COMP 322: Fundamentals of Parallel Programming (Spring

...

2024)

Course Syllabus

A summary PDF file containing the course syllabus for the course can be found here.  Much of the syllabus information is also included below in this course web site, along with some additional details that are not included in the syllabus.

...

The desired learning outcomes fall into three major areas:

1) Parallelism: functional programming, Java streams, creation and coordination of parallelism (async, finish), abstract performance metrics (work, critical paths), Amdahl's Law, weak vs. strong scaling, data races and determinism, data race avoidance (immutability, futures, accumulators, dataflow), deadlock avoidance, abstract vs. real performance (granularity, scalability), collective & point-to-point synchronization (phasers, barriers), parallel algorithms, systolic algorithms.

...

To ensure that students gain a strong knowledge of parallel programming foundations, the classes and homeworks homework will place equal emphasis on both theory and practice. The programming component of the course will mostly use the  Habanero-Java Library (HJ-lib)  pedagogic extension to the Java language developed in the  Habanero Extreme Scale Software Research project  at Rice University.  The course will also introduce you to real-world parallel programming models including Java Concurrency, MapReduce. An important goal is that, at the end of COMP 322, you should feel comfortable programming in any parallel language for which you are familiar with the underlying sequential language (Java or C). Any parallel programming primitives that you encounter in the future should be easily recognizable based on the fundamentals studied in COMP 322.

...

• Fork-Join Parallelism with a Data-Structures Focus (FJP) by Dan Grossman (Chapter 7 in Topics in Parallel and Distributed Computing)
• Java Concurrency in Practice by Brian Goetz with Tim Peierls, Joshua Bloch, Joseph Bowbeer, David Holmes and Doug Lea
• Principles of Parallel Programming by Calvin Lin and Lawrence Snyder
• The Art of Multiprocessor Programming by Maurice Herlihy and Nir Shavit

Lecture Schedule

...

 

Week	Day	Date (

2022 

2024)	Lecture	Assigned Reading	Assigned Videos (see Canvas site for video links)	In-class Worksheets	Slides	Work Assigned	Work Due

 

Worksheet Solutions
1	Mon	Jan

10

08	Lecture 1: Introduction

  

worksheet1

lec1-

slides

 

 

  

 

slides			WS1-solution
	Wed	Jan

12

10	Lecture 2:  Functional Programming

  

worksheet2

lec2 

lec02-slides

 

 

  

WS2-solution
	Fri	Jan

14worksheet3

12	Lecture 3: Higher order functions

  

worksheet3

lec3-

slides 

slides

  

WS3-solution
2	Mon	Jan

17

15	No class: MLK

        

 

	Wed	Jan

19

17	Lecture 4: Lazy Computation

   

worksheet4

lec4-slides

    

WS4-solution
	Fri	Jan

21

19	Lecture 5: Java Streams

   

worksheet5

lec5-slides

Homework 1

  

WS5-solution
3	Mon	Jan

24 

22

Lecture 6: Map Reduce with Java Streams

Module 1: Section 2.4

Topic 2.4 Lecture, Topic 2.4 Demonstration

worksheet6

lec6-slides

 

   

WS6-solution
	Wed	Jan

26

24

Lecture 7: Futures

Module 1: Section 2.1

Topic 2.1 Lecture , Topic 2.1 Demonstration

worksheet7

lec7-slides

 

   

 

		WS7-solution
	Fri	Jan

28

26	Lecture 8:  Async, Finish, Computation Graphs

, Ideal Parallelism

Module 1: Sections 1.

2

1, 1.

3

2

Topic 1.

2

1 Lecture, Topic 1.

2

1 Demonstration, Topic 1.

3

2 Lecture, Topic 1.

3

2 Demonstration

worksheet8

lec8 

lec8-slides

   

WS8-solution
4	Mon

 

Jan

31

29

Lecture 9: Ideal Parallelism, Data-Driven Tasks

Module 1: Section 1.3, 4.5

 

Topic 1.3 Lecture, Topic 1.3 Demonstration, Topic 4.5 Lecture

 

, Topic 4.5 Demonstration

worksheet9

lec9-slides

 

WS9-solution

 

	Wed

  

Jan 31

Lecture

 WedFeb 02Lecture

10: Event-based programming model

 

  

worksheet10

lec10-slides

     

	Homework 1	WS10-solution
	Fri	Feb

04  worksheet11

02	Lecture 11: GUI programming

as an example of event-based,
futures/callbacks in GUI programming

, Scheduling/executing computation graphs

Module 1: Section 1.4

Topic 1.4 Lecture , Topic 1.4 Demonstration

worksheet11

lec11-slides

Homework 2

   

	WS11-solution
5	Mon	Feb

07

05	Lecture 12:

Scheduling/executing computation graphs

Abstract performance metrics, Parallel Speedup, Amdahl's Law

Module 1: Section 1.

4

5

Topic 1.

4

5 Lecture , Topic 1.

4

5 Demonstration

worksheet12

lec12-slides

    

 

		WS12-solution
	Wed	Feb

09

07	Lecture 13:

Lightweight task parallelism

Accumulation and reduction. Finish

/async

accumulators

Module 1: Section

1

2.

1

3

Topic

1

2.

1

3 Lecture

,

Topic

1

2.

1

 

3 Demonstration

worksheet13

lec13-slides

   

	WS13-solution

Fri

Feb

11

09	No class: Spring Recess

 

        

6	Mon	Feb

14

12	Lecture 14:

Parallel Speedup, Critical Path, Amdah's Law

Data Races, Functional & Structural Determinism

Module 1:

Section 1

Sections 2.5, 2.6

Topic

1

2.5 Lecture ,  Topic

1 

2.5 Demonstration,  Topic 2.6 Lecture,  Topic 2.6 Demonstration

worksheet14

lec14-slides

    

WS14-solution
	Wed	Feb

16

14	Lecture 15:

Recursive Task Parallelism   

Limitations of Functional parallelism. Abstract vs. real performance. Cutoff Strategy

worksheet15

lec15-slides

 

 

   

Homework 2

WS15-solution

 

	Fri	Feb

18

16	Lecture 16:

Accumulation and reduction. Finish accumulatorsModule 1: Section 2.3Topic 2.3 Lecture , Topic 2.3 Demonstration 

Recursive Task Parallelism

worksheet16

lec16-slides

Homework 3

  

WS16-solution
7	Mon	Feb

21

19	Lecture 17: Midterm Review

   lec17-slides    

 

Wed

Feb 23

Lecture 18: Limitations of Functional parallelism.
Abstract vs. real performance. Cutoff Strategy

  

			lec17-slides
	Wed	Feb 21	Lecture 18: Midterm Review

 

lec18-slides

    

 

	Fri	Feb

25 

23	Lecture 19:

Data Races, Functional & Structural DeterminismModule 1: Sections 2.5, 2.6

Fork/Join programming model. OS Threads. Scheduler Pattern

Topic 2.

5

7 Lecture, Topic 2.

5

7 Demonstration, Topic 2.

6

8 Lecture, Topic 2.

6  

8 Demonstration

worksheet19

lec19-slides

   

WS19-solution
8	Mon	Feb

28

26	Lecture 20:

Confinement & Monitor Pattern. Critical sections
Global lockModule 2: Sections 5.1, 5.2, 5.6 Topic 5

Data-Parallel Programming model. Loop-Level Parallelism, Loop Chunking

Module 1: Sections 3.1, 3.2, 3.3

Topic 3.1 Lecture, Topic

5

3.1 Demonstration , Topic

5

3.2 Lecture,

Topic 5

Topic 3.2 Demonstration, Topic

5

3.

6

3 Lecture,

Topic 5

Topic 3.

6

3 Demonstration

worksheet20

lec20

lec20-slides

       

WS20-solution
	Wed

Mar 02

Feb 28

Lecture 21:

N-Body problem, applications and implementations

Barrier Synchronization with Phasers

Module 1: Sections 3.4

Topic 3.4 Lecture, Topic 3.4 Demonstration

worksheet21

lec21-slides

WS21-solution

  worksheet21lec21-slides     

	Fri	Mar

04

01	Lecture 22:

Fork/Join programming model. OS Threads. Scheduler Pattern Module 2: Sections 2.7, 2.8Topic 2.7 Lecture, Topic 2.7 Demonstration, Topic 2.8 Lecture, Topic 2.8 Demonstration,  

Stencil computation. Point-to-point Synchronization with Phasers

Module 1: Sections 4.2, 4.3

Topic 4.2 Lecture, Topic 4.2 Demonstration, Topic 4.3 Lecture, Topic 4.3 Demonstration

worksheet22

lec22-slides

Homework 4

 

 

WS22-solution
9	Mon	Mar

07

04	Lecture 23:

 Locks, Atomic variables

Fuzzy Barriers with Phasers

Module

2

1:

7

Section 4.

3

1

Topic 7

Topic 4.

3 Lecture

1 Lecture, Topic 4.1 Demonstration

 

worksheet23

lec23-slides

 

 

  

	Homework 3 (CP 1)	WS23-solution

Wed

Mar

09

06	Lecture 24:

Parallel Spanning Tree, other graph algorithms

Confinement & Monitor Pattern. Critical sections Global lock	Module 2: Sections 5.1, 5.2	Topic 5.1 Lecture, Topic 5.1 Demonstration, Topic 5.2 Lecture, Topic 5.2 Demonstration, Topic 5.6 Lecture, Topic 5.6 Demonstration	worksheet24	lec24-slides			WS24-solution
	Fri	Mar 08	Lecture 25:  Atomic variables, Synchronized statements	Module 2: Sections 5.4, 7.2	Topic 5.4 Lecture, Topic 5.4 Demonstration, Topic 7.2 Lecture	worksheet25	lec25-slides			WS25-solution
	Mon	Mar 11

  worksheet24 lec24-slides 

 

  

 

Fri

Mar 11

 Lecture 25: Linearizability of Concurrent ObjectsModule 2: 7.4Topic 7.4 Lectureworksheet25lec25-slides 

 

   

Mon

Mar 14

No class: Spring Break

 

	Wed

   

 

  

Mar 13	No class: Spring Break
	Fri	Mar 15

 WedMar 16

No class: Spring Break

 

10

   

 

   

 

Fri

Mar 18

No class: Spring Break

     

 

  

10

Mon

Mon

Mar 18

Mar 21

Lecture 26: Java Threads and Locks

- Soundness and progress guarantees

Module 2: Sections 7.

5

1, 7.3

Topic 7.

5 Lecture

 

1 Lecture, Topic 7.3 Lecture

worksheet26

lec26-slides

    

		WS26-solution

Wed

Mar

23

20	Lecture 27:

Dining Philosophers Problem

Read-Write Locks,  Soundness and progress guarantees

Module 2: Section 7.

6

3

Topic 7.

4

3 Lecture, Topic 7.

6

5 Lecture

worksheet27

lec27-slides

 

   

 

	Homework 3 (CP 2)	WS27-solution
	Fri	Mar

25

22	Lecture 28:

Read-Write Pattern. Read-Write Locks. Fairness & starvation

Dining Philosophers Problem

Topic 7.6 Lecture

Module 2: 7.3, 7.5Topic 7.3 Lecture, Topic 7.5 Lecture,  

worksheet28

lec28-slides

 

 

 

 

WS28-solution
11	Mon	Mar

28

25	Lecture 29:

Task Affinity and locality. Memory hierarchy

Linearizability of Concurrent Objects

Module 2: Sections 7.4

Topic 7.4 Lecture

  

worksheet29

lec29-slides

 

 

  

 

		WS29-solution
	Wed	Mar

30

27	Lecture 30:

Reactor Pattern. Web servers 

Parallel Spanning Tree, other graph algorithms

worksheet30

lec30-slides

 

   

 

Fri

Apr 01

Lecture 31: Scan Pattern. Parallel Prefix Sum, uses and algorithms

 

		WS30-solution
	Fri	Mar 29	Lecture 31: Message-Passing programming model with Actors	Module 2: Sections 6.1, 6.2	Topic 6.1 Lecture, Topic 6.1 Demonstration,   Topic 6.2 Lecture, Topic 6.2 Demonstration

  

worksheet31

lec31-slides

Homework 5

 

 

WS31-solution
12	Mon	Apr

04

01	Lecture 32:

Data-Parallel Programming model. Loop-Level Parallelism, Loop ChunkingModule 1: Sections 3.1, 3.2, 3.3Topic 3.1 Lecture , Topic 3.1 Demonstration , Topic 3.2 Lecture,  Topic 3.2 Demonstration,

Active Object Pattern. Combining Actors with task parallelism

Module 2: Sections 6.3, 6.4

Topic 6.3 Lecture, Topic 6.3 Demonstration,   Topic 6.4 Lecture, Topic 6.4

Topic 3.3 Lecture,  Topic 3.3

Demonstration

worksheet32

lec32-slides

 

 

  

Homework 4

Homework 3 (All)

WS32-solution

 

	Wed	Apr

06

03	Lecture 33:

Barrier Synchronization with phasersModule 1: Section 3.4Topic 3.4 Lecture ,   Topic 3.4 Demonstration

Task Affinity and locality. Memory hierarchy

worksheet33

lec33-slides

 

   

 

		WS33-solution
	Fri	Apr

08

05	Lecture 34:

  Stencil computation. Point-to-point Synchronization with PhasersModule 1: Section 4.2, 4.3Topic 4.2 Lecture ,   Topic 4.2 Demonstration, Topic 4.3 Lecture,  Topic 4.3 Demonstration

Eureka-style Speculative Task Parallelism

worksheet34

lec34-slides

 

 

  

		WS34-solution
13	Mon	Apr

11

08

Lecture 35: Message-Passing programming model with ActorsModule 2: 6.1, 6.2Topic 6.1 Lecture ,   Topic 6.1 Demonstration ,   Topic 6.2 Lecture, Topic 6.2 Demonstration

No class: Solar Eclipse
	Wed	Apr 10	Lecture 35: Scan Pattern. Parallel Prefix Sum			worksheet35	lec35-slides

 

 

  

Homework 4 (CP 1)

WS35-solution

 WedApr 13Lecture 36: Active Object Pattern. Combining Actors with task parallelismModule 2: 6.3, 6.4

Topic 6.3 Lecture ,   Topic 6.3 Demonstration ,   Topic 6.4 Lecture, Topic 6.4 Demonstration

worksheet36lec36-slides     

	Fri	Apr

15

12

Lecture

37: Eureka-style Speculative Task Parallelism  worksheet37lec37-slides   

36: Parallel Prefix Sum applications

worksheet36

lec36-slides

WS36-solution

 

14	Mon	Apr

18

15

Lecture

38lec38

37: Overview of other models and frameworks

   

lec37-slides

     

	Wed	Apr

20

17

Lecture

39

38: Course Review (Lectures 19-

38lec39

34)

  

lec38-slides

     

	Homework 4 (All)
	Fri	Apr

22

19

Lecture

40

39: Course Review (Lectures 19-

38

34)

   lec40

lec39-slides

    

Lab Schedule

Grading, Honor Code Policy, Processes and Procedures

...

Labs must be submitted by the following Monday at 11:59pm3pm.  Labs must be checked off by a TA.

Worksheets should be completed by the deadline listed in Canvas before the start of the following class (for full credit) so that solutions to the worksheets can be discussed in the next class.

...