Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

COMP 322: Fundamentals of Parallel Programming (Spring

...

2022)

 

Austin Bae, Avery Whitaker, Aydin Zanager, Eduard Danalache, Frank Chen, Hamza Nauman, Harrison Brown, Jahid Adam, Jeemin Sim, Kitty Cai, Madison Lewis, Ryan Han, Teju Manchenella, Victor Gonzalez, Victoria Nazari

InstructorInstructors:

Mackale Joyner, DH 2071

Head TA:Abbey Baker

Co-Instructor:

2063

Zoran Budimlić, DH 30813003

Graduate TAs:Jonathan Sharman, Srdjan MilakovicAdrienne Li, Austin Hushower, Claire Xu, Diep Hoang, Hunena Badat, Maki Yu, Mantej Singh, Rose Zhang, Victor Song, Yidi Wang  
Admin Assistant:Annepha Hurlock, annepha@rice.edu , DH 3122, 713-348-5186Undergraduate TAs: 

 

Piazza site:

https://piazza.com/classrice/j3w0pi8pl9s8sspring2022/comp322 (Piazza is the preferred medium for all course communications, but you can also send email to comp322-staff at rice dot edu if needed)

Cross-listing:

ELEC 323

Lecture location:

Sewall Hall 301Herzstein Amphitheater (online 1st 2 weeks)

Lecture times:

MWF 1:00pm - 1:50pm

Lab locations:

Sewall Hall 301Keck 100 (online 1st 2 weeks)

Lab times:

Thursday, 4Mon  3:00pm - 43:50pm (Austin, Claire)

Wed 4:30pm - 5:20pm (Hunena, Mantej, Yidi, Victor, Rose, Adrienne, Diep, Maki)

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 (course modules):

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.

...

3) Locality & Distribution: memory hierarchies, locality, cache affinity, data movement, message-passing (MPI), communication overheads (bandwidth, latency), MapReduce, accelerators, GPGPUs, CUDA, OpenCL.

To achieve these learning outcomes, each class period will include time for both instructor lectures and in-class exercises based on assigned reading and videos.  The lab exercises will be used to help students gain hands-on programming experience with the concepts introduced in the lectures.

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, MPI, OpenCL and CUDA. 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.

...

There are no required textbooks for the class. Instead, lecture handouts are provided for each module as follows.  You are expected to read the relevant sections in each lecture handout before coming to the lecture.  We will also provide a number of references in the slides and handouts.The links to the latest versions of the lecture handouts are included below:

There

...

There are also a few optional textbooks that we will draw from during the course.  You are encouraged to get copies of any or all of these books.  They will serve as useful references both during and after this course:

 

Finally, here are some additional resources that may be helpful for you:

Lecture Schedule

 

 

19 4  Parallel Speedup and Amdahl's Law 24 Finish Accumulators3 3   Jan 31 Java’s Fork/Join LibraryQuiz for Unit 2 02 Loop-Level Parallelism, Parallel Matrix Multiplication, Iteration Grouping (Chunking) 05  Barrier Synchronization 3 3 07Topic 3.7 Java Streams, Topic 3.7 Java Streams Parallelism in Java Streams, Parallel Prefix Sums 12 Iterative Averaging Revisited, SPMD pattern 35 Lecture , Topic .5 Demonstration , Topic 3.6 , .6   16 Point-to-point Synchronization with Phasers 4.2 42 lec22Lecture 24: Java Threads, Java synchronized statementlec2410Mon 19 21

Homework 4

(includes one intermediate checkpoint)

Homework 3 (all) Passing Interface (MPI), (start of Module 3)Topic 8 8. 8.3 Lecture,Fri Distributed Map-Reduce using Hadoop and Spark frameworksTopic 9.1 Lecture (optional, overlaps with video 2.4), Topic 9.2 Lecture, Topic 9.3 LectureAlgorithms based on (Scan) operationsApr 16Algorithms based on (Scan) operations, contd.Wed GPU ComputingFri 18

Week

Day

Date (20182022)

Lecture

Assigned Reading

Assigned Videos (see Canvas site for video links)

In-class Worksheets

Slides

Work Assigned

Work Due

Worksheet Solutions 

1

Mon

Jan 0810

Lecture 1: Task Creation and Termination (Async, Finish)

Module 1: Section 1.1

Topic 1.1 Lecture, Topic 1.1 Demonstration Introduction

 

 

worksheet1lec1-slidesslides  

 

 

WS1-solution 

 

Wed

Jan 1012

Lecture 2:  Computation Graphs, Ideal Parallelism

Module 1: Sections 1.2, 1.3Topic 1.2 Lecture, Topic 1.2 Demonstration, Topic 1.3 Lecture, Topic 1.3 Demonstrationworksheet2lec2-slides

Homework 1

 

 FriJan 12Lecture 3: Abstract Performance Metrics, Multiprocessor SchedulingModule 1: Section 1.4Topic 1.4 Lecture, Topic 1.4 Demonstrationworksheet3lec3-slides

 

 

2

Mon

Jan 15

No lecture, School Holiday (Martin Luther King, Jr. Day)Functional Programming

GList.java worksheet2lec02-slides

 

 

WS2-solution 
 FriJan 14Lecture 3: Higher order functions  worksheet3 lec3-slides   

 

 WS3-solution 

2

Mon

Jan 17

No class: MLK

        

 

Wed

Jan 17

No lecture, Rice closed due to weather

  

19

Lecture 4: Lazy Computation

LazyList.java

Lazy.java

 worksheet4lec4-slides  Quiz for Unit 1WS4-solution 

 

Fri

Jan

21

Lecture

5: Java Streams

 Module 1: Section 1.5Topic 1.5 Lecture, Topic 1.5 Demonstrationworksheet4lec4-slides   worksheet5lec5-slidesHomework 1 WS5-solution 
3MonJan 2224

Lecture 5: Future Tasks, Functional Parallelism ("Back to the Future")6: Map Reduce with Java Streams

Module 1: Section 2.14Topic 2.1 4 Lecture, Topic 2.1 4 Demonstration  worksheet5worksheet6lec5lec6-slides

 

 WS6-solution 

 

Wed

Jan

26

Lecture 7:

Futures

Module 1: Section 2.31Topic 2.1 Lecture , Topic 2.1 Demonstrationworksheet7lec7-slidesHomework 2

 

Homework 1 WS7-solution 

 

Fri

Jan 2628

Lecture 8: Memoization, Map Reduce  Computation Graphs, Ideal Parallelism

Module 1: Section 2Sections 1.2 & 2, 1.43Topic 21.2 Lecture, Topic 21.2 Demonstration, Topic 21.4 3 Lecture, Topic 21.4 3 Demonstrationworksheet8lec8-slides Quiz for Unit 1 WS8-solution 

4

Mon

 

Jan 2931 Lecture 9: Data Races, Functional & Structural DeterminismAsync, Finish, Data-Driven Tasks 

Module 1:

Sections 2

Section 1.

5

1,

2

4.

6

5

 

Topic

2

1.

5

1 Lecture, Topic

2

1.

5

1 Demonstration, Topic

2

4.

6

5 Lecture, Topic

2

4.

6

5 Demonstration

   

worksheet9

lec9-slidesslides   WS9-solution 
 WedFeb 02Module 1: Sections 2.7, 2.8Topic 2.7 Lecture, Topic 2.8 Lecture,Lecture 10: Event-based programming model

 

  worksheet10lec10-slides  WS10-solution 
 FriFeb 04Module 1: Sections 3.1, 3.2, 3.3

Topic 3.1 Lecture , Topic 3.1 Demonstration , Topic 3.2 Lecture, Topic 3.2 Demonstration, Topic 3.3 Lecture , Topic 3.3 Demonstration

Lecture 11: GUI programming as an example of event-based,
futures/callbacks in GUI programming
  worksheet11lec11-slides Homework 2Homework 1WS11-solution 
5

Mon

Feb

07

Lecture 12: Scheduling/executing computation graphs
Abstract performance metrics
Module 1: Section 31.4Topic 1.4 Lecture , Topic 1.4 Demonstrationworksheet12lec12-slides  WS12-solution 

 

Wed

Feb

09

 

Lecture 13:

Parallel Speedup, Critical Path, Amdahl's Law

Module 1: Section 1.5

Topic 1.5 Lecture , Topic 1.5 Demonstration

worksheet13lec13-slides

Homework 3 (includes 2 intermediate checkpoints)

Homework 2

-

  WS13-solution 

 

Fri

Feb 0911

No class: Spring Recess

 

        
6

Mon

Feb

14

Lecture 14:

Accumulation and reduction. Finish accumulators

Module 1: Sections 3Section 2.5, 3.6Topic 2.3 Lecture   Topic 2.3 Demonstrationworksheet14lec14-slidesQuiz for Unit 3Quiz for Unit 2

 

  WS14-solution 

 

Wed

Feb 1416Topic 4.5 Lecture   Topic 4.5 Demonstration, Topic 4.3 Lecture,  Topic 4.3 Demonstration

Lecture 15:   Data-Driven Tasks

Module 1: Sections 4.5, 4.2, 4.3

Recursive Task Parallelism  

  worksheet15lec15-slides

 

 

 WS15-solution 
 FriFeb 18

Lecture 16:

Data Races, Functional & Structural Determinism

Module 1: Sections 42.5, 2.6Topic 2.5 Lecture ,  Topic 2.5 Demonstration,  Topic 2.6 Lecture,  Topic 2.6 Demonstrationworksheet16 lec16-slidesHomework 3Homework 2WS16-solution Quiz for Unit 3

7

Mon

Feb 1921

Lecture 17: Midterm Summary Midterm Review

   lec17-slides   

Wed

Feb 21

Midterm Review (interactive Q&A)

 

 

    

 

Fri

Wed

Feb 23

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

  worksheet18lec18lec18-slides Homework 3, Checkpoint-1

8

Mon

Feb 26

Lecture 19: Pipeline Parallelism, Signal Statement, Fuzzy Barriers

Module 1: Sections 4.4, 4.1Topic 4.4 Lecture ,   Topic 4.4 Demonstration, Topic 4.1 Lecture,  Topic 4.1 Demonstration, WS18-solution 

 

Fri

Feb 25 

Lecture 19: Fork/Join programming model. OS Threads. Scheduler Pattern 

 Topic 2.7 Lecture, Topic 2.7 Demonstration, Topic 2.8 Lecture, Topic 2.8 Demonstration, worksheet19lec19-slidesQuiz for Unit 4  WS19-solution Wed

8

Mon

Feb 28

Lecture 20: Confinement & Monitor Pattern. Critical sections, Isolated construct, Parallel Spanning Tree algorithm, Atomic variables (start of Module 2)
Global lock

Module 2: Sections 5.1, 5.2, 5.3, 5.4, 5.6 Topic 5.1 Lecture, Topic 5.1 Demonstration, Topic 5.2 Lecture, Topic 5.2 Demonstration, Topic 5.3 Lecture, Topic 5.3 Demonstration, Topic 5.4 Lecture, Topic 5.4 Demonstration, Topic 5.6 Lecture, Topic 5.6 Demonstrationworksheet20lec20-slides        WS20-solution Fri

 

Wed

Mar 02

Lecture 21:  Read-Write Isolation, Review of Phasers Atomic variables, Synchronized statements

Module 2:

Section

Sections 5.

5

4, 7.2

Topic 5.5 4 Lecture, Topic 5.5 Demonstration4 Demonstration, Topic 7.2 Lectureworksheet21lec21-slidesQuiz for Unit 5

Quiz for Unit 4

  WS21-solution 

 

Fri

Mar 04

Lecture 22: Parallel Spanning Tree, other graph algorithms 

  worksheet22lec22-slidesHomework 4

Homework 3

WS22-solution 

9

Mon

Mar 0507

Lecture 22: Actors23: Java Threads and Locks

Module 2: 6Sections 7.1, 67.23

Topic

6

7.1 Lecture,

 

Topic

6.1 Demonstration ,   Topic 6.2 Lecture, Topic 6.2 Demonstration
worksheet22

7.3 Lecture

worksheet23 lec23-slides  

 

WS23-solution 

 

Wed

Mar 0709

Lecture 23:  Actors (contd)24: Java Locks - Soundness and progress guarantees  

Module 2: 6.3, 6.4, 6.5, 6.6Topic 6.3 Lecture, Topic 6.3 Demonstration, Topic 6.4 Lecture , Topic 6.4 Demonstration,   Topic 6.5 Lecture, Topic 6.5 Demonstration, Topic 6.6 Lecture, Topic 6.6 Demonstrationworksheet23 lec23-slides

Quiz for Unit 6

Homework 3, Checkpoint-2

 

Fri

Mar 09

7.5Topic 7.5 Lecture worksheet24 lec24-slides 

 

WS24-solution 

 

Fri

Mar 11

 Lecture 25: Dining Philosophers Problem  Module 2: 7.1, 7.26Topic 7.1 Lecture, Topic 7.2 Lectureworksheet246 Lectureworksheet25lec25-slides  Quiz for Unit 5

 

WS25-solution

M-F

Mar 12 - Mar 16

 
 

Mon

Mar 14

No class: Spring Break

     

 

  
 WedMar 16

Lecture 25: Java synchronized statement (contd), wait/notify

Module 2: 7.2Topic 7.2 Lectureworksheet25 lec25-slidesNo class: Spring Break    

 

   Wed

 

Fri

Mar

18

Lecture 26: Java Locks, Linearizability of Concurrent Objects

Module 2: 7.3, 7.4Topic 7.3 Lecture, Topic 7.4 Lectureworksheet26 lec26-slides

No class: Spring Break

     

 

  

10

Mon

Mar 21

Lecture 26: N-Body problem, applications and implementations 

  worksheet26lec26-slides   WS26-solution Fri

 

Wed

Mar 23

Lecture 27: Safety and Liveness Properties, Java Synchronizers, Dining Philosophers Problem Read-Write Locks, Linearizability of Concurrent Objects

Module 2: 7.53, 7.64Topic 7.5 3 Lecture, Topic 7.6 4 Lectureworksheet27lec27-slides Quiz for Unit 7

Quiz for Unit 6

11

Mon

Mar 26 

 WS27-solution 

 

Fri

Mar 25

Lecture 28: Message

 

-Passing programming model with Actors

Module 2: 6.1, 6.2Topic 6.1 Lecture, Topic 6.1 Demonstration,   Topic 6.2 Lecture, Topic 6.2 Demonstrationworksheet28lec28-slides

 

 

 

Wed

WS28-solution 

11

Mon

Mar 28

Topic 8.4 Lecture, Topic 8.5 Lecture, Topic 8 Demonstration Video

Lecture 29:   Message Passing Interface (MPI, contd)

 

Active Object Pattern. Combining Actors with task parallelism 

Module 2: 6.3, 6.4Topic 6.3 Lecture, Topic 6.3 Demonstration,   Topic 6.4 Lecture, Topic 6.4 Demonstrationworksheet29lec29-slides

Quiz for Unit 8 

 

WS29-solution 

 

Wed

Mar 30

Lecture 30:

 

Task Affinity and locality. Memory hierarchy 

  worksheet30lec30-slides

 

Quiz for Unit 7

12

Mon

Apr 02 WS30-solution 

 

Fri

Apr 01

Topic 9.4 Lecture, Topic 9.5 Lecture, Unit 9 Demonstration

Lecture 31: TF-IDF and PageRank Algorithms with Map-Reduce

 

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

Module 1: Sections 3.1, 3.2, 3.3Topic 3.1 Lecture, Topic 3.1 Demonstration , Topic 3.2 Lecture,  Topic 3.2 Demonstration, Topic 3.3 Lecture,  Topic 3.3 Demonstrationworksheet31lec31-slidesQuiz for Unit 9

 

 

Wed

Homework 5

Homework 4

WS31-solution 

12

Mon

Apr 04

Lecture 32:  Partitioned Global Address Space (PGAS) programming models  Barrier Synchronization with PhasersModule 1: Section 3.4Topic 3.4 Lecture,  Topic 3.4 Demonstrationworksheet32lec32-slides

 

Homework 4 Checkpoint-1 

WS32-solution Fri

 

Wed

Apr 06

Lectures 10.1 - 10.5, Unit 10 Demonstration (all videos optional – unit 10 has no quiz)

Lecture 33: Combining Distribution and Multithreading

 

  Stencil computation. Point-to-point Synchronization with Phasers

Module 1: Section 4.2, 4.3

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

worksheet33lec33-slides

 

Quiz for Unit 8

13

Mon

Apr 09 WS33-solution 

 

Fri

Apr 08

 

Lecture 34: Task Affinity with Places

 

 Fuzzy Barriers with Phasers

Module 1: Section 4.1Topic 4.1 Lecture, Topic 4.1 Demonstrationworksheet34lec34-slides 

 

Wed

WS34-solution 

13

Mon

Apr 11

Lecture 35: Eureka-style Speculative Task Parallelism 

 

worksheet35lec35-slides

Homework 5

Homework 4 (all)

 

Fri

 

 

WS35-solution 
 WedApr 13Lecture 36: Scan Pattern. Parallel Prefix Sum 

 

worksheet36lec36-slides Quiz for Unit 9

14

Mon

 WS36-solution 
 FriApr 15Lecture 37: Parallel Prefix Sum applications  worksheet37lec37-slides    
14MonApr 18Lecture 38: Overview of other models and frameworks  worksheet38 lec38-slides    
 WedApr 20Lecture 39: Course Review (Lectures 19-38)   lec39-slides 

Homework 5

-   
  Fri  Apr 22Lecture 40: Course Review (Lectures 19-38)    lec40-slides  Homework 5      

Lab Schedule

0  Setup 25Futures- Java's ForkJoin Framework-Mar 15 - Spring BreakIsolated Statement and Atomic VariablesJava Threads, Java LocksMessage Passing Interface (MPI)  19Apache Spark  

Lab #

Date (20182022)

Topic

Handouts

Code Examples

1

Jan 10

Infrastructure

setup

lab0-handout

-

1

Jan 11

Async-Finish Parallel Programming with abstract metrics

lab1-handout

-
 
2Jan 17Functional Programminglab2-handout 

3

Feb 01

Cutoff Strategy and Real World Performance

Jan 24

Java Streams

lab3-handout
-
 
4

Feb 15

Jan 31Futureslab4-handout 

5

-

5

Mar 01

DDFs

 

Feb

22

 

No lab this week - Midterm Exam -

07

Data-Driven Tasks

lab5-handout- 
6

Mar 05

Loop-level ParallelismFeb 14

Async / Finish

lab6-handout handout- 
-

Feb 21

No lab this week

(Midterm)

  
7

Mar 22

Feb 28Recursive Task Cutoff Strategylab7-handout 
8Mar 2907ActorsJava Threadslab8-handout 9

Apr 05

-

Mar 14

No lab this week (Spring Break)

  
9Mar 21Concurrent Listslab9-handout 
10

Apr 12

Mar 28Actorslab10-handout 
11

Apr

04

Loop Parallelism

lab11-handout 

 -

 

Eureka-style Speculative Task ParallelismApr 11

No lab this week

  

-

 

Apr 18

No lab this week

  

Grading, Honor Code Policy, Processes and Procedures

Grading will be based on your performance on five homeworks four homework assignments (weighted 40% in all), two exams (weighted 40% in all), weekly lab exercises (weighted 10% in all), online quizzes (weighted 5% in all), and in-class worksheets (weighted 5% in all).

The purpose of the homeworks homework is to give you practice in solving problems that deepen your understanding of concepts introduced in class. Homeworks are Homework is due on the dates and times specified in the course schedule.  No late submissions (other than those using slip days mentioned below) will be accepted.

The slip day policy for COMP 322 is similar to that of COMP 321. All students will be given 3 slip days to use throughout the semester. When you use a slip day, you will receive up to 24 additional hours to complete the assignment. You may use these slip days in any way you see fit (3 days on one assignment, 1 day each on 3 assignments, etc.). Slip days will be automatically tracked through the Autograder, more details are available later in this document and in the Autograder user guide. using the README.md file. Other than slip days, no extensions will be given unless there are exceptional circumstances (such as severe sickness, not because you have too much other work). Such extensions must be requested and approved by the instructor (via e-mail, phone, or in person) before the due date for the assignment. Last minute requests are likely to be denied.

Labs must be checked off submitted by a TA by the following Monday Wednesday at 114:59pm.30pm.  Labs must be checked off by a TA.

Worksheets should be completed in class for full credit.  For partial credit, a worksheet can be turned in before the start of the class following the one in which the worksheet for distributed, by the deadline listed in Canvas so that solutions to the worksheets can be discussed in the next class.

You will be expected to follow the Honor Code in all homeworks and homework and exams.  The following policies will apply to different work products in the course:

  • In-class worksheets: You are free to discuss all aspects of in-class worksheets with your other classmates, the teaching assistants and the professor during the class. You can work in a group and write down the solution that you obtained as a group. If you work on the worksheet outside of class (e.g., due to an absence), then it must be entirely your individual effort, without discussion with any other students.  If you use any material from external sources, you must provide proper attribution.
  • Weekly lab assignments: You are free to discuss all aspects of lab assignments with your other classmates, the teaching assistants and the professor during the lab.  However, all code and reports that you submit are expected to be the result of your individual effort. If you work on the lab outside of class (e.g., due to an absence), then it must be entirely your individual effort, without discussion with any other students.  If you use any material from external sources, you must provide proper attribution (as shown here).
  • HomeworksHomework: All submitted homeworks are homework is expected to be the result of your individual effort. You are free to discuss course material and approaches to problems with your other classmates, the teaching assistants and the professor, but you should never misrepresent someone else’s work as your own. If you use any material from external sources, you must provide proper attribution.
  • Quizzes: Each online quiz will be an open-notes individual test.  The student may consult their course materials and notes when taking the quizzes, but may not consult any other external sources.
  • Exams: Each exam will be a closedopen-book, closedopen-notes, and closedopen-computer individual written test, which must be completed within a specified time limit.  No class notes or external materials may be consulted when taking the exams.

...