COMP 322: Fundamentals of Parallel Programming (Spring
...
2023)
Instructor: | Mackale Joyner, DH 2063 | Head TAs: | Admin Assistant: | Annepha Hurlock, annepha@rice.edu, DH 3122, 713-348-5186 | Undergraduate TAs: | Mohamed Abead, Chase Hartsell, Taha Hasan, Harrison Huang, Jerry Jiang, Jasmine Lee, Michelle Lee, Hung Nguyen, Quang Nguyen, Ryan Ramos, Oscar Reynozo, Delaney Schultz, Tina Wen, Raiyan Zannat, Kailin Zhang | |
---|---|---|---|---|---|---|---|
Piazza site: | https://piazza.com/rice/spring2021spring2022/comp322 (Piazza is the preferred medium for all course communications) | Cross-listing: | ELEC 323 | ||||
Lecture location: | Fully OnlineTBD | Lecture times: | MWF 1:30pm 00pm - 21:25pm50pm | ||||
Lab locations: | Fully OnlineTBD | Lab times: | Tu 1Mon 3:30pm 00pm - 2:25pm, Th 3:50pm () Tue 4:50pm 00pm - 54:45pm50pm () |
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:
- Module 1 handout (Parallelism)
- Module 2 handout (Concurrency)There is no lecture handout for Module 3 (Distribution and Locality). The instructors will refer you to optional resources to supplement the lecture slides and videos.
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:
- 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
Finally, here are some additional resources that may be helpful for you:
- Slides titled "MPI-based Approaches for Java" by Bryan Carpenter
Lecture Schedule
Week | Day | Date (20212022) | Lecture | Assigned Reading | Assigned Videos (see Canvas site for video links) | In-class Worksheets | Slides | Work Assigned | Work Due | Worksheet Solutions | ||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Mon | Jan 2509 | Lecture 1: Task Creation and Termination (Async, Finish) | Module 1: Section 1.1 | Topic 1.1 Lecture, Topic 1.1 DemonstrationIntroduction |
| worksheet1 | lec1-slides | worksheet1 | lec1-slides |
|
| WS1-solution | |||||||||||||||||
| Wed | Jan 2711 | Lecture 2: Computation Graphs, Ideal Parallelism | Module 1: Sections 1.2, 1.3 | Topic 1.2 Lecture, Topic 1.2 Demonstration, Topic 1.3 Lecture, Topic 1.3 Demonstration | worksheet2 | lec2-slides | Homework 1Functional Programming | GList.java | worksheet2 | lec02-slides |
|
| WS2-solution | ||||||||||||||||
Fri | Jan 2913 | Lecture 3: Abstract Performance Metrics, Multiprocessor Scheduling | Module 1: Section 1.4 | Topic 1.4 Lecture, Topic 1.4 Demonstration | worksheet3 | Higher order functions | worksheet3 | lec3-slides lec3-slides |
| WS3-solution | ||||||||||||||||||||
2 | Mon | Jan 16 | No class: MLK | |||||||||||||||||||||||||||
| Wed | Jan 18 | Lecture 4: Lazy Computation | LazyList.java Lazy.java | Feb 01 | Lecture 4: Parallel Speedup and Amdahl's Law | Module 1: Section 1.5 | Topic 1.5 Lecture, Topic 1.5 Demonstration | worksheet4 | lec4-slides | Quiz for Unit 1 | WS4-solution | ||||||||||||||||||
| WedFri | Feb 03Jan 20 | Lecture 5: | Future Tasks, Functional Parallelism ("Back to the Future")Java Streams | Module 1: Section 2.1 | Topic 2.1 Lecture, Topic 2.1 Demonstrationworksheet5 | lec5-slides | Homework 1 | WS5-solution | |||||||||||||||||||||
3 | FriMon | Feb 05Jan 23 | Lecture 6: Finish Accumulators Map Reduce with Java Streams | Module 1: Section 2.34 | Topic 2.3 4 Lecture, Topic 2.3 4 Demonstration | worksheet6 | lec6-slides |
| Quiz for Unit 1 | WS6-solution | ||||||||||||||||||||
| 3 | Mon | Feb 08Wed | Jan 25 | Lecture 7: | Map ReduceFutures | Module 1: Section 2.41 | Topic 2. | 41 Lecture , Topic 2. | 41 Demonstration | worksheet7 | lec7-slides |
| WS7-solution | ||||||||||||||||
| WedFri | Feb 10Jan 27 | Lecture 8: Data Races, Functional & Structural Determinism Computation Graphs, Ideal Parallelism | Module 1: Section Sections 1.2.5, 21.63 | Topic 1.2 .5 Lecture, Topic 1.2 .5 Demonstration, Topic 21.6 3 Lecture, Topic 21.6 3 Demonstration | worksheet8 | lec8-slides | Homework 2 | WS8-solutionHomework 1 | |||||||||||||||||||||
4 | Mon
| Fri | Feb 12 | Jan 30 | Lecture 9: Java’s Fork/Join LibraryAsync, Finish, Data-Driven Tasks | Module 1: Sections 2Section 1. 71, 24. 85
| Topic 21. 71 Lecture, Topic 2.8 Lecture1.1 Demonstration, Topic 4.5 Lecture, Topic 4.5 Demonstration | worksheet9 | lec9-slidesslides | Quiz for Unit 2 | WS9-solution | |||||||||||||||||||
4 | MonWed | Feb | 1501 | Lecture 10: | Loop-Level Parallelism, Parallel Matrix Multiplication Event-based programming model
| Module 1: Sections 3.1, 3.2 | Topic 3.1 Lecture , Topic 3.1 Demonstration , Topic 3.2 Lecture, Topic 3.2 Demonstration | worksheet10 | lec10-slides | Homework 1 | WS10-solution | |||||||||||||||||||
WedFri | Feb 17 | Spring "Sprinkle" Day (no class) | 03 | Lecture 11: GUI programming as an example of event-based, futures/callbacks in GUI programming | worksheet11 | lec11-slides | Homework 2 | WS11-solution | Fri | |||||||||||||||||||||
5 | Mon | Feb | 1906 | Lecture | 11: Iteration Grouping (Chunking), Barrier Synchronization12: Scheduling/executing computation graphs Abstract performance metrics | Module 1: Sections 3.3, 3Section 1.4 | Topic | 3.3 Lecture , Topic 3.3 Demonstration, Topic 3.1.4 Lecture | , | Topic | 31.4 Demonstration | worksheet11worksheet12 | lec11lec12-slides | Quiz for Unit 2 | WS12-solution | |||||||||||||||
5 | Wed | Mon | Feb 2208 | Lecture 12: Parallelism in Java Streams, Parallel Prefix Sums 13: Parallel Speedup, Critical Path, Amdahl's Law | Module 1: Section 31.75 | Topic Topic 3.7 Java Streams1.5 Lecture , Topic 31. 7 Java Streams5 Demonstration | worksheet12worksheet13 | lec12lec13-slides | WS13-solution | |||||||||||||||||||||
| WedFri | Feb | 2410 | Lecture 13: Iterative Averaging Revisited, SPMD pattern | Module 1: Sections 3.5, 3.6 | Topic 3.5 Lecture , Topic 3.5 Demonstration , Topic 3.6 Lecture, Topic 3.6 Demonstration | worksheet13 | lec13-slides | Homework 3 (includes one intermediate checkpoint) Quiz for Unit 3 | Homework 2No class: Spring Recess
| ||||||||||||||||||||
6 | FriMon | Feb | 2613 | Lecture 14: | Data-Driven Tasks Accumulation and reduction. Finish accumulators | Module 1: Sections 4Section 2.53 | Topic | 42. | 53 Lecture Topic | 42. | 53 Demonstration | worksheet14 | lec14-slides | WS14-solution | ||||||||||||||||
| 6Wed | Mon | Mar 01 | Feb 15 | Lecture 15: Recursive Task Parallelism | Spring "Sprinkle" Day (no class) | worksheet15 | lec15-slides |
| WS15-solution | ||||||||||||||||||||
Wed | Fri | Mar 03Feb 17 | Lecture | 15: Point-to-point Synchronization with Phasers16: Data Races, Functional & Structural Determinism | Module 1: Section 4Sections 2.5, 2, 4.36 | Topic | 42. | 25 Lecture , Topic | 42. | 25 Demonstration, Topic | 42. | 36 Lecture, | Topic 4.3Topic 2.6 Demonstration | worksheet15worksheet16 | lec15lec16-slides | Homework 3 | Homework 2 | WS16-solution | ||||||||||||
7 | FriMon | Mar 05Feb 20 | Lecture 16: Pipeline Parallelism, Signal Statement, Fuzzy Barriers | Module 1: Sections 4.4, 4.1 | Topic 4.4 Lecture , Topic 4.4 Demonstration, Topic 4.1 Lecture, Topic 4.1 Demonstration | worksheet16 | lec16-slides | Quiz for Unit 4 | Quiz for Unit 3 | 17: Midterm Review | lec17-slides | 7 | Mon | Mar 08 | Lecture 17: Midterm Review | lec17-slides | ||||||||||||||
| WedMar 10 | Feb 22 | Lecture 18: Limitations of Functional parallelism. | worksheet18 | lec18lec18-slides | WS18-solution | ||||||||||||||||||||||||
| FriMar 12 | Feb 24 | Lecture 19: Critical Sections, Isolated construct (start of Module 2) | Module 2: Sections 5.1, 5.2, 5.6, | Fork/Join programming model. OS Threads. Scheduler Pattern | Topic 2.7 Lecture, Topic 2.7 Demonstration, Topic 2.8 Lecture, Topic 2.8 Demonstration, Topic 5.1 Lecture, Topic 5.1 Demonstration, Topic 5.2 Lecture, Topic 5.2 Demonstration, Topic 5.6 Lecture, Topic 5.6 Demonstration | worksheet19 | lec19-slides | Homework 3, Checkpoint-1 | WS19-solution | ||||||||||||||||||||
8 | MonMar | 15Feb 27 | Lecture 20: Parallel Spanning Tree algorithm, Atomic variables Confinement & Monitor Pattern. Critical sections | Module 2: Sections 5.31, 5.42, 5.56 | Topic 5.1 Lecture, Topic 5.3 1 Demonstration, Topic 5.4 2 Lecture, Topic 5.4 2 Demonstration, Topic 5.5 6 Lecture, Topic 5.5 6 Demonstration | worksheet20 | lec20-slides | WS20-solution | ||||||||||||||||||||||
| Wed | Mar 1701 | Lecture 21: Actors Atomic variables, Synchronized statements | Module 2: 6Sections 5. 14, 67.2 | Topic 65.1 4 Lecture, Topic 65.1 4 Demonstration, Topic 67.2 Lecture, Topic 6.2 Demonstration | worksheet21 | lec21-slides | WS21-solution | ||||||||||||||||||||||
| Fri | Mar 1903 | Lecture 22: Actors (contd) | Module 2: 6.3, 6.4, 6.5 | Parallel Spanning Tree, other graph algorithms | Topic 6.3 Lecture, Topic 6.3 Demonstration, Topic 6.4 Lecture , Topic 6.4 Demonstration, Topic 6.5 Lecture, Topic 6.5 Demonstration | worksheet22 | lec22-slides | Quiz for Unit 4 | Homework 4 | Homework 3 | WS22-solution | ||||||||||||||||||
9 | Mon | Mar 2206 | Lecture 23: Actors (contd)Java Threads and Locks | Module 2: 6.6Sections 7.1, 7.3 | Topic 67. 61 Lecture, Topic 67. 6 Demonstration3 Lecture | worksheet23 | lec23-slides | Quiz for Unit 5
| WS23-solution | |||||||||||||||||||||
| Wed | Mar | 2408 | Lecture 24: | Java Threads, Java synchronized statementJava Locks - Soundness and progress guarantees | Module 2: 7.1, 7.25 | Topic 7. | 1 Lecture, Topic 7.25 Lecture | worksheet24 | lec24-slides |
| WS24-solution | ||||||||||||||||||
| Fri | Mar 10 | Lecture 25: Dining Philosophers Problem | Module 2: 7.6 | Topic 7.6 Lecture | worksheet25 | lec25-slides |
| WS25-solution | |||||||||||||||||||||
Mon | Mar 13 | No class: Spring Break 26Spring "Sprinkle" Day (no class) |
| |||||||||||||||||||||||||||
Mon | Wed | Mar | 2915 | Lecture 25: Java Threads, Java synchronized statement (contd), wait/notify | Module 2: 7.1, 7.2 | Topic 7.1 Lecture, Topic 7.2 Lecture | lec25-slides | No class: Spring Break |
| |||||||||||||||||||||
| WedFri | Mar 3117 | Lecture 26: Java Threads (exercise)No class: Spring Break | lec26-handout | Homework 3 (all) |
| Fri | |||||||||||||||||||||||
10 | Mon | Mar 20Apr 02 | Lecture 27: Java Locks | Module 2: 7.3 | Topic 7.3 Lecture | lec27-slides | Quiz for Unit 6 | 26: N-Body problem, applications and implementations | worksheet26 | lec26-slides | WS26-solutionQuiz for Unit 5 | |||||||||||||||||||
| 11 | Mon | Apr 05Wed | Mar 22 | Lecture | 2827: Read-Write Locks, Linearizability of Concurrent Objects | Module 2: 7.43, 7.4 | Topic 7.3 Lecture, Topic 7.4 Lecture | worksheet27 | lec28lec27-slides | Homework 4 (includes one intermediate checkpoint) |
| WS27-solution | |||||||||||||||||
| WedFri | Apr 07Mar 24 | Lecture | 29: Java Locks (exercise)lec29-handout |
| |||||||||||||||||||||||||
| Fri | Apr 09 | Lecture 30: Safety and Liveness Properties, Java Synchronizers, Dining Philosophers Problem | Module 2: 7.5, 7.6 | Topic 7.5 Lecture, Topic 7.6 Lecture | lec30-slides | Quiz for Unit 7 | Quiz for Unit 6 | ||||||||||||||||||||||
12 | Mon | Apr 12 | Lecture 31: Message Passing Interface (MPI), (start of Module 3) | Topic 8.1 Lecture, Topic 8.2 Lecture, Topic 8.3 Lecture | lec31-slides |
| ||||||||||||||||||||||||
28: Message-Passing programming model with Actors | Module 2: 6.1, 6.2 | Topic 6.1 Lecture, Topic 6.1 Demonstration, Topic 6.2 Lecture, Topic 6.2 Demonstration | worksheet28 | lec28-slides |
|
| WS28-solution | |||||||||||||||||||||||
11 | Mon | Mar 27 | Lecture 29: Active Object Pattern. Combining Actors with task parallelism | Module 2: 6.3, 6.4 | Topic 6.3 Lecture, Topic 6.3 Demonstration, Topic 6.4 Lecture, Topic 6.4 Demonstration | worksheet29 | lec29-slides |
|
| WS29-solution | ||||||||||||||||||||
| Wed | Mar 29 | Lecture 30: Task Affinity and locality. Memory hierarchy | worksheet30 | lec30-slides |
| WS30-solution | |||||||||||||||||||||||
| Fri | Mar 31 | Lecture 31: Data-Parallel Programming model. Loop-Level Parallelism, Loop Chunking | Module 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 | worksheet31 | lec31-slides | Homework 5 | Homework 4 | WS31-solution | ||||||||||||||||||||
12 | Mon | Apr 03 | Lecture 32: Barrier Synchronization with Phasers | Module 1: Section 3.4 | Topic 3.4 Lecture, Topic 3.4 Demonstration | worksheet32 | lec32-slides |
|
| WS32-solution | ||||||||||||||||||||
| Wed | Apr 05 | Lecture 33: 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 | worksheet33 | lec33-slides |
| WS33-solution |
| Wed | Apr 14 | Lecture 32: Message Passing Interface (MPI, contd) | Topic 8.4 Lecture | lec32-slides | Homework 4 Checkpoint-1 | ||||||||||||||
| Fri | Apr 1607 | Lecture 33: Message Passing Interface (MPI, contd) | Topic 8.5 Lecture, Topic 8 Demonstration Video | 34: Fuzzy Barriers with Phasers | Module 1: Section 4.1 | Topic 4.1 Lecture, Topic 4.1 Demonstration | worksheet34 | lec34lec33-slides |
| WS34-solution | |||||||||||||||||||
13 | Mon | Apr 1910 | Lecture 34: Task Affinity with Places35: Eureka-style Speculative Task Parallelism |
| worksheet35 | lec34lec35-slides |
| Quiz for Unit 8 | Quiz for Unit 7 |
| WS35-solution | |||||||||||||||||||
Wed | Apr | 2112 | Lecture | 35: Eureka-style Speculative Task Parallelism36: Scan Pattern. Parallel Prefix Sum |
| worksheet36 | lec35lec36-slides | WS36-solution | ||||||||||||||||||||||
Fri | Apr | 2414 | Lecture | 3637: | Algorithms based onParallel Prefix | (Scan) operationsSum applications | worksheet37 | lec36lec37-slides | ||||||||||||||||||||||
14 | Mon | Apr | 26TBD17 | Lecture 38: Overview of other models and frameworks | lec38-slides | |||||||||||||||||||||||||
Wed | Apr 2819 | Lecture 3839: Course Review (Lectures 19-3438) | lec38lec39-slides | Homework 4 (all) | ||||||||||||||||||||||||||
Fri | Apr 30TBD21 | Lecture 40: Course Review (Lectures 19-38) | lec40-slides | Quiz for Unit 8 | Homework 5 |
Lab Schedule
Lab # | Date (20212023) | Topic | Handouts | Examples | 0|||||||
---|---|---|---|---|---|---|---|---|---|---|---|
1 | Jan 09 | Infrastructure | Setupsetup | lab0-handout | 1 | Jan 26 | Async-Finish Parallel Programming with abstract metrics | lab1-handout | |||
- | Feb 02Jan 16 | No lab this week (MLK) | |||||||||
2 | Feb 09 | FuturesJan 23 | Functional Programming | lab2-handout | |||||||
3 | Feb 16 | Jan 30 | Java Streams Cutoff Strategy and Real World Performance | lab3-handout | |||||||
4 | Feb | 2306 | DDFsFutures | lab4- | handouthandout | - | Mar 02 | No lab this week | |||
5 | Mar 09Feb 13Loop | Data- Driven Tasks | lab5-handout handoutlab5-intro | ||||||||
- | Mar 16Feb 20 | No lab this week ( | Spring "Sprinkle" DayMidterm) | ||||||||
6 | Feb 27 | Async / Finish | lab6-handout | ||||||||
7 | Mar 06 | Recursive Task Cutoff Strategy | lab7-handout | Isolated Statement and Atomic Variables | |||||||
- | Mar 13 | No lab this week (Spring Break)Actors | |||||||||
8 | Mar 20 | Java Threads, Java Locks | - | lab8-handout | Message Passing Interface (MPI) | ||||||
9 | Mar 27 | Concurrent Lists | lab9-handout | -||||||||
10 | Apr 03 | Actors | lab10-handout | ||||||||
11 | Apr 10 | Loop Parallelism | lab11-handout | Eureka-style Speculative Task Parallelism | |||||||
- | Java's ForkJoin Framework | Apr 17 | No lab this week |
Grading, Honor Code Policy, Processes and Procedures
...
Labs must be submitted by the following Monday Wednesday at 114:59pm30pm. 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.
...