Syllabus - CMSC714 (Fall 2018)

Class meets TTh 2:00--3:15PM in CSIC 2120.

Selected topics in high-performance systems, including contemporary architectures, interconnection topologies, shared memory and message-passing systems, multi-threaded kernels, latency avoidance and hiding techniques, methods for data and workload partitioning, performance profiling, debugging.

The objective of this course is to study the theory and practice of high performance parallel and distributed computing. The overall goal is to expose you to tools and techniques for parallel and/or distributed environments and to study current trends and technologies. 

This course will focus on current practices in high performance computing technologies, including systems, architectures, programming models, languages and software tools. The course will require students to produce questions for multiple papers during the semester and complete two introductory and one more extensive group research project in parallel and/or distributed programming.

CMSC411 and CMSC412; or permission of instructor.

None - Reading List will provide materials

• Introduction (1 week)
   
• Programming Models (3 weeks)
   
• Parallel Architectures and Networks (2 weeks)
   
• Grid and Cloud Computing(2 weeks)
   
• Debugging & Instrumentation (1 week)
   
• Performance Tools (2 weeks)
   
• Runtime systems and Parallel I/O (1 week)
   
• Commercial and Scientific Applications (1 week)
   

Your grade will be determined as follows:

• The first two projects, in C and/or Fortran, will use MPI and OpenMP to become familiar with parallel/distributed programming on a Linux cluster. Accounts on the university deepthought2 cluster will be forthcoming.
  
  
• After the first 2 projects, you will do a group project, in C/C++ (or Fortran or Java), with milestones along the way.
  This project will be done in teams of 3 or 4 people.
  
  
• The projects should be submitted by 6:00PM on the day they are due.
  Late projects will be not be accepted.
  
  
• The instructor reserves the right to fail, regardless of overall numeric score, students who do not submit on-time a good faith attempt to complete all programming assignments and projects.
  
  
• All submissions of assignments will be via email to the instructor. Posted deadlines for program submission are sharp; standard Unix time of submission is used.
  
  

• Do not post your assignments' implementation online (e.g., GitHub, PasteBin) where they can be seen by others. Making your code accessible to others can lead to academic integrity violations.
• Posting of your projects in a private repository where only selected people (e.g., potential employers) can see them is OK. Just make sure is not a public site.
• Even once the course is over do not make your code publicly available to others.

See the section titled "Attendance, Absences, or Missed Assignments" available at Course Related Policies.

See the section titled "Accessibility" available at Course Related Policies.

The college policy on academic dishonesty is strictly followed.
All graded materials (whether exams or programming assignments) must be strictly individual efforts, except for group projects.

Dr. Sussman is the information owner for all Linux cluster accounts in this class, and all files in those accounts may be inspected by him at any time.

Note that academic dishonesty includes not only cheating, fabrication, and plagiarism, but also includes helping other students commit acts of academic dishonesty by allowing them to obtain copies of your work. In short, all submitted work must be your own. Cases of academic dishonesty will be pursued to the fullest extent possible as stipulated by the Office of Student Conduct.

It is very important for you to be aware of the consequences of cheating, fabrication, facilitation, and plagiarism. For more information on the Code of Academic Integrity or the Student Honor Council, please visit http://www.shc.umd.edu.

The CS Department takes academic integrity seriously. Information on how the CS Department views and handle academic integrity matters can be found at Academic Integrity.

The following are examples of academic integrity violations:

• Hardcoding of results in a project assignment. Hardcoding refers to attempting to make a program appear as if it works correctly (e.g., printing expected results for a test).
• Using any code available on the internet/web or any other source. For example, using code from GitHub.
• Hiring any online service to complete an assignment for you.
• Sharing your code or your tests with any student.
• Using online forums in order to ask for help regarding assignments.

• At the end of the semester visit (www.courseevalum.umd.edu) to complete your course evaluations.
• If you are experiencing difficulties in keeping up with the academic demands of this course, you may contact the Counseling Center located in the Shoemaker Building.
• UMD Course related policies can be found at http://www.ugst.umd.edu/courserelatedpolicies.html

All course materials are copyright UMCP, Department of Computer Science © 2018. All rights reserved. Students are permitted to use course materials for their own personal use only. Course materials may not be distributed publicly or provided to others (excepting other students in the course), in any way or format.

  

Web Accessibility