1 Overview

1.1 Welcome

CSE113: Parallel and Concurrent Programming
UCSC CSE
Fall 2026

Instructor:
    Mohsen Lesani <mlesani@ucsc.edu>
Time:
    Tuesdays Thursdays 09:50am-11:25am
Location:
    Porter Acad 148
TAs:
    Md Hasanul Islam <mislam5@ucsc.edu>
Tutor:
    Pratham Kotkar <pkotkar@ucsc.edu>

Hello and welcome to the parallel and concurrent programming class! Over the last decade, systems have become increasingly parallel, from our phones to supercomputers. Nearly every device today contains multiple compute units, CPUs and GPUs among them. Working together, these units can solve problems such as model training far faster than a single core can. But they must be programmed with care, for both performance and safety. In this class, you will learn parallel programming models, synchronization idioms and their implementation, and reasoning about concurrency.

This class is scheduled to be in person. We will follow the university guidelines and adapt if necessary. We will do our best to accommodate temporary remote attendance if needed (e.g., if you get sick); however, you are expected to make an effort to attend in-person classes.

1.2 Websites and Forums

• Github.io
  https://mohsenlesani.github.io/slugcse113/
  Non-protected materials will be hosted on this website. This includes the schedule, lecture slides, and references, etc.

• Canvas
  https://canvas.ucsc.edu/courses/94996
  Protected materials will be hosted on a Canvas website that you will need your university credentials to access. These materials include homeworks, zoom links, lecture recordings, tests, grades, etc.

• Piazza
  https://piazza.com/class/mubljwbhjval1/
  A Class forum will be provided in Piazza. If you organize other forums outside of the class Piazza (e.g. discord), you must adhere to academic integrity and be kind and respectful.

1.3 Acknowledgement

The material for this course is adopted from Professor Tyler Sorensen.

2 Schedule

The schedule may adapt to our pace. The slides for each lecture are uploaded before the lecture.

The example code snippets from the lectures are available at this Code Repo.

2.1 Module 1: Introduction, Background and ILP

2.2 Module 2: Mutual Exclusion

2.3 Module 3: Concurrent Data Structures

2.4 Module 4: GPU Computing

2.5 Module 5: Advanced topics

3 Description

3.1 Summary

Welcome to CSE 113: Parallel and Concurrent Programming! In this class, we will explore many aspects of parallel computing, from instruction-level parallelism in seemingly sequential programs to thread-level parallel programs that can efficiently execute across the many cores of today’s multiprocessors and accelerators (e.g., GPUs).
Further, we will learn how to write programs that execute efficiently and correctly in concurrent environments.
This class will give you a powerful skillset considering that today’s computers are increasingly parallel.

3.2 Modules

This class will be split into 5 modules, each of which are roughly two weeks:

• Module 1: Introduction, Background and ILP
  This module will introduce the course and provide a programming, compiler, and architectural refresher. We will discuss how modern hardware exploits parallelism within a sequential thread, known as instruction-level parallelism (ILP), and how to write parallel code in C++.

• Module 2: Mutual Exclusion
  This module will discuss the fundamental problem of mutual exclusion. We will discuss the theory behind mutual exclusion, how it is implemented in practice, and further, specialized mutual exclusion implementations.

• Module 3: Concurrent Data Structures
  This module will discuss concurrent objects, and how to reason about them. We will discuss several implementations, and show how they can be used in load balancing and software pipelining.

• Module 4: Parallel Programming on GPGPUs
  This module will discuss general purpose (GP) GPU programming. We will discuss the single-instruction multiple-threads (SIMT) programming model, hierarchical execution, and different architectural considerations when optimizing programs.

• Module 5: Advanced topics
  This module will discuss advanced topics, including memory consistency and fairness.

3.3 References

We cover the required material in the class and provide the slides. We do not require a physical textbook for this class; however, we list the following. Each are available online from the UCSC library.

• The Art of Multiprocessor Programming.
  This book has a nice collection of concurrent data structures.
• Computer Architecture: A Quantitative Approach
  For those of you who have not taken CSE 120, you may find this helpful. Chapter 2 deals with the memory hierarchy and Chapter 3 deals with hardware for instruction level parallelism.
• CUDA by Example
  We recommend this book for more information on GPU computing.

3.4 Required Background

The prerequisites for this class are CSE 12 (systems), CSE 101 (data-structures and algorithms), and recommended CSE 120 (architecture). You will need some foundation in all of those topics to succeed in this class. For example: you will need to know data-structures and algorithms, as we will extend some of these sequential concepts to their natural concurrent counterparts. You will need some systems background, as we will discussing many aspects of the hardware/software interface. Parallel programming is most efficiently executed on parallel hardware; thus, it is helpful to understand shared hardware resources (e.g. the memory hierarchy) of the underlying architectures.

Because this is an upper division class, we expect a general CS foundation. For the homeworks, we will assume that you are:

• comfortable using a linux command-line
• programming in a high-level language (e.g. Python)
• programming in a low-level language (e.g. C)
• a high-level understanding of computer architecture
• a basic ability to use Github
• a basic ability to use Docker

• Attendance/Quiz

Live discussions and synchronous class attendance are a valuable part of the learning experience. I expect you to make an effort to synchronously attend this class in-person. I plan to upload recordings of the class to canvas, but this is not a substitute for attendance.

Attendance will be graded using a small quiz given at the end of the class. Please do not submit the quiz unless you either attended or watched the lecture.

If synchronous attendance drops significantly then we will stop recording lectures and make attendance a part of the grade.

3.5 Accessibility

UC Santa Cruz is committed to creating an academic environment that supports its diverse student body. If you are a student with a disability who requires accommodations, please submit your Accommodation Authorization Letter from the Disability Resource Center (DRC) to me by email, preferably within the first two weeks of the quarter. I would also like us to discuss ways we can ensure your full participation in the course. I encourage all students who may benefit from learning more about DRC services to contact DRC by phone at 831-459-2089 or by email at drc@ucsc.edu.

3.6 Privacy

We plan to record lectures in class. Please be aware that:

• Things you say or do will be recorded. We doubt that this will be an issue, but if you want us to remove any part of the recording, please just let us know.
• Many forums (e.g. zoom chats, Piazza posts, etc.) chats are not private. Please be respectful and kind, and assume everyone can see what you are typing.

4 Teaching Team

We have a great teaching staff this quarter! All of them are passionate about parallel programming. Please get to know them and take advantage of the office hours and mentoring sessions they provide.

4.1 Office Hours:

• TA
  Md Hasanul Islam
  <mislam5@ucsc.edu>
  Monday, Wednesday 10:30am-11:30am
  BE2, room 315,
  Zoom

• Tutor
  Pratham Kotkar
  <pkotkar@ucsc.edu>
  3:00pm - 5:00pm on Mondays
  BE2 315 room

• Instructor
  Mohsen Lesani
  <mlesani@ucsc.edu>
  Tuesday Thursday 3-4pm. Office: E2 331
  My office hours can be remote or in-person.
  My physical office is E2-331.
  I announced a Zoom link and its passcode on canvas.

4.2 Asynchronous Communication

For any questions outside of office hours: Please post to the class Piazza.

Link to Piazza:
https://piazza.com/class/mubljwbhjval1/

If your question is more general, make it visible to the rest of class. If it isn’t clear if it is a sensitive question or not, please start out by making the question to the teaching staff and we can advise on making it public or not. Feel free to answer questions that your classmates post or freely participate in discussions there.

• If it is a sensitive topic, you can post only to the teaching staff. Please do not email us individually directly unless it is an emergency. Those emails get buried, or they might not be seen by the right member of the teaching staff. This especially applies to grading questions; if you have questions about your grade do not email a grader directly. Write a private post on Piazza to the teaching staff. Typically grades are a collaborative effort between several readers, and it helps if we can all see the issue.

We will strive to reply to homework questions and discussions within 24 hours. Please do not plan on, or expect help, outside of regular business hours (after 5 pm, weekends, or holidays)

5 Assessment

5.1 Grade Breakdown

• Attendance/Quiz: 10%
• Homeworks: 50% (10% each)
• Midterm Exam: 10%
• Final Exam: 30%

If you want to discuss a grade, please contact the teaching staff no later than 1 week after the grades are posted.

5.2 Attendance/Quiz

A small quiz is given at the end of each class. Each quiz has a few multiple-choice or short-answer questions that are designed to make you think about what you have learned in the class, and the questions might be open-ended. Please do not submit the quiz unless you either attended or watched the lecture. You can have up to 3 absences that will not affect your grade.

5.3 Homework:

There will be one assignment per module, for a total of 5 homeworks.

Homeworks are due at midnight on their due date. (Please do not plan on help after 5 pm.) You will have 10 days for each assignment. You have an additional 3 days to turn homework in late without penalty. After that, late submission will not be accepted. The due date for the last homework may need to be adjusted to account for the end of the quarter.

5.4 Exams:

There will be two exams in this course: a midterm and a final. The midterm will be worth as much as a single homework assignment (10%). The final will be worth 30%.

The midterm will be given halfway through the class. The final will be given in the finals week. Dates will be announced in the schedule section.

You will be allowed 3 pages (front and back) of notes in any format (printed, hand-written, colored, etc). Feel free to print slides to use as your notes.

5.5 Academic Integrity

One of the joys of university life is socializing and working with your classmates. We want you to make friends with each other and discuss the material. That said, I expect all assignments (code, write-ups, and tests) to be your own original work. If you work together with a classmate on an assignment, please mention this, e.g. in the comments of your code. If you use a figure you didn’t create in a write-up, then it needs a citation. Please review the universities policy on plagiarism. This class has a zero-tolerance policy on cheating. Please don’t do it. We would much rather get a hundred emails asking for help than have to refer anyone for academic misconduct.

As a final note on cheating: It is crucial that you benefit from your time at the university, and learn the concepts thoroughly. If you cheat, you will not be able to stand out from others who put in the effort when it comes time to find a job. Cheating will have a devastating impact on your own career opportunities. Just don’t do it.

5.6 Using AI Tools

We are in an exciting time for AI, especially with LLMs. These tools have incredible potential and they are improving every day. However, the educational community has not had sufficient time to understand their impact on learning objectives. This class has been designed to be taken without the use of AI tools. You can use them to retrieve information but you cannot be replace by them in this course. You yourself should consider the problem, think about it and solve it step by step. If we suspect abuse, then we will do random audits of assignments, where you will be asked to explain your implementation in detail.

If you are interested in seeing how these tools can help with parallel programming, please feel free to use them after you have submitted a non-AI version of the homework. We would be very interested in hearing about your experience, e.g., as a piazza post.