In this module, we will learn the fundamentals of task parallelism. Tasks are the most basic unit of parallel programming. An increasing number of programming languages (including Java and C++) are moving from older thread-based approaches to more modern task-based approaches for parallel programming. We will learn about task creation, task termination, and the “computation graph” theoretical model for understanding various properties of task-parallel programs. These properties include work, span, ideal parallelism, parallel speedup, and Amdahl’s Law. We will also learn popular Java APIs for task parallelism, most notably the Fork/Join framework.

Welcome to Module 2! In this module, we will learn about approaches to parallelism that have been inspired by functional programming. Advocates of parallel functional programming have argued for decades that functional parallelism can eliminate many hard-to-detect bugs that can occur with imperative parallelism. We will learn about futures, memoization, and streams, as well as data races, a notorious class of bugs that can be avoided with functional parallelism. We will also learn Java APIs for functional parallelism, including the Fork/Join framework and the Stream API’s.

Join Professor Vivek Sarkar as he talks with Two Sigma Managing Director, Jim Ward, and Software Engineers, Margaret Kelley and Jake Kornblau, at their downtown Houston, Texas office about the importance of parallel programming.

Welcome to Module 3, and congratulations on reaching the midpoint of this course! It is well known that many applications spend a majority of their execution time in loops, so there is a strong motivation to learn how loops can be sped up through the use of parallelism, which is the focus of this module. We will start by learning how parallel counted-for loops can be conveniently expressed using forall and stream APIs in Java, and how these APIs can be used to parallelize a simple matrix multiplication program. We will also learn about the barrier construct for parallel loops, and illustrate its use with a simple iterative averaging program example. Finally, we will learn the importance of grouping/chunking parallel iterations to reduce overhead.

Welcome to the last module of the course! In this module, we will wrap up our introduction to parallel programming by learning how data flow principles can be used to increase the amount of parallelism in a program. We will learn how Java’s Phaser API can be used to implement “fuzzy” barriers, and also “point-to-point” synchronizations as an optimization of regular barriers by revisiting the iterative averaging example. Finally, we will also learn how pipeline parallelism and data flow models can be expressed using Java APIs.

The next two videos will showcase the importance of learning about Concurrent Programming and Distributed Programming in Java. Professor Vivek Sarkar will speak with industry professionals at Two Sigma about how the topics of our other two courses are utilized in the field.