|
|
|
|
Course on Advanced GPU
computing
Dr.
Manuel Carcenac
([email protected])
The following course lecture notes are freely available to
anyone interested.
They may be used and modified without any restrictions.
Chapters: Compute Unified
Device Architecture (CUDA) - Graphic Processing Unit (GPU) - coprocessor -
General Purpose Graphic Processing Unit (GPGPU) - compute capability
- Tesla C2050 - NVIDIA CUDA architecture and programming model [ pdf ; doc ] shared memory system - Single Instruction Multiple Threads (SIMT) - data-based parallelism - grid - block - thread - multiprocessor - core - host - device
CUDA programming interface - CUDA C [ pdf ; doc ] host memory - device memory - host code - device code - kernel - thread synchronization - nvcc
Optimization of a CUDA code [ pdf ; doc ] cache memory - shared memory
CUBLAS linear algebra library [ pdf ; doc ] CUda Basic Linear Algebra Subprograms (CUBLAS) - helper function - core function - leading dimension - column-major storage - matrix multiplication
Application: linear
system resolution with Gauss method [ pdf ; doc ] CUDA - CUBLAS
Using OpenGL with
CUDA [ pdf ;
doc
] installation - basic use - interoperability
Application: all
pairs n-body problem [ pdf
; doc ] time integration - law of gravity - collision of galaxies Some
programs: matrix multiplication with
either CUDA or CUBLAS Gauss method with either CUDA or
CUBLAS other available courses
(Java, Graphics in Java, Advanced 3D Graphics in Java,
|
|
|