http://en.wikipedia.org/wiki/CUDA

CUDA

From Wikipedia, the free encyclopedia
 
 
Developer(s)NVIDIA Corporation
Stable release6.0 / November 14, 2013; 4 days ago
Operating systemWindows XP and later,
Mac OS XLinux
PlatformSupported GPUs
TypeGPGPU
LicenseFreeware
Websitewww.nvidia.com/object/cuda_home_new.html

CUDA (aka Compute Unified Device Architecture) is a parallel computing platform and programming model created by NVIDIA and implemented by the graphics processing units (GPUs) that they produce.CUDA gives program developers direct access to the virtual instruction set and memory of the parallel computational elements in CUDA GPUs.

Using CUDA, the GPUs can be used for general purpose processing (i.e., not exclusively graphics); this approach is known as GPGPU. Unlike CPUs, however, GPUs have a parallel throughput architecture that emphasizes executing many concurrent threads slowly, rather than executing a single thread very quickly.

The CUDA platform is accessible to software developers through CUDA-accelerated libraries, compiler directives (such as OpenACC), and extensions to industry-standard programming languages, including C,C++ and Fortran. C/C++ programmers use 'CUDA C/C++', compiled with "nvcc", NVIDIA's LLVM-based C/C++ compiler, and Fortran programmers can use 'CUDA Fortran', compiled with the PGI CUDA Fortran compiler from The Portland Group.

In addition to libraries, compiler directives, CUDA C/C++ and CUDA Fortran, the CUDA platform supports other computational interfaces, including the Khronos Group's OpenCL, Microsoft's DirectCompute, and C++ AMP. Third party wrappers are also available for PythonPerlFortranJavaRubyLuaHaskell,MATLABIDL, and native support in Mathematica.

In the computer game industry, GPUs are used not only for graphics rendering but also in game physics calculations (physical effects like debris, smoke, fire, fluids); examples include PhysX and Bullet. CUDA has also been used to accelerate non-graphical applications in computational biologycryptography and other fields by an order of magnitude or more.

CUDA provides both a low level API and a higher level API. The initial CUDA SDK was made public on 15 February 2007, for Microsoft Windows and LinuxMac OS Xsupport was later added in version 2.0, which supersedes the beta released February 14, 2008. CUDA works with all Nvidia GPUs from the G8x series onwards, including GeForceQuadro and the Tesla line. CUDA is compatible with most standard operating systems. Nvidia states that programs developed for the G8x series will also work without modification on all future Nvidia video cards, due to binary compatibility.

NVIDIA---CUDA-LMLPHP

NVIDIA---CUDA-LMLPHP

Example of CUDA processing flow
1. Copy data from main mem to GPU mem
2. CPU instructs the process to GPU
3. GPU execute parallel in each core
4. Copy the result from GPU mem to main mem

Background[edit]

See also: GPU

The GPU, as a specialized processor, addresses the demands of real-time high-resolution 3D graphics compute-intensive tasks. As of 2012, GPUs have evolved into highly parallel multi-core systems allowing very efficient manipulation of large blocks of data. This design is more effective than general-purpose CPUs for algorithmswhere processing of large blocks of data is done in parallel, such as:

Advantages[edit]

CUDA has several advantages over traditional general-purpose computation on GPUs (GPGPU) using graphics APIs:

  • Scattered reads – code can read from arbitrary addresses in memory
  • Shared memory – CUDA exposes a fast shared memory region (up to 48KB per Multi-Processor) that can be shared amongst threads. This can be used as a user-managed cache, enabling higher bandwidth than is possible using texture lookups.
  • Faster downloads and readbacks to and from the GPU
  • Full support for integer and bitwise operations, including integer texture lookups

Limitations[edit]

  • CUDA does not support the full C standard, as it runs host code through a C++ compiler, which makes some valid C (but invalid C++) code fail to compile.
  • Texture rendering is not supported (CUDA 3.2 and up addresses this by introducing "surface writes" to CUDA arrays, the underlying opaque data structure).
  • Copying between host and device memory may incur a performance hit due to system bus bandwidth and latency (this can be partly alleviated with asynchronous memory transfers, handled by the GPU's DMA engine)
  • Threads should be running in groups of at least 32 for best performance, with total number of threads numbering in the thousands. Branches in the program code do not affect performance significantly, provided that each of 32 threads takes the same execution path; the SIMD execution model becomes a significant limitation for any inherently divergent task (e.g. traversing a space partitioning data structure during ray tracing).
  • Unlike OpenCL, CUDA-enabled GPUs are only available from Nvidia
  • Valid C/C++ may sometimes be flagged and prevent compilation due to optimization techniques the compiler is required to employ to use limited resources.
  • CUDA (with compute capability 1.x) uses a recursion-free, function-pointer-free subset of the C language, plus some simple extensions. However, a single process must run spread across multiple disjoint memory spaces, unlike other C language runtime environments.
  • CUDA (with compute capability 2.x) allows a subset of C++ class functionality, for example member functions may not be virtual (this restriction will be removed in some future release). [See CUDA C Programming Guide 3.1 – Appendix D.6]
  • Double precision floats (CUDA compute capability 1.3 and above) deviate from the IEEE 754 standard: round-to-nearest-even is the only supported rounding mode for reciprocal, division, and square root. In single precisiondenormals and signalling NaNs are not supported; only two IEEE rounding modes are supported (chop and round-to-nearest even), and those are specified on a per-instruction basis rather than in a control word; and the precision of division/square root is slightly lower than single precision.

Supported GPUs[edit]

Compute capability table (version of CUDA supported) by GPU and card. Also available directly from Nvidia:

Compute
capability
(version)
GPUsCards
1.0G80, G92, G92b, G94, G94bGeForce 8800GTX/Ultra, 9400GT, 9600GT, 9800GT, Tesla C/D/S870, FX4/5600, 360M, GT 420
1.1G86, G84, G98, G96, G96b, G94, G94b, G92, G92bGeForce 8400GS/GT, 8600GT/GTS, 8800GT/GTS, 9600 GSO, 9800GTX/GX2, GTS 250, GT 120/30/40, FX 4/570, 3/580, 17/18/3700, 4700x2, 1xxM, 32/370M, 3/5/770M, 16/17/27/28/36/37/3800M, NVS290, NVS420/50
1.2GT218, GT216, GT215GeForce 210, GT 220/240, FX380 LP, 1800M, 370/380M, NVS300, NVS 2/3100M
1.3GT200, GT200bGeForce GTX 260, GTX 275, GTX 280, GTX 285, GTX 295, Tesla C/M1060, S1070, Quadro CX, FX 3/4/5800
2.0GF100, GF110GeForce (GF100) GTX 465, GTX 470, GTX 480, Tesla C2050, C2070, S/M2050/70, Quadro Plex 7000, Quadro 4000, 5000, 6000, GeForce (GF110) GTX 560 TI 448, GTX570, GTX580, GTX590
2.1GF104, GF106 GF108,GF114, GF116, GF119GeForce 500M series, 610M, GT630M,GTX 670M, GeForce GTX 675M, GT 430, GT 440, GTS 450, GTX 460, GT 545, GTX 550 Ti, GTX 560, GTX 560 Ti, 605,615,620, Quadro 600, 2000
3.0GK104, GK106, GK107GeForce GTX 770, GTX 760, GTX 690, GTX 680, GTX 670, GTX 660 Ti, GTX 660, GTX 650 Ti BOOST, GTX 650 Ti, GTX 650, GT 640, GT 630, GeForce GTX 780M, GeForce GTX 775M(for Apple OEM only), GeForce GTX 770M, GeForce GTX 765M, GeForce GTX 760M, GeForce GT 755M(for Apple OEM only), GeForce GT 750M, GeForce GT 745M, GeForce GT 740M, GeForce GTX 680MX(for Apple OEM only), GeForce GTX 680M, GeForce GTX 675MX, GeForce GTX 670MX, GTX 660M, GeForce GT 650M, GeForce GT 645M, GeForce GT 640M, Quadro K600, Quadro K2000, Quadro K4000, Quadro K5000, Quadro K2100M,Quadro K4100M,Quadro K5100M
3.5GK110, GK208Tesla K40, K20X, K20, GeForce GTX TITAN, GTX780Ti, GTX 780, Quadro K510M, Quadro K610M, Quadro K6000, GT 640(Rev.2)

A table of devices officially supporting CUDA:

Nvidia GeForce
GeForce GTX TITAN
GeForce GTX 780 Ti
GeForce GTX 780
GeForce GTX 770
GeForce GTX 760
GeForce GTX 690
GeForce GTX 680
GeForce GTX 670
GeForce GTX 660 Ti
GeForce GTX 660
GeForce GTX 650 Ti BOOST
GeForce GTX 650 Ti
GeForce GTX 650
GeForce GT 640
GeForce GTX 590
GeForce GTX 580
GeForce GTX 570
GeForce GTX 560 Ti
GeForce GTX 560
GeForce GTX 550 Ti
GeForce GT 520
GeForce GTX 480
GeForce GTX 470
GeForce GTX 465
GeForce GTX 460
GeForce GTX 460 SE
GeForce GTS 450
GeForce GT 440
GeForce GT 430
GeForce GT 420
GeForce GTX 295
GeForce GTX 285
GeForce GTX 280
GeForce GTX 275
GeForce GTX 260
GeForce GTS 250
GeForce GTS 240
GeForce GT 240
GeForce GT 220
GeForce 210/G210
GeForce GT 140
GeForce 9800 GX2
GeForce 9800 GTX+
GeForce 9800 GTX
GeForce 9800 GT
GeForce 9600 GSO
GeForce 9600 GT
GeForce 9500 GT
GeForce 9400 GT
GeForce 9400 mGPU
GeForce 9300 mGPU
GeForce 9100 mGPU
GeForce 8800 Ultra
GeForce 8800 GTX
GeForce 8800 GTS
GeForce 8800 GT
GeForce 8800 GS
GeForce 8600 GTS
GeForce 8600 GT
GeForce 8600 mGT
GeForce 8500 GT
GeForce 8400 GS
GeForce 8300 mGPU
GeForce 8200 mGPU
GeForce 8100 mGPU

GeForce GT 630

Nvidia GeForce Mobile
GeForce GTX 780M
GeForce GTX 770M
GeForce GTX 765M
GeForce GTX 760M
GeForce GT 750M
GeForce GT 745M
GeForce GT 740M
GeForce GT 735M
GeForce GT 730M
GeForce GTX 680MX
GeForce GTX 680M
GeForce GTX 675MX
GeForce GTX 675M
GeForce GTX 670MX
GeForce GTX 670M
GeForce GTX 660M
GeForce GT 650M
GeForce GT 645M
GeForce GT 640M
GeForce GTX 580M
GeForce GTX 570M
GeForce GTX 560M
GeForce GT 555M
GeForce GT 550M
GeForce GT 540M
GeForce GT 525M
GeForce GT 520M
GeForce GTX 480M
GeForce GTX 470M
GeForce GTX 460M
GeForce GT 445M
GeForce GT 435M
GeForce GT 425M
GeForce GT 420M
GeForce GT 415M
GeForce GTX 285M
GeForce GTX 280M
GeForce GTX 260M
GeForce GTS 360M
GeForce GTS 350M
GeForce GTS 260M
GeForce GTS 250M
GeForce GT 335M
GeForce GT 330M
GeForce GT 325M
GeForce GT 320M
GeForce 310M
GeForce GT 240M
GeForce GT 230M
GeForce GT 220M
GeForce G210M
GeForce GTS 160M
GeForce GTS 150M
GeForce GT 130M
GeForce GT 120M
GeForce G110M
GeForce G105M
GeForce G103M
GeForce G102M
GeForce G100
GeForce 9800M GTX
GeForce 9800M GTS
GeForce 9800M GT
GeForce 9800M GS
GeForce 9700M GTS
GeForce 9700M GT
GeForce 9650M GT
GeForce 9650M GS
GeForce 9600M GT
GeForce 9600M GS
GeForce 9500M GS
GeForce 9500M G
GeForce 9400M G
GeForce 9300M GS
GeForce 9300M G
GeForce 9200M GS
GeForce 9100M G
GeForce 8800M GTX
GeForce 8800M GTS
GeForce 8700M GT
GeForce 8600M GT
GeForce 8600M GS
GeForce 8400M GT
GeForce 8400M GS
GeForce 8400M G
GeForce 8200M G
Nvidia Quadro
Quadro K6000
Quadro K5000
Quadro K4000
Quadro K2000D
Quadro K2000
Quadro K600
Quadro 6000
Quadro 5000
Quadro 4000
Quadro 2000
Quadro 600
Quadro FX 5800
Quadro FX 5600
Quadro FX 4800
Quadro FX 4700 X2
Quadro FX 4600
Quadro FX 3800
Quadro FX 3700
Quadro FX 1800
Quadro FX 1700
Quadro FX 580
Quadro FX 570
Quadro FX 380
Quadro FX 370
Quadro NVS 510
Quadro NVS 450
Quadro NVS 420
Quadro NVS 295
Quadro Plex 1000 Model IV
Quadro Plex 1000 Model S4
Nvidia Quadro Mobile
Quadro K5100M
Quadro K5000M
Quadro K4100M
Quadro K4000M
Quadro K3100M
Quadro K3000M
Quadro K2100M
Quadro K2000M
Quadro K1100M
Quadro K1000M
Quadro K610M
Quadro K510M
Quadro K500M
Quadro 5010M
Quadro 5000M
Quadro 4000M
Quadro 3000M
Quadro 2000M
Quadro 1000M
Quadro FX 3800M
Quadro FX 3700M
Quadro FX 3600M
Quadro FX 2800M
Quadro FX 2700M
Quadro FX 1800M
Quadro FX 1700M
Quadro FX 1600M
Quadro FX 880M
Quadro FX 770M
Quadro FX 570M
Quadro FX 380M
Quadro FX 370M
Quadro FX 360M
Quadro NVS 320M
Quadro NVS 160M
Quadro NVS 150M
Quadro NVS 140M
Quadro NVS 135M
Quadro NVS 130M
Nvidia Tesla
Tesla K40
Tesla K20X
Tesla K20
Tesla K10
Tesla C2050/2070
Tesla M2050/M2070
Tesla S2050
Tesla S1070
Tesla M1060
Tesla C1060
Tesla C870
Tesla D870
Tesla S870

Version features and specifications[edit]

Feature support (unlisted features are
supported for all compute capabilities)
Compute capability (version)
1.01.11.21.32.x3.03.5
Integer atomic functions operating on
32-bit words in global memory
NoYes
atomicExch() operating on 32-bit
floating point values in global memory
Integer atomic functions operating on
32-bit words in shared memory
NoYes
atomicExch() operating on 32-bit
floating point values in shared memory
Integer atomic functions operating on
64-bit words in global memory
Warp vote functions
Double-precision floating-point operationsNoYes
Atomic functions operating on 64-bit
integer values in shared memory
NoYes
Floating-point atomic addition operating on
32-bit words in global and shared memory
_ballot()
_threadfence_system()
_syncthreads_count(),
_syncthreads_and(),
_syncthreads_or()
Surface functions
3D grid of thread block
Warp shuffle functionsNoYes
Funnel shiftNoYes
Dynamic parallelism
Technical specificationsCompute capability (version)
1.01.11.21.32.x3.03.5
Maximum dimensionality of grid of thread blocks23
Maximum x-, y-, or z-dimension of a grid of thread blocks655352-1
Maximum dimensionality of thread block3
Maximum x- or y-dimension of a block5121024
Maximum z-dimension of a block64
Maximum number of threads per block5121024
Warp size32
Maximum number of resident blocks per multiprocessor816
Maximum number of resident warps per multiprocessor24324864
Maximum number of resident threads per multiprocessor768102415362048
Number of 32-bit registers per multiprocessor8 K16 K32 K64 K
Maximum number of 32-bit registers per thread12863255
Maximum amount of shared memory per multiprocessor16 KB48 KB
Number of shared memory banks1632
Amount of local memory per thread16 KB512 KB
Constant memory size64 KB
Cache working set per multiprocessor for constant memory8 KB
Cache working set per multiprocessor for texture memoryDevice dependent, between 6 KB and 8 KB12 KBBetween 12 KB
and 48 KB
Maximum width for 1D texture
reference bound to a CUDA array
819265536
Maximum width for 1D texture
reference bound to linear memory
2
Maximum width and number of layers
for a 1D layered texture reference
8192 × 51216384 × 2048
Maximum width and height for 2D
texture reference bound to a CUDA array
65536 × 3276865536 × 65535
Maximum width and height for 2D
texture reference bound to a linear memory
65000 × 6500065000 × 65000
Maximum width and height for 2D
texture reference bound to a CUDA array
supporting texture gather
N/A16384 × 16384
Maximum width, height, and number
of layers for a 2D layered texture reference
8192 × 8192 × 51216384 × 16384 × 2048
Maximum width, height and depth
for a 3D texture reference bound to linear
memory or a CUDA array
2048 × 2048 × 20484096 × 4096 × 4096
Maximum width (and height) for a cubemap
texture reference
N/A16384
Maximum width (and height) and number
of layers for a cubemap layered texture reference
N/A16384 × 2046
Maximum number of textures that
can be bound to a kernel
128256
Maximum width for a 1D surface
reference bound to a CUDA array
Not
supported
65536
Maximum width and number of layers
for a 1D layered surface reference
65536 × 2048
Maximum width and height for a 2D
surface reference bound to a CUDA array
65536 × 32768
Maximum width, height, and number
of layers for a 2D layered surface reference
65536 × 32768 × 2048
Maximum width, height, and depth
for a 3D surface reference bound to a CUDA array
65536 × 32768 × 2048
Maximum width (and height) for a cubemap
surface reference bound to a CUDA array
32768
Maximum width (and height) and number
of layers for a cubemap layered surface reference
32768 × 2046
Maximum number of surfaces that
can be bound to a kernel
816
Maximum number of instructions per
kernel
2 million512 million
Architecture specificationsCompute capability (version)
1.01.11.21.32.02.13.03.5
Number of cores for integer and floating-point arithmetic functions operations83248192192
Number of special function units for single-precision floating-point transcendental functions2483232
Number of texture filtering units for every texture address unit or render output unit (ROP)2483232
Number of warp schedulers12244
Number of instructions issued at once by scheduler11222

For more information please visit this site: http://www.geeks3d.com/20100606/gpu-computing-nvidia-cuda-compute-capability-comparative-table/ and also read Nvidia CUDA programming guide.

Example[edit]

This example code in C++ loads a texture from an image into an array on the GPU:

texture<float, 2, cudaReadModeElementType> tex;

void foo()
{
cudaArray* cu_array; // Allocate array
cudaChannelFormatDesc description = cudaCreateChannelDesc<float>();
cudaMallocArray(&cu_array, &description, width, height); // Copy image data to array
cudaMemcpyToArray(cu_array, image, width*height*sizeof(float), cudaMemcpyHostToDevice); // Set texture parameters (default)
tex.addressMode[0] = cudaAddressModeClamp;
tex.addressMode[1] = cudaAddressModeClamp;
tex.filterMode = cudaFilterModePoint;
tex.normalized = false; // do not normalize coordinates // Bind the array to the texture
cudaBindTextureToArray(tex, cu_array); // Run kernel
dim3 blockDim(16, 16, 1);
dim3 gridDim((width + blockDim.x - 1)/ blockDim.x, (height + blockDim.y - 1) / blockDim.y, 1);
kernel<<< gridDim, blockDim, 0 >>>(d_data, height, width); // Unbind the array from the texture
cudaUnbindTexture(tex);
} //end foo() __global__ void kernel(float* odata, int height, int width)
{
unsigned int x = blockIdx.x*blockDim.x + threadIdx.x;
unsigned int y = blockIdx.y*blockDim.y + threadIdx.y;
if (x < width && y < height) {
float c = tex2D(tex, x, y);
odata[y*width+x] = c;
}
}

Below is an example given in Python that computes the product of two arrays on the GPU. The unofficial Python language bindings can be obtained from PyCUDA.

import pycuda.compiler as comp
import pycuda.driver as drv
import numpy
import pycuda.autoinit mod = comp.SourceModule("""
__global__ void multiply_them(float *dest, float *a, float *b)
{
const int i = threadIdx.x;
dest[i] = a[i] * b[i];
}
""") multiply_them = mod.get_function("multiply_them") a = numpy.random.randn(400).astype(numpy.float32)
b = numpy.random.randn(400).astype(numpy.float32) dest = numpy.zeros_like(a)
multiply_them(
drv.Out(dest), drv.In(a), drv.In(b),
block=(400,1,1)) print dest-a*b

Additional Python bindings to simplify matrix multiplication operations can be found in the program pycublas.

import numpy
from pycublas import CUBLASMatrix
A = CUBLASMatrix( numpy.mat([[1,2,3]],[[4,5,6]],numpy.float32) )
B = CUBLASMatrix( numpy.mat([[2,3]],[4,5],[[6,7]],numpy.float32) )
C = A*B
print C.np_mat()

Language bindings[edit]

Current CUDA architectures[edit]

The current generation CUDA architecture (codename: Fermi) which is standard on Nvidia's released (GeForce 400 Series [GF100] (GPU) 2010-03-27) GPU is designed from the ground up to natively support more programming languages such as C++. It has significantly increased the peak double-precision floating-point performance compared to Nvidia's prior-generation Tesla GPU. It also introduced several new features including:

  • up to 1024 CUDA cores and 6.0 billion transistors on the GTX 590
  • Nvidia Parallel DataCache technology
  • Nvidia GigaThread engine
  • ECC memory support
  • Native support for Visual Studio

Current and future usages of CUDA architecture[edit]

See also[edit]

External links[edit]

 
 
[hide]

 
 
Early chipsets
 
RIVA series
 
GeForce series
 
Other series
 
Technologies
 
Multimedia Acceleration Technologies
 
 
[hide]

 
 
GeForce Series
 
nForce Series
 
Technologies
 
 
[hide]

Other products
 
 
Consoles
 
CPUs
 
Bridge chips
 
Handheld
 
Software
 
Acquisitions
 
 

 
Architecture
 
Instruction set
 
Word size
 
Pipeline
 
Parallel level
 
Multithreading
 
Flynn's taxonomy
 
Types
 
Components
 
Power
management

 
GeneralNVIDIA---CUDA-LMLPHP
 
Levels
 
Multithreading
 
Theory
 
Elements
 
Coordination
 
Programming
 
Hardware
 
APIs
 
Problems
05-11 09:37