Computer Science 400
Parallel Processing and High Performance Computing

Fall 2017, Siena College

Syllabus

Instructor:

Dr. James D. Teresco, Roger Bacon 321, (518) 782-6992
Electronic mail: jteresco AT siena.edu (best contact method)
Twitter: @JTerescoSienaCS
Office hours: Monday, Tuesday, Wednesday 2-3:30, and by appointment
Class URL: [Link]
Class hour: Tuesday, Thursday 9:50-11:15, Roger Bacon 304

Disclaimer

Everything on this syllabus is subject to change. Changes will be announced in class and updated in the online version of the syllabus.

School of Science Mission Statement

The School of Science mission statement can be found on the School of Science web site.

Course Objectives

Performance increases in modern computers no longer come from faster clock speeds. The multi-core revolution has put parallel computing on everyone's desktop, meaning parallel programming techniques are needed to take full advantage of today's computers. Parallelism complicates everything from computer hardware design to operating systems and compilers to application programming, yet it is now pervasive. This course examines methods, techniques, and languages for parallel programming, libraries and extensions to sequential languages to support parallel programming, parallel computing environments, performance analysis, efficiency and complexity of parallel algorithms, and applications of parallel computation.

This course counts as an upper-level elective for the Computer Science Major, the Computer Science 3/2 Program, the Information Systems Certificate, and the Computer Science Minor, and the Information Systems Minor.

Our specific learning objectives for CSIS 400 include the following:

Prerequisites

Texts

The required text for the course is An Introduction to Parallel Programming, First Edition (Morgan Kaufmann, 2011, ISBN 9780123742605) by Peter Pacheco. This is available from the Siena Bookstore (and elsewhere). If you buy elsewhere, be sure to get the correct edition.

Other readings may also be assigned from freely available sources.

Lecture/Lab Meetings

We have the advantage of holding all of our class meetings in a computer lab. This allows a significant fraction of our class time to be used for lab activities. To minimize the need for traditional lectures, it is important to do any readings or small assignments before coming to class. This will permit the majority of class time to be used for hands-on work. Come to class prepared and on time. Assigned readings are listed on the lecture and reading schedule. Everyone is expected to attend class and participate in discussions and to complete in-class exercises.

Assignments

Assignments will be made regularly, and will include several types of items. Different requirements apply to each, as described below. It is important that you adhere to file format and naming requirements to facilitate grading. Submissions that do not meet these requirements will not be accepted.

Lab Questions
The answers to all "lab questions" for a particular lab should be included within a single plain-text or PDF file named labn.txt or labn.pdf, as appropriate, where n is the lab number. The file should start with your name and the lab number. If you use plain text, your file should be appropriately line-wrapped for easy reading in a window or on a printed page with a width of 80 characters.

Note: for lab questions that ask you to draw a memory diagram, you have a few options. You may attempt to represent the memory with plain text, but that is difficult to construct and read. Ideally, you would use a drawing program and submit in your PDF file, but you are also permitted to draw the diagram on paper and submit a scan or photograph, as long as the diagram is legible in that form.

Practice Programs
Your submissions for practice programs are graded primarily on correctness, but you will be required to include your name at the top of the program, and you must use the file name specified. You are of course encouraged to practice good documentation, formatting, and style for these programs, but the grade will depend only on correctness (and having your name in a properly-named file).
Programming Assignments
These are the most formal submissions and will be graded on design, documentation, style, correctness, and efficiency (where appropriate). A good design will use an appropriate algorithm, data structures, and language constructs to solve the problem. A well-documented program will include a comment at the top of each file that includes your name, the assignment, and a description of the contents of the file. There should also be comments for each constructor or method definition (including a brief description of the method's purpose, its parameters, and return value), each variable or group of related variables, and any section of code whose purpose and/or behavior is not obvious from context or the code itself. Style requirements include appropriate formatting (sufficient and consistent indentation, spacing, and punctuation, wrapping long lines of code), good use of named constants, and meaningful and appropriate names for variables, methods, constants, and parameters. Correctness, of course, requires the expected output be produced for a set of test inputs (which will normally not be provided in advance). Efficiency will be more important in some assignments than others, and requires that the program does not do any unnecessary computation or use any more memory than needed. And, of course, you must use the file name(s) specified.

The number of points available will vary with the size and complexity of the assignment.

You may develop your programs on any platform that has the necessary hardware and software, but grading will be done on the platform specified in each assignment. It is your responsibility to ensure that your program works on the grading platform. Programming projects will be graded on design, documentation, style, correctness, and efficiency. All assignments and projects are to be submitted electronically unless otherwise specified. Please submit plain text where appropriate and PDF otherwise. Please avoid other formats such as Word documents. If in doubt about a file format, please check with me first.

In-class exercises must be submitted in class to earn credit. Most assignments, unless otherwise specified, may be turned in late with a penalty computed as 1.08h%, where h is the number of hours late. Extensions will only be granted in serious situations and requests must be accompanied by appropriate written documentation from the appropriate college office. You can find a C program that prints out a table of the late penalties here. Work turned in after solutions have been made available cannot receive credit.

Final Project

For the last part of the semester, the workload of the regular assignments will be reduced and you will focus your efforts on a large project on a relevant topic of your own choosing. The project will include a proposal, a progress report, a paper and a presentation. Most projects will also involve significant software development. Details will be made available around mid-semester, but you should expect a series of intermediate due dates during the last few weeks of the semester and a minisymposium where you will present your projects to your classmates. The final project submission will be due on the last day of classes.

Exams

There will a midterm exam during the semester and a final at the end of the semester. Details will be announced as the exams approach.

Grading

Grades for individual exams and assignments are not scaled. Any scaling deemed appropriate will take place at the end of the semester by adjusting the above thresholds. The following thresholds may be adjusted downward (thereby raising grades) but will never be adjusted upward.

Breakdown:

Scale:
Assignments/Labs 30% A >= 93% A- >= 90%
Programming Projects 25% B+ >= 87% B >= 83% B- >= 80%
Midterm Exam 10% C+ >= 77% C >= 73% C- >= 70%
Final Project 15% D+ >= 67% D >= 63% D- >= 60%
Final Exam 20% F < 60%

Attendance

Please be sure you are familiar with the Siena College Attendance Policy.

Every college student should be motivated to attend every lecture and lab meeting for all the right reasons (e.g., desire for knowledge, desire to get the most out of every very expensive minute, etc.). As experienced college students, you understand that regular attendance is essential to your ability to master the course material.

Therefore, there is no formal attendance policy. You are expected to attend regularly, and should still see the instructor about any excused absences. An excused absence may be any of the following:

  1. A documented athletic or academic event that conflicts with a class meeting. The required paperwork must be presented in person at least one week prior to the event.
  2. A family emergency. These must be documented through the Office of Academic Affairs (783-2307), who will then contact your instructors.
  3. Personal illness. These must be documented by the Office of Student Affairs (783-2328), who will then contact your instructors.

While there is no formal penalty for unexecused absences, missing class regularly, frequent tardiness, or being distracted in class (e.g., checking your phone or Facebook) will be considered a sign that you are not taking the course seriously. Common sense suggests and experience validates that students who are frequently absent, late, or inattentive perform poorly on graded work. Do not expect compassion when final grades are assigned or extensive extra help if you do not understand a topic that was covered while you were absent without a valid excuse.

Disability Accommodations

In compliance with the Americans with Disabilities Act and with Section 504 of the Rehabilitation Act, Siena College is committed to ensuring educational access and accommodations for all its registered students.

Any student with a documented disability needing academic adjustments or accommodations should provide documentation of such during the first two weeks of class. All discussions will remain confidential. Accommodations must be arranged with Mr. Rob Bahny, Director of Services for Students with Disabilities (Foy 109, 783-4239).

Complaints about services provided or not provided may be brought to the attention of Public Safety at 783-2376 or Ms. Lois Goland, JD, Title IX Coordinator and Equal Opportunity Specialist (SSU 235, 782-6673).

Academic Integrity

You are encouraged to discuss the concepts related to course assignments and exams with your classmates. This is an essential part of a healthy academic environment. However, work submitted for grading must be your own (or the combined work of group members, for group assignments). Any unauthorized copying or collaboration is considered a breach of academic integrity and will not be tolerated. Academic dishonesty cases are unpleasant and uncomfortable for everyone involved. You are responsible for reading and understanding the document Academic Integrity and the Siena Student and the Computer Science Department's Academic Integrity statement. The minimum penalties for a first violation will include failure (0 grade) for the assignment or exam in question and the filing of a Academic Integrity Violation Accusation Form. A second violation will result in failure of the course and a formal letter describing your misconduct will be sent to the head of the Computer Science Department and the Office of Academic Affairs. Students suspected of violating academic integrity will be referred to the Academic Integrity Committee for final determination.

If there is any doubt about the degree of collaboration allowed or the permitted sources for a particular assignment, please ask for clarification before collaborating or consulting the source. Any such collaborations or sources must be cited properly.