Dynamischer Default-Fachbereich geändert auf INF

Module INF-14-54-M-6

High Performance Computing with GPUs (M, 6.0 LP)

Module Identification

Module Number	Module Name	CP (Effort)
INF-14-54-M-6	High Performance Computing with GPUs	6.0 CP (180 h)

Basedata

CP, Effort	6.0 CP = 180 h
Position of the semester	1 Sem. in WiSe
Level	[6] Master (General)
Language	[DE/EN] German or English as required
Module Manager	Gauger, Nicolas, Prof. Dr. (PROF \| DEPT: INF)
Lecturers	Schüle, Josef, Dr. (WMA \| DEPT: INF)
Area of study	[INF-VIS] Visualisation and Scientific Computing
Reference course of study	[INF-88.79-SG] M.Sc. Computer Science
Livecycle-State	[NORM] Active

Notice

Lecuture plus programming exercises.

Courses

Type/SWS	Course Number	Title	Choice in Module-Part	Presence-Time / Self-Study		SL	SL is required for exa.	PL	CP	Sem.
3V+1U	INF-14-54-K-6	High Performance Computing with GPUs	P	56 h	124 h	U-Schein	ja	PL1	6.0	WiSe

About [INF-14-54-K-6]: Title: "High Performance Computing with GPUs"; Presence-Time: 56 h; Self-Study: 124 h
About [INF-14-54-K-6]: The study achievement [U-Schein] proof of successful participation in the exercise classes (ungraded) must be obtained. It is a prerequisite for the examination for PL1.

Examination achievement PL1

Form of examination: oral examination (20-60 Min.)
Examination Frequency: each semester
Examination number: 61454 ("High Performance Computing with GPGPUs")

Evaluation of grades

The grade of the module examination is also the module grade.

From [INF-14-54-K-6] High Performance Computing with GPUs:

Up-to-date graphical devices allow not only video games but may be used for scientific computations. They are heavily used in deep learning and artificial intelligence. With their superior performance it is not surprising that many of the fastest computers in the world contain these cards. This course offers basic knowledges on high performance computing on graphical devices. The course focuses on programming graphic cards with Compute Unified Device Architecture (CUDA) which is trained and deepened in examples and exercises.

Comprehension in SIMD programming model (single instruction multiple data) and relation between hardware architecture and performance
evaluation of parallel algorithms with regard to their performance
theoretical and practical application of CUDA

Competencies / intended learning achievements

After successfully completing the module, students will be able to

analyze simple algorithms with regard to parallelization and transform them into parallel algorithms,
implement qualified and unaided given algorithms in CUDA on graphical processing units (GPUs).

Literature