Wiley-IEEE Press, 2004. {Seems to be similar to the previous text, though it uses Java as an implementation language to make the topics concrete.}
• Nancy A. Lynch, Distributed Algorithms, Morgan Kaufmann, 1997. {This text contains just about everything, and features many correctness proofs. What would you expect from this publisher?}
• Gerard Tel, Introduction to Distributed Algorithms, Second Edition, Cambridge University Press, 2001. {A text focusing on algorithms.}
• Henri Bal, Jennifer Steiner, and Andrew S. Tanenbaum, Programming languages for distributed computing systems, Computing Surveys 21(3): 261–322. {A paper that discusses over 90 distributed languages.}
• Rance Cleaveland and Scott Smolka, Strategic Directions in Concurrency Research, Computing Surveys 28(4):607–625.

Java: It seems like most non-trivial Java programs use threads or networking. Fortunately the language was designed with multithreading in mind, and networked code is easy to write.

• Scott Oaks and Henry Wong, Java Threads, 3rd edition, O'Reilly, 2004. {Perhaps the best starter book for Java threading; the third edition covers Java 5 and is a huge improvement over the previous editions.}
• Elliotte Rusty Harold, Java Network Programming 3rd edition, O'Reilly, 2004.{Works great as a Java book; don't expect to learn TCP/IP from it, though. Don't get anything less than the 3rd edition!}
• Doug Lea, Concurrent Programming in Java, 2nd edition, 1999. {Could be considered a classic — this is the design patterns book for concurrency; recommended; not for beginners; Contains an extensive bibliography. Many of the ideas in this book have been incorporated into Java 5.0.}

Ada: Although not as widely used as it was several years ago, Ada remains one of the best case studies for a language that directly supports distributed, concurrent, and real-time features.

• Burns and Wellings, Concurrency in Ada, Cambridge University Press, 1998. {This is a great all-around text; it has plenty of examples with good commentary.}
• Ken Shumate, Understanding Concurrency in Ada, McGraw-Hill, 1990. {I remember having read this book when it first came out and liking it. But it went out of print rather quickly. I remember some fairly cool diagrams and a lot of coverage of real-time issues.}

WinAPI Processes and Threads: You'll need to get good at consulting the online documentation for Windows threads; there aren't too many books, but the one by Beveridge and Weiner is pretty good.

• Microsoft, Windows API, Online documentation. {This is good, and indispensable for programmers but it's a terse reference and some readers may want a book as well.}
• Jim Beveridge and Robert Weiner, Multithreading Applications in Win32, Addison-Wesley Professional, 1996. {Approaching 10 years old, but still okay. I'm not aware of too many texts in this area.}

Perl: I don't know of any good books or papers specifically covering Perl threads, but several presentations are online. As always with Perl, check out the tutorials and references on perl.org especially perlthrtut

Miscellaneous Papers: Here are a few papers we'll look at. They aren't classics.

• Cormac Flanagan and Stephen N. Freund, Type-based race detection for Java, ACM SIGPLAN Notices, 35(5):219–232, 2000.
• Radu Rugina and Martin C. Rinard, Automatic Parallelization of Divide and Conquer Algorithms, Proc. 7th ACM Symposium on Principles and Practice of Parallel Programming, 72–83, 1999.
• Jurgen Lind. Specifying Agent Interaction Protocols with Standard UML. In Agent-Oriented Software Engineering, 136—147, 2001.

Various papers and readings will be assigned throughout the course (including my own course notes, practice problems, and sample code). These readings will be posted on individual assignment pages. If you have projects or papers to work on, you’ll have to find some additional readings on your own. Make sure you take the time for effective self-study. Take advantage of classmates and friends; the computing industry is one of the most collaborative fields in which to work, and your course experience should reflect this.

Workload

In accordance with the LMU Credit Hour Policy, this 3-unit course will require 9 hours of work per week (including the time spent in lecture and lab).

Assignments and Grading

You’ll have several homework sets containing in-depth theoretical questions and non-trivial programming problems, and quizzes and a final exam with less difficult material. To help prepare you to meet industry expectations for college graduates, programming assignments will sometimes be required to be placed in version-controlled public repositories (most students tend to prefer GitHub). Exams may cover material from lectures not previously assigned for homework, so take care to learn concepts and not just memorize technical steps and recipes to answer questions.

Occasionally, you may be permitted to work in groups of students; however, while only one solution set is turned in per group, both students are responsible for understanding all of its content and may be asked at any time for an oral explanation of any solution. Collaboration with other groups is fine but must be limited: you may share ideas and approaches but not solutions. You must acknowledge any help received. Academic dishonesty may result in expulsion; be certain your work meets the standards set forth in the LMU Honor Code.

Your final grade will be weighted as follows:

Letter grades are figured according to the usual scale: 90% or more of the total points guarantees you an A, 80% a B, 70% a C, and so on. Note the word guarantee: 82 points will earn you at least a B-; you might still get an A if 82 is or is near the top score. The lower bounds ensure grades measure your achievement of the learning outcomes, and can never punish you because you did very well but on the low end of a class full of high-achievers.

To ensure a degree of fairness for those who consistently perform timely course work, and to encourage everyone to pace themselves properly in completing assignments, late work is normally penalized 15% per 24-hour period. If there is an issue that prevents you from submitting an assignment on time (e.g., excellent surf conditions, personal or family issues, sickness, conference attendance, job interviews, a family ski trip, or personal emergencies), let the instructor know ahead of time.

Where assignments involve programming, the quality of your code, not just its correctness, will play a large part in determining your grade. Please refer to these resources and notes on clean code for information on expectations of code quality. Appearance of the grading policy in this syllabus constitutes fair warning of the consequences of poorly written code.

If the course has a contribution score, it will be computed by awarding you 1 point every time you: (1) correct me during class, (2) ask a profound question during class (where profound elevates the discussion in some sense), (3) answer a question during class as the first response, (4) post a good question on the class Slack channel, (5) post a detailed answer to another student’s question on the Slack channel, (6) recommend an article or video or tutorial (on the Slack channel) directly related to the course material. There is an expectation of 10 points throughout the semester, well-spaced. (Please do not dump 10 video recommendations on Slack during the first week of class and then shut down.)

Student Rights and Responsibilities

You have the right to:

• A syllabus with stated learning outcomes
• Have those learning outcomes addressed
• Chats with the instructor during office hours
• Have questions answered via email (one-day turnaround max)
• Video and text chats with the instructor
• Feedback on your work
• A challenging experience (you should be working a little outside your comfort zone)
• All lecture sessions recorded and stored online for you to access
• An experience beyond what you would get from books, websites, online tutorials, and discussion forums
• Feel like you belong (if you are experiencing impostor syndrome or stereotype threat, please see the instructor)
• Leave the course better skilled than when you came in

In return, you are expected to:

• Use the lab resources, TA time, and instructor time when you need them
• Stay on top of all course announcements, both on BrightSpace and Slack
• Watch the course recording if you miss class
• Perform work conscientiously, neatly, and in a timely manner
• Be considerate and respectful to your fellow students
• Be honest with yourself and others
• Finish the course or withdraw by the withdrawal deadline

For online courses, I recommend your camera be turned on unless you have bandwidth problems or need to step away.

Two Tips for Success

You've probably seen hundreds of these, but there are two that stand out.

• Be aware of how students trick themselves into thinking they’re learning. Cramming, rereading, and highlighting do not move information from your working memory into long-term memory. Short-term working memory is not “learned.” Recall is the most effective way to load your long-term memory. Please work through Nicky Case’s interactive essay How to remember everything forever-ish. (Also, this article backs up the importance of recall.)
• Use checklists, as these are incredibly effective. Read Atul Gawande’s The Checklist and think about how you apply these to your studies.

Generative AI and Your Learning

Academic honesty includes the appropriate use of technology as an aid for learning and productivity. This includes but is not limited to LLM-based generative artificial intelligence tools such as ChatGPT, Gemini, and Copilot.

Modern GenAI provides fast solutions to a variety of computing problems but must be used responsibly to get the most out of your education. GenAI will not always be helpful, especially when asked to write code for scenarios it cannot recognize. It cannot independently validate the code that it produces, and will frequently produce bugs and security vulnerabilities, making things worse for you. It will not be available for job interviews. As a student and future software professional, you have a moral and ethical responsibility to deeply understand every line of the software you author and should therefore employ GenAI in your coursework sparingly, especially when its use might rob you of the stretch-zone learning that comes from crafting programs on your own.

That said, there are several acceptable use cases of GenAI in programmatic assignments, including:

• Generating short documentation strings and type hints.
• Finding examples and explanations for syntactic forms (e.g., how to use the spread operator; how to assign variables directly from an array or object; how to construct a dictionary comprehension; how to unwrap an optional, and so on). Prompt the AI for the how-to’s, not your final code.
• Interpreting error messages, as these might be pretty cryptic! Sometimes the AI can be a useful tutor when explaining things it is familiar with.
• Generating unit tests for edge cases. When it finds something you overlooked, make a mental or written note so you can find these cases on your own next time.
• Brainstorming new ideas for apps.

Remember the four As

DO use technologies such as GenAI to augment ✅, amplify ✅, and accelerate ✅ your learning; NEVER use technology to avoid ❌ learning.

Be responsible!

Generally, using GenAI and related tools without knowing what you are doing often leads to poor homework submissions that rarely achieve a grade higher than an F on their own merits, simply due to hallucinations and an inability to carry out tasks as they were assigned (even with good prompts).

Also, instructors and TAs are GenAI users themselves and will be able to easily detect inappropriate student use of these tools. The teaching staff will try to steer you back on track should we feel like you are giving up on your learning.

If you would like to use GenAI appropriately as a learner, see this short introductory article. If you are a TA, see this article by Jeff Olson on how to help teach students to use GenAI.

Topics

This is a tentative schedule.

• INTRODUCTION: Motivations for the study of concurrency; examples of concurrent and distributed systems; Basic definitions; Overview of modeling, granularity, scheduling and language integration; Simple examples coded in Java, Ada, Perl, C++/pthreads and C++/WinAPI.
• CORRECTNESS OF CONCURRENT PROGRAMS: Safety and liveness; Race conditions, atomicity, critical regions and mutual exclusion; Deadlock, livelock, starvation, and fairness.
• COMMUNICATION: Synchrony and asynchrony; Shared-variable communication and mutual exclusion: from Peterson's algorithm to test/set to barriers to semaphores to monitors to Ada protected objects; Message-based communication: client/server models (rendezvous, receipt, callback, etc.), channels, mailboxes, and pipes.
• JAVA THREADS: The Thread class and the Runnable and Callable interfaces; Thread creation and termination; Synchronization: volatile fields, atomic classes, locks, and synchronization objects; The concurrent and blocking collections; Threads in graphical applications; Scheduling with thread pools and timers.
• WINDOWS API CONCURRENCY: Overview of WinAPI; processes and threads; Process and thread API; Scheduling, priorities, boosts; Synchronization: kernel object approaches (semaphores, events, mutexes, waitable timers) and non-kernel object approaches.
• ADA CONCURRENCY: Tasks and protected objects; initialization, activation, communication and termination of Ada tasks; events, buffers, blackboards, broadcasting; Asynchronous Transfer of Control (ATC), abort; systems programming: interrupt handling, resource control and periodic scheduling. Real time facilities: priorities, queuing policies and synchronization primitives.
• DISTRIBUTED PROGRAMMING: Network and Internet fundamentals; Protocols and layers; Basic socket-level programming in Java, C and Perl; Ada distributed systems facilities: partitions, RPC, distributed objects, replication and distributed fault tolerance; Windows API interprocess communication: mailslots, slots, clipboard, DDE, messages, and sockets; HTTP and Web Applications; Remoting (RPC, RMI, Hessian/Burlap, Swing HTTP Invoker, Ajax, Web Services).
• JAVA ENTERPRISE COMPUTING: Java EE Overview; JNDI, Java Mail, JMS, Web tier (Java Servlets, Java Server Pages, Java Server Faces, Struts), Java Web Services Technologies (JAXB, JAX-RPC, SAAJ, etc.); The Struts MVC Framework; Persistence technologies (JDBC, JDO, iBATIS, Hibernate). Comparisons with .NET.
• SURVEY OF DISTRIBUTED LANGUAGES: A look at an old paper with 90 of them, supplemented with some modern ones.
• ADVANCED THEORY: Theory of parallel algorithms, covering models for parallel computation, SIMD vs. MIMD, PRAMs vs. connection machines, and the complexity of parallel algorithms (NC and P-completeness).
• MODELING AND VERIFICATION: Using Temporal Logic for specification of concurrent systems; Correctness proofs; Mechanical Verification Systems; Well-known formalisms and Process Algebras (e.g. Milner's p-calculus); a look at some classic distributed algorithms, with a focus on correctness proofs.
• IMPLEMENTATION: How to build a concurrency kernel: single- and multiple-processor implementation issues, and scheduling algorithms; Brief look at threads in the Java Virtual Machine; Parallelism in the Pentium family of processors.

Additional Information

Tentative Nature of the Syllabus. If necessary, this syllabus and its contents are subject to revision; students are responsible for any changes or modifications announced or distributed in class or posted online.

Course Evaluation. Student feedback on this course provides valuable information for continued improvement. All students are expected to fairly and thoughtfully complete a course evaluation for this course. Course evaluations for the Seaver College of Science and Engineering are administered online through the Blue™ evaluation system. You will receive an e-mail notification at your Lion e-mail address when the evaluation form is available. You may also access the evaluation form on Brightspace during the evaluation period. A few minutes of class time will be reserved for you to complete a course evaluation near the end of the semester. Please bring a laptop, smart phone, tablet or other mobile device to class on this date so that you can access the online evaluation platform.

Academic Honesty. Loyola Marymount University is a community dedicated to academic excellence, student- centered education and the Jesuit and Marymount traditions. As such, the University expects all members of its community to act with honesty and integrity at all times, especially in their academic work. Academic honesty respects the intellectual and creative work of others, flows from dedication and pride in performing one’s own best work, and is essential if true learning is to take place. As an LMU Lion, you are pledged to join the discourse of the academy with honesty of voice and integrity of scholarship.

Academic dishonesty will be treated as an extremely serious matter with severe consequences that can range from receiving no credit for an assignment or test to failing the class, to expulsion. It is never permissible to turn in any work not been authored by you without properly acknowledging the source. It is your responsibility to make sure that your work meets the standard of academic honesty set forth in the “LMU Honor Code and Process” which appears in the LMU Bulletin.

Special Accommodations. Students with special needs who require reasonable modifications, special assistance, or accommodations in this course should promptly direct their request to the Disability Support Services (DSS) Office. Any student who currently has a documented disability (ADHD, Autism Spectrum Disorder, Learning, Physical, or Psychiatric) needing academic accommodations should contact the DSS Office in Daum Hall, as early in the semester as possible. All discussions will remain confidential. Please visit LMU DSS for additional information. Please request any needed assistance as early in the semester as possible.

Wellness. Please familiarize yourself with, and make use, of the resources at Lion Wellness as needed.

Reporting Requirements of Sexual or Interpersonal Misconduct. As responsible employees, faculty are required to report any case of suspected sexual or interpersonal misconduct and cannot protect student confidentiality. For information about confidential counseling on campus and for general information about consensual relationships, sexual harassment, and sexual assault, please visit LMU Cares.

Emergency Preparedness. To report an emergency or suspicious activity, phone the LMU Department of Public Safety (x222 or 310-338-2893) or at the nearest emergency call box. In the event of an evacuation, follow the evacuation signage throughout the building to the designated safe refuge area where you will receive further instruction from Public Safety or a Building Captain. For more safety information and preparedness tips, visit LMU DPS.