Navigation To modulepage
Display language

Parallel Systems

6 LP

English

#40626 / #4

SS 2017 - SS 2018

Fakultät IV

EN 6

Institut für Telekommunikationssysteme

34331300 FG Kommunikations- und Betriebssysteme

Heiß, Hans-Ulrich

No information

lehre@kbs.tu-berlin.de

POS-Nummer PORD-Nummer Modultitel
61380 18146 Parallel Systems

Learning Outcomes

Students who have successfully finished this module have a good knowledge of different parallel architectures. They are capable to design and analyze parallel algorithms for different target architectures. They know how to implement them on different parallel architectures with state-of-the-art tools. They also know, how to manage parallel computers for multiprogramming and how to solve the partitioning, mapping, load balancing and scheduling problems.

Content

Parallel machine architectures. Performance models for parallel computing. Allocation problems in parallel systems and (heuristic) algorithms for solving: partitioning, mapping, scheduling and load balancing. Parallel machine models. Analysis of parallel algorithms. Methodology of parallel software engineering. Design of parallel algorithms with examples. Parallel programming environments.

Module Components

Pflichtteil:

All Courses are mandatory.

Course Name Type Number Cycle Language SWS
Cluster Computing VL 0432 L 520 SS No information 2
Parallel Programming IV 0432 L 596 SS No information 4

Workload and Credit Points

Cluster Computing (VL):

Workload description Multiplier Hours Total
Preparation and follow-up 15.0 1.0h 15.0h
Presence 15.0 2.0h 30.0h
45.0h (~2 LP)

Parallel Programming (IV):

Workload description Multiplier Hours Total
Assignments 5.0 9.0h 45.0h
Presence 15.0 4.0h 60.0h
105.0h (~4 LP)

Course-independent workload:

Workload description Multiplier Hours Total
Exam preparation 1.0 30.0h 30.0h
30.0h (~1 LP)
The Workload of the module sums up to 180.0 Hours. Therefore the module contains 6 Credits.

Description of Teaching and Learning Methods

The lecture conveys the material in traditional form. The integrated course is blend of lecture, tutorial and lab hours.

Requirements for participation and examination

Desirable prerequisites for participation in the courses:

Bachelor in Computer Science or related; basic knowledge in computer architecture, operating systems, algorithms and data structures; C programming.

Mandatory requirements for the module test application:

1. Requirement:
[KBS] Parallel Programming Assignments

Module completion

Grading:

graded

Type of exam:

Oral exam

Language:

English

Duration/Extent:

30 minutes

Duration of the Module

This module can be completed in one semester.

Maximum Number of Participants

The maximum capacity of students is 35.

Registration Procedures

See http://www.kbs.tu-berlin.de

Recommended reading, Lecture notes

Lecture notes

Availability:  unavailable

Electronical lecture notes

Availability:  available
Additional information:
http://www.kbs.tu-berlin.de

Literature

Recommended literature
Andrews, G. A.: Foundations of Multithreaded, Parallel and Distributed Programming. Addison-Wesley, 2000.
Buyya, R.: High Performance Cluster Computing, Vol. 1+2. Prentice Hall, 1999.
Heiss, H.-U.: Prozessorzuteilung in Parallelrechnern. BI-Verlag, Mannheim, 1996.
Herlihy, M.; Shavit, N.: The art of multiprocessor programming. Morgan Kaufmann, 2008.
Mattson, T.; Sanders, B.; Massingill, B.: Patterns for Parallel Programming. Addison-Wesley, 2004.
Quinn, M.: Parallel programming in C with MPI and OpenMP. Mgraw-Hill, 2003.
Rauber, Th.; Rünger, G.: Parallele Programmierung, 2. Aufl. Springer, 2007.

Assigned Degree Programs

This module is used in the following modulelists:

Students of other degrees can participate in this module without capacity testing.

Miscellaneous

The module is offered every year.