SDU Education Information System
   Home   |  Login Türkçe  | English   
 
   
 
 


 
Course Information
Course Unit Title : Introduction to Parallel Computing
Course Unit Code : 01FZK5171
Type of Course Unit : Optional
Level of Course Unit : Second Cycle
Year of Study : Preb
Semester : 255.Semester
Number of ECTS Credits Allocated : 6,00
Name of Lecturer(s) : ---
Course Assistants : ---
Learning Outcomes of The Course Unit : - Improving C and Fortran coding skills,
- Learning MPI and PVM parallel coding techniques
- Application of coding on physics problems
- Finding a solution for challenging physics problems requring high performance,
- Improving physics and engeneering students? programming skills.
Mode of Delivery : Face-To-Face
Prerequisities and Co-requisities Courses : Unavailable
Recommended Optional Programme Components : Unavailable
Course Contents : Computer Architecture, High Performance Computing ? Parallelism, Fast processors architectures, MIMD - memory organization, Standard MPI, Standard PVM, Elementary MPI: What is MPI?; Binding to C and Fortran; MPI Initialization; Who Am I?; Finalizing MPI; Building an MPI code; Running an MPI code, Message Passing, Point to Point Communication: Communication Modes; Communication Envelope; Deadlock; Timing; The Cost of Communication; Simple examples.
Languages of Instruction : Turkish-English
Course Goals : - Introduce the parallel systems
- Understand what the parallel programming is
- Doing basic parallel computing with MPI and PVM.
- Converting basic serial C and Fortran codes into parallel ones.
Course Aims : - Digest to students what the parallel computing is
- How to get high performance with parallel computing
- Digest the message passing enviroment
- Digest distributed memory paradigm
WorkPlacement   doing homework
Recommended or Required Reading
Textbook : Lecture Notes
Additional Resources : 1. MPI in Wikipedia
en.wikipedia.org/wiki/Message_Passing_Interface
2. Tutorial on MPI
www.llnl.gov/computing/tutorials/mpi
3. Using MPI - 2nd Edition by
William Gropp, Ewing Lusk, and Anthony Skjellum
4. Parallel Programming with MPI
by Peter Pacheco
Material Sharing
Documents :
Assignments : There are 7-8 homework during the term.
Exams : 1 midterm exam and 1 final exam
Additional Material :
Planned Learning Activities and Teaching Methods
Lectures, Practical Courses, Presentation, Seminar, Project, Laboratory Applications (if necessary)
ECTS / Table Of Workload (Number of ECTS credits allocated)
Student workload surveys utilized to determine ECTS credits.
Activity :
Number Duration Total  
Course Duration (Excluding Exam Week) :
14 3 42  
Time Of Studying Out Of Class :
14 3 42  
Homeworks :
7 2 14  
Presentation :
0 0 0  
Project :
0 0 0  
Lab Study :
0 0 0  
Field Study :
0 0 0  
Visas :
1 10 10  
Finals :
1 10 10  
Workload Hour (30) :
30  
Total Work Charge / Hour :
0  
Course's ECTS Credit :
0      
Assessment Methods and Criteria
Studies During Halfterm :
Number Co-Effient
Visa :
1 40
Quiz :
0 0
Homework :
7 60
Attendance :
0 0
Application :
0 0
Lab :
0 0
Project :
0 0
Workshop :
0 0
Seminary :
0 0
Field study :
0 0
   
TOTAL :
100
The ratio of the term to success :
60
The ratio of final to success :
40
TOTAL :
100
Weekly Detailed Course Content
Week Topics  
1 Introduction to Parallel Computing: Computer Architecture
 
2 High Performance Computing - Parallelism
 
3 Fast processors architectures
 
4 MIMD - memory organization
 
5 Standard MPI
 
6 Standard PVM
 
7 Elementary MPI: What is MPI?; Binding to C and Fortran;
 
8 Elementary MPI: MPI Initialization; Who Am I?; Finalizing MPI
 
9 Elementary MPI: Building an MPI code; Running an MPI code
 
10 Message Passing
 
11 Point to Point Communication: Communication Modes; Communication Envelope
 
12 Point to Point Communication: Deadlock; Timing; The Cost of Communication
 
13 Simple examples
 
14 Some examples
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0
 
0