This page might be updated without prior notice. Please look out for notifications on Canvas and check this page frequently.

Last updated on 01/12/2026

Course Information

Lectures: 10 am - 1pm Tues, Room 601@Dubois Center at UNC Charlotte Center City

Office hours: 4pm - 5pm Wed or by appointment

Instructor: Jinzhen Wang, jwang96@charlotte.edu

TA: Kunal Shah, kshah65@charlotte.edu

Course Description

This course provides an in-depth exploration of parallel computing, with a primary focus on single-node architectures. It examines the why, what, and how of parallel computing, introducing the fundamental principles and practical techniques required to design, implement, and evaluate parallel programs on modern computers.

Students will study shared-memory programming models and GPU parallelism, gaining hands-on experience in developing and optimizing parallel algorithms for multi-core CPUs and many-core GPUs. The course integrates theoretical foundations with practical programming approaches and performance optimization strategies.

Learning Objectives

By the end of the course, you will:

• Understand key concepts in parallel computing, including parallel architectures, concurrency, synchronization, communication, and performance analysis.

• Develop scalable parallel applications while critically evaluating trade-offs among performance, correctness, and portability.

Prerequisites

There is no prerequisite needed to take the course. However, you probably need the following skill sets and knowledge.

• A solid foundation in computer architecture.

• Intermediate proficiency in C/C++.

• Basic understanding of algorithms and data structures.

• Basic familiarity with operating systems and concurrency concepts.

• Basic proficiency with Linux environment, git, ssh/scp, and build systems.

Learning Resources

• Primary Resource: lecture slides

• Assignments

• Practice questions prior to exams

• Various external resources

• No required textbook

• Some reading materials listed in syllabus and appended in lecture slides.

Grading

• 5% x 10 Homework assignments

• 20% Midterm exam

• 20% Final exam

• 10% Class participation (attendance)

Lectures

• Lecture 1 - Introduction / Logistics (how to HPC cluster)

• Lecture 2 - Introduction to Parallel Computing

• Lecture 3 - Extracting Parallelism

• Lecture 4 - Representing and Scheduling Parallelism

• Lecture 5 - Code Redesign

• Lecture 6 - Basics of Shared Memory Programming

• Lecture 7 - Race condition, Deadlock and False sharing

• Lecture 8 - Concurrent Data Structures

• Lecture 9 - GPU Architecture and Programming Model

• Lecture 10 -  Multi-GPU programming

Assignments

10 homework assignments (written and coding) where students practice the conceptual understanding of parallel computing, and hands-on techniques of building large-scale compute/data-intensive applications.

Late Days Policy

• 5 late days total for the semester

• Maximum 2 late days per assignment

• After late days are exhausted, no submissions accepted

In general, we do not grant extensions on assignments. There are several exceptions:

• Medical Emergencies

• Family/Personal Emergencies

• University-Approved Travel