Misplaced Pages

This article has multiple issues. Please help improve it or discuss these issues on the talk page. (Learn how and when to remove these messages) This article is written like a manual or guide. Please help rewrite this article and remove advice or instruction. (February 2024) This article contains a pro and con list. Please help rewriting it into consolidated sections based on topics. (February 2024) (Learn how and when to remove this message)

CUDA

Developer(s)	Nvidia
Initial release	February 15, 2007; 17 years ago (2007-02-15)
Stable release	12.6 / August 2024; 4 months ago (2024-08)
Operating system	Windows, Linux
Platform	Supported GPUs
Type	GPGPU
License	Proprietary
Website	developer.nvidia.com/cuda-zone

In computing, CUDA is a proprietary parallel computing platform and application programming interface (API) that allows software to use certain types of graphics processing units (GPUs) for accelerated general-purpose processing, an approach called general-purpose computing on GPUs. CUDA was created by Nvidia in 2006. When it was first introduced, the name was an acronym for Compute Unified Device Architecture, but Nvidia later dropped the common use of the acronym and now rarely expands it.

CUDA is a software layer that gives direct access to the GPU's virtual instruction set and parallel computational elements for the execution of compute kernels. In addition to drivers and runtime kernels, the CUDA platform includes compilers, libraries and developer tools to help programmers accelerate their applications.

CUDA is designed to work with programming languages such as C, C++, Fortran and Python. This accessibility makes it easier for specialists in parallel programming to use GPU resources, in contrast to prior APIs like Direct3D and OpenGL, which require advanced skills in graphics programming. CUDA-powered GPUs also support programming frameworks such as OpenMP, OpenACC and OpenCL.

Background

The graphics processing unit (GPU), as a specialized computer processor, addresses the demands of real-time high-resolution 3D graphics compute-intensive tasks. By 2012, GPUs had evolved into highly parallel multi-core systems allowing efficient manipulation of large blocks of data. This design is more effective than general-purpose central processing unit (CPUs) for algorithms in situations where processing large blocks of data is done in parallel, such as:

• cryptographic hash functions
• machine learning
• molecular dynamics simulations
• physics engines

Ian Buck, while at Stanford in 2000, created an 8K gaming rig using 32 GeForce cards, then obtained a DARPA grant to perform general purpose parallel programming on GPUs. He then joined Nvidia, where since 2004 he has been overseeing CUDA development. In pushing for CUDA, Jensen Huang aimed for the Nvidia GPUs to become a general hardware for scientific computing. CUDA was released in 2007. Around 2015, the focus of CUDA changed to neural networks.

Ontology

The following table offers a non-exact description for the ontology of CUDA framework.

Programming abilities

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++, Fortran and Python. C/C++ programmers can use 'CUDA C/C++', compiled to PTX with nvcc, Nvidia's LLVM-based C/C++ compiler, or by clang itself. Fortran programmers can use 'CUDA Fortran', compiled with the PGI CUDA Fortran compiler from The Portland Group. Python programmers can use the cuNumeric library to accelerate applications on Nvidia GPUs.

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, OpenGL Compute Shader and C++ AMP. Third party wrappers are also available for Python, Perl, Fortran, Java, Ruby, Lua, Common Lisp, Haskell, R, MATLAB, IDL, Julia, and native support in Mathematica.

In the computer game industry, GPUs are used for graphics rendering, and for game physics calculations (physical effects such as debris, smoke, fire, fluids); examples include PhysX and Bullet. CUDA has also been used to accelerate non-graphical applications in computational biology, cryptography and other fields by an order of magnitude or more.

CUDA provides both a low level API (CUDA Driver API, non single-source) and a higher level API (CUDA Runtime API, single-source). The initial CUDA SDK was made public on 15 February 2007, for Microsoft Windows and Linux. Mac OS X support 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 GeForce, Quadro and the Tesla line. CUDA is compatible with most standard operating systems.

CUDA 8.0 comes with the following libraries (for compilation & runtime, in alphabetical order):

• cuBLAS – CUDA Basic Linear Algebra Subroutines library
• CUDART – CUDA Runtime library
• cuFFT – CUDA Fast Fourier Transform library
• cuRAND – CUDA Random Number Generation library
• cuSOLVER – CUDA based collection of dense and sparse direct solvers
• cuSPARSE – CUDA Sparse Matrix library
• NPP – NVIDIA Performance Primitives library
• nvGRAPH – NVIDIA Graph Analytics library
• NVML – NVIDIA Management Library
• NVRTC – NVIDIA Runtime Compilation library for CUDA C++

CUDA 8.0 comes with these other software components:

• nView – NVIDIA nView Desktop Management Software
• NVWMI – NVIDIA Enterprise Management Toolkit
• GameWorks PhysX – is a multi-platform game physics engine

CUDA 9.0–9.2 comes with these other components:

• CUTLASS 1.0 – custom linear algebra algorithms,
• NVIDIA Video Decoder was deprecated in CUDA 9.2; it is now available in NVIDIA Video Codec SDK

CUDA 10 comes with these other components:

• nvJPEG – Hybrid (CPU and GPU) JPEG processing

CUDA 11.0–11.8 comes with these other components:

• CUB is new one of more supported C++ libraries
• MIG multi instance GPU support
• nvJPEG2000 – JPEG 2000 encoder and decoder

Advantages

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.
• Unified virtual memory (CUDA 4.0 and above)
• Unified memory (CUDA 6.0 and above)
• Shared memory – CUDA exposes a fast shared memory region that can be shared among 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

• Whether for the host computer or the GPU device, all CUDA source code is now processed according to C++ syntax rules. This was not always the case. Earlier versions of CUDA were based on C syntax rules. As with the more general case of compiling C code with a C++ compiler, it is therefore possible that old C-style CUDA source code will either fail to compile or will not behave as originally intended.
• Interoperability with rendering languages such as OpenGL is one-way, with OpenGL having access to registered CUDA memory but CUDA not having access to OpenGL memory.
• 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).
• No emulation or fallback functionality is available for modern revisions.
• Valid C++ may sometimes be flagged and prevent compilation due to the way the compiler approaches optimization for target GPU device limitations.
• C++ run-time type information (RTTI) and C++-style exception handling are only supported in host code, not in device code.
• In single-precision on first generation CUDA compute capability 1.x devices, denormal numbers are unsupported and are instead flushed to zero, and the precision of both the division and square root operations are slightly lower than IEEE 754-compliant single precision math. Devices that support compute capability 2.0 and above support denormal numbers, and the division and square root operations are IEEE 754 compliant by default. However, users can obtain the prior faster gaming-grade math of compute capability 1.x devices if desired by setting compiler flags to disable accurate divisions and accurate square roots, and enable flushing denormal numbers to zero.
• Unlike OpenCL, CUDA-enabled GPUs are only available from Nvidia as it is proprietary. Attempts to implement CUDA on other GPUs include:
  • Project Coriander: Converts CUDA C++11 source to OpenCL 1.2 C. A fork of CUDA-on-CL intended to run TensorFlow.
  • CU2CL: Convert CUDA 3.2 C++ to OpenCL C.
  • GPUOpen HIP: A thin abstraction layer on top of CUDA and ROCm intended for AMD and Nvidia GPUs. Has a conversion tool for importing CUDA C++ source. Supports CUDA 4.0 plus C++11 and float16.
  • ZLUDA is a drop-in replacement for CUDA on AMD GPUs and formerly Intel GPUs with near-native performance. The developer, Andrzej Janik, was separately contracted by both Intel and AMD to develop the software in 2021 and 2022, respectively. However, neither company decided to release it officially due to the lack of a business use case. AMD's contract included a clause that allowed Janik to release his code for AMD independently, allowing him to release the new version that only supports AMD GPUs.
  • chipStar can compile and run CUDA/HIP programs on advanced OpenCL 3.0 or Level Zero platforms.

Example

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 = cudaAddressModeClamp;
  tex.addressMode = 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 = 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 = a * b;
}
"""
)
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(, ], numpy.float32))
B = CUBLASMatrix(numpy.mat(, , ], numpy.float32))
C = A * B
print(C.np_mat())

while CuPy directly replaces NumPy:

import cupy
a = cupy.random.randn(400)
b = cupy.random.randn(400)
dest = cupy.zeros_like(a)
print(dest - a * b)

GPUs supported

Supported CUDA Compute Capability versions for CUDA SDK version and Microarchitecture (by code name):

Note: CUDA SDK 10.2 is the last official release for macOS, as support will not be available for macOS in newer releases.

CUDA Compute Capability by version with associated GPU semiconductors and GPU card models (separated by their various application areas):

'*' – OEM-only products

Version features and specifications

This section needs to be updated. The reason given is: Missing CUDA compute capability 10.x (Blackwell). Please help update this article to reflect recent events or newly available information. (March 2024)

Data types

Note: Any missing lines or empty entries do reflect some lack of information on that exact item.

Tensor cores

Note: Any missing lines or empty entries do reflect some lack of information on that exact item.

Technical specification

Multiprocessor architecture

For more information read the Nvidia CUDA programming guide.

Current and future usages of CUDA architecture

• Accelerated rendering of 3D graphics
• Accelerated interconversion of video file formats
• Accelerated encryption, decryption and compression
• Bioinformatics, e.g. NGS DNA sequencing BarraCUDA
• Distributed calculations, such as predicting the native conformation of proteins
• Medical analysis simulations, for example virtual reality based on CT and MRI scan images
• Physical simulations, in particular in fluid dynamics
• Neural network training in machine learning problems
• Large Language Model inference
• Face recognition
• Volunteer computing projects, such as SETI@home and other projects using BOINC software
• Molecular dynamics
• Mining cryptocurrencies
• Structure from motion (SfM) software

Comparison with competitors

CUDA competes with other GPU computing stacks: Intel OneAPI and AMD ROCm.

Whereas Nvidia's CUDA is closed-source, Intel's OneAPI and AMD's ROCm are open source.

Intel OneAPI

oneAPI is an initiative based in open standards, created to support software development for multiple hardware architectures. The oneAPI libraries must implement open specifications that are discussed publicly by the Special Interest Groups, offering the possibility for any developer or organization to implement their own versions of oneAPI libraries.

Originally made by Intel, other hardware adopters include Fujitsu and Huawei.

Unified Acceleration Foundation (UXL)

Unified Acceleration Foundation (UXL) is a new technology consortium working on the continuation of the OneAPI initiative, with the goal to create a new open standard accelerator software ecosystem, related open standards and specification projects through Working Groups and Special Interest Groups (SIGs). The goal is to offer open alternatives to Nvidia's CUDA. The main companies behind it are Intel, Google, ARM, Qualcomm, Samsung, Imagination, and VMware.

AMD ROCm

ROCm is an open source software stack for graphics processing unit (GPU) programming from Advanced Micro Devices (AMD).

See also

• SYCL – an open standard from Khronos Group for programming a variety of platforms, including GPUs, with single-source modern C++, similar to higher-level CUDA Runtime API (single-source)
• BrookGPU – the Stanford University graphics group's compiler
• Array programming
• Parallel computing
• Stream processing
• rCUDA – an API for computing on remote computers
• Molecular modeling on GPUs
• Vulkan – low-level, high-performance 3D graphics and computing API
• OptiX – ray tracing API by NVIDIA
• CUDA binary (cubin) – a type of fat binary
• Numerical Library Collection – by NEC for their vector processor

References

1. ^ Shah, Agam. "Nvidia not totally against third parties making CUDA chips". www.theregister.com. Retrieved 2024-04-25.
2. "Nvidia CUDA Home Page". 18 July 2017.
3. Shimpi, Anand Lal; Wilson, Derek (November 8, 2006). "Nvidia's GeForce 8800 (G80): GPUs Re-architected for DirectX 10". AnandTech. Retrieved May 16, 2015.
4. "Introduction — nsight-visual-studio-edition 12.6 documentation". docs.nvidia.com. Retrieved 2024-10-10.
5. ^ Abi-Chahla, Fedy (June 18, 2008). "Nvidia's CUDA: The End of the CPU?". Tom's Hardware. Retrieved May 17, 2015.
6. Zunitch, Peter (2018-01-24). "CUDA vs. OpenCL vs. OpenGL". Videomaker. Retrieved 2018-09-16.
7. "OpenCL". NVIDIA Developer. 2013-04-24. Retrieved 2019-11-04.
8. Witt, Stephen (2023-11-27). "How Jensen Huang's Nvidia Is Powering the A.I. Revolution". The New Yorker. ISSN 0028-792X. Retrieved 2023-12-10.
9. "CUDA LLVM Compiler". 7 May 2012.
10. First OpenCL demo on a GPU on YouTube
11. DirectCompute Ocean Demo Running on Nvidia CUDA-enabled GPU on YouTube
12. Vasiliadis, Giorgos; Antonatos, Spiros; Polychronakis, Michalis; Markatos, Evangelos P.; Ioannidis, Sotiris (September 2008). "Gnort: High Performance Network Intrusion Detection Using Graphics Processors" (PDF). Recent Advances in Intrusion Detection. Lecture Notes in Computer Science. Vol. 5230. pp. 116–134. doi:10.1007/978-3-540-87403-4_7. ISBN 978-3-540-87402-7.
13. Schatz, Michael C.; Trapnell, Cole; Delcher, Arthur L.; Varshney, Amitabh (2007). "High-throughput sequence alignment using Graphics Processing Units". BMC Bioinformatics. 8: 474. doi:10.1186/1471-2105-8-474. PMC 2222658. PMID 18070356.
14. Manavski, Svetlin A.; Giorgio, Valle (2008). "CUDA compatible GPU cards as efficient hardware accelerators for Smith-Waterman sequence alignment". BMC Bioinformatics. 10 (Suppl 2): S10. doi:10.1186/1471-2105-9-S2-S10. PMC 2323659. PMID 18387198.
15. "Pyrit – Google Code".
16. "Use your Nvidia GPU for scientific computing". BOINC. 2008-12-18. Archived from the original on 2008-12-28. Retrieved 2017-08-08.
17. "Nvidia CUDA Software Development Kit (CUDA SDK) – Release Notes Version 2.0 for MAC OS X". Archived from the original on 2009-01-06.
18. "CUDA 1.1 – Now on Mac OS X". February 14, 2008. Archived from the original on November 22, 2008.
19. "CUDA 11 Features Revealed". 14 May 2020.
20. "CUDA Toolkit 11.1 Introduces Support for GeForce RTX 30 Series and Quadro RTX Series GPUs". 23 September 2020.
21. "Enhancing Memory Allocation with New NVIDIA CUDA 11.2 Features". 16 December 2020.
22. "Exploring the New Features of CUDA 11.3". 16 April 2021.
23. Silberstein, Mark; Schuster, Assaf; Geiger, Dan; Patney, Anjul; Owens, John D. (2008). "Efficient computation of sum-products on GPUs through software-managed cache" (PDF). Proceedings of the 22nd annual international conference on Supercomputing – ICS '08 (PDF). Proceedings of the 22nd annual international conference on Supercomputing – ICS '08. pp. 309–318. doi:10.1145/1375527.1375572. ISBN 978-1-60558-158-3.
24. "CUDA C Programming Guide v8.0" (PDF). nVidia Developer Zone. January 2017. p. 19. Retrieved 22 March 2017.
25. "NVCC forces c++ compilation of .cu files". 29 November 2011.
26. Whitehead, Nathan; Fit-Florea, Alex. "Precision & Performance: Floating Point and IEEE 754 Compliance for Nvidia GPUs" (PDF). Nvidia. Retrieved November 18, 2014.
27. "CUDA-Enabled Products". CUDA Zone. Nvidia Corporation. Retrieved 2008-11-03.
28. "Coriander Project: Compile CUDA Codes To OpenCL, Run Everywhere". Phoronix.
29. Perkins, Hugh (2017). "cuda-on-cl" (PDF). IWOCL. Retrieved August 8, 2017.
30. "hughperkins/coriander: Build NVIDIA® CUDA™ code for OpenCL™ 1.2 devices". GitHub. May 6, 2019.
31. "CU2CL Documentation". chrec.cs.vt.edu.
32. "GitHub – vosen/ZLUDA". GitHub.
33. Larabel, Michael (2024-02-12), "AMD Quietly Funded A Drop-In CUDA Implementation Built On ROCm: It's Now Open-Source", Phoronix, retrieved 2024-02-12
34. "GitHub – chip-spv/chipStar". GitHub.
35. "PyCUDA".
36. "pycublas". Archived from the original on 2009-04-20. Retrieved 2017-08-08.
37. "CuPy". Retrieved 2020-01-08.
38. "NVIDIA CUDA Programming Guide. Version 1.0" (PDF). June 23, 2007.
39. "NVIDIA CUDA Programming Guide. Version 2.1" (PDF). December 8, 2008.
40. "NVIDIA CUDA Programming Guide. Version 2.2" (PDF). April 2, 2009.
41. "NVIDIA CUDA Programming Guide. Version 2.2.1" (PDF). May 26, 2009.
42. "NVIDIA CUDA Programming Guide. Version 2.3.1" (PDF). August 26, 2009.
43. "NVIDIA CUDA Programming Guide. Version 3.0" (PDF). February 20, 2010.
44. "NVIDIA CUDA C Programming Guide. Version 3.1.1" (PDF). July 21, 2010.
45. "NVIDIA CUDA C Programming Guide. Version 3.2" (PDF). November 9, 2010.
46. "CUDA 11.0 Release Notes". NVIDIA Developer.
47. "CUDA 11.1 Release Notes". NVIDIA Developer.
48. "CUDA 11.5 Release Notes". NVIDIA Developer.
49. "CUDA 11.8 Release Notes". NVIDIA Developer.
50. "NVIDIA Quadro NVS 420 Specs". TechPowerUp GPU Database. 25 August 2023.
51. Larabel, Michael (March 29, 2017). "NVIDIA Rolls Out Tegra X2 GPU Support In Nouveau". Phoronix. Retrieved August 8, 2017.
52. Nvidia Xavier Specs on TechPowerUp (preliminary)
53. "Welcome — Jetson LinuxDeveloper Guide 34.1 documentation".
54. "NVIDIA Bringing up Open-Source Volta GPU Support for Their Xavier SoC".
55. "NVIDIA Ada Lovelace Architecture".
56. Dissecting the Turing GPU Architecture through Microbenchmarking
57. "H.1. Features and Technical Specifications – Table 13. Feature Support per Compute Capability". docs.nvidia.com. Retrieved 2020-09-23.
58. "CUDA C++ Programming Guide".
59. Fused-Multiply-Add, actually executed, Dense Matrix
60. as SASS since 7.5, as PTX since 8.0
61. unofficial support in SASS
62. unofficial support in SASS
63. "Technical brief. NVIDIA Jetson AGX Orin Series" (PDF). nvidia.com. Retrieved 5 September 2023.
64. "NVIDIA Ampere GA102 GPU Architecture" (PDF). nvidia.com. Retrieved 5 September 2023.
65. Luo, Weile; Fan, Ruibo; Li, Zeyu; Du, Dayou; Wang, Qiang; Chu, Xiaowen (2024). "Benchmarking and Dissecting the Nvidia Hopper GPU Architecture". arXiv:2402.13499v1 .
66. "Datasheet NVIDIA A40" (PDF). nvidia.com. Retrieved 27 April 2024.
67. "NVIDIA AMPERE GA102 GPU ARCHITECTURE" (PDF). 27 April 2024.
68. "Datasheet NVIDIA L40" (PDF). 27 April 2024.
69. In the Whitepapers the Tensor Core cube diagrams represent the Dot Product Unit Width into the height (4 FP16 for Volta and Turing, 8 FP16 for A100, 4 FP16 for GA102, 16 FP16 for GH100). The other two dimensions represent the number of Dot Product Units (4x4 = 16 for Volta and Turing, 8x4 = 32 for Ampere and Hopper). The resulting gray blocks are the FP16 FMA operations per cycle. Pascal without Tensor core is only shown for speed comparison as is Volta V100 with non-FP16 datatypes.
70. "NVIDIA Turing Architecture Whitepaper" (PDF). nvidia.com. Retrieved 5 September 2023.
71. "NVIDIA Tensor Core GPU" (PDF). nvidia.com. Retrieved 5 September 2023.
72. "NVIDIA Hopper Architecture In-Depth". 22 March 2022.
73. shape x converted operand size, e.g. 2 tensor cores x 4x4x4xFP16/cycle = 256 Bytes/cycle
74. = product first 3 table rows
75. = product of previous 2 table rows; shape: e.g. 8x8x4xFP16 = 512 Bytes
76. Sun, Wei; Li, Ang; Geng, Tong; Stuijk, Sander; Corporaal, Henk (2023). "Dissecting Tensor Cores via Microbenchmarks: Latency, Throughput and Numeric Behaviors". IEEE Transactions on Parallel and Distributed Systems. 34 (1): 246–261. arXiv:2206.02874. doi:10.1109/tpds.2022.3217824. S2CID 249431357.
77. "Parallel Thread Execution ISA Version 7.7".
78. Raihan, Md Aamir; Goli, Negar; Aamodt, Tor (2018). "Modeling Deep Learning Accelerator Enabled GPUs". arXiv:1811.08309 .
79. "NVIDIA Ada Lovelace Architecture".
80. shape x converted operand size, e.g. 2 tensor cores x 4x4x4xFP16/cycle = 256 Bytes/cycle
81. = product first 3 table rows
82. = product of previous 2 table rows; shape: e.g. 8x8x4xFP16 = 512 Bytes
83. Jia, Zhe; Maggioni, Marco; Smith, Jeffrey; Daniele Paolo Scarpazza (2019). "Dissecting the NVidia Turing T4 GPU via Microbenchmarking". arXiv:1903.07486 .
84. Burgess, John (2019). "RTX ON – The NVIDIA TURING GPU". 2019 IEEE Hot Chips 31 Symposium (HCS). pp. 1–27. doi:10.1109/HOTCHIPS.2019.8875651. ISBN 978-1-7281-2089-8. S2CID 204822166.
85. Jia, Zhe; Maggioni, Marco; Smith, Jeffrey; Daniele Paolo Scarpazza (2019). "Dissecting the NVidia Turing T4 GPU via Microbenchmarking". arXiv:1903.07486 .
86. Burgess, John (2019). "RTX ON – The NVIDIA TURING GPU". 2019 IEEE Hot Chips 31 Symposium (HCS). pp. 1–27. doi:10.1109/HOTCHIPS.2019.8875651. ISBN 978-1-7281-2089-8. S2CID 204822166.
87. dependent on device
88. "Tegra X1". 9 January 2015.
89. NVIDIA H100 Tensor Core GPU Architecture
90. H.1. Features and Technical Specifications – Table 14. Technical Specifications per Compute Capability
91. NVIDIA Hopper Architecture In-Depth
92. can only execute 160 integer instructions according to programming guide
93. 128 according to . 64 from FP32 + 64 separate units?
94. 64 by FP32 cores and 64 by flexible FP32/INT cores.
95. "CUDA C++ Programming Guide".
96. 32 FP32 lanes combine to 16 FP64 lanes. Maybe lower depending on model.
97. only supported by 16 FP32 lanes, they combine to 4 FP64 lanes
98. ^ depending on model
99. Effective speed, probably over FP32 ports. No description of actual FP64 cores.
100. Can also be used for integer additions and comparisons
101. 2 clock cycles/instruction for each SM partition Burgess, John (2019). "RTX ON – The NVIDIA TURING GPU". 2019 IEEE Hot Chips 31 Symposium (HCS). pp. 1–27. doi:10.1109/HOTCHIPS.2019.8875651. ISBN 978-1-7281-2089-8. S2CID 204822166.
102. Durant, Luke; Giroux, Olivier; Harris, Mark; Stam, Nick (May 10, 2017). "Inside Volta: The World's Most Advanced Data Center GPU". Nvidia developer blog.
103. The schedulers and dispatchers have dedicated execution units unlike with Fermi and Kepler.
104. Dispatching can overlap concurrently, if it takes more than one cycle (when there are less execution units than 32/SM Partition)
105. Can dual issue MAD pipe and SFU pipe
106. No more than one scheduler can issue 2 instructions at once. The first scheduler is in charge of warps with odd IDs. The second scheduler is in charge of warps with even IDs.
107. ^ shared memory only, no data cache
108. ^ shared memory separate, but L1 includes texture cache
109. "H.6.1. Architecture". docs.nvidia.com. Retrieved 2019-05-13.
110. "Demystifying GPU Microarchitecture through Microbenchmarking" (PDF).
111. Jia, Zhe; Maggioni, Marco; Staiger, Benjamin; Scarpazza, Daniele P. (2018). "Dissecting the NVIDIA Volta GPU Architecture via Microbenchmarking". arXiv:1804.06826 .
112. "Tegra X1". 9 January 2015.
113. Jia, Zhe; Maggioni, Marco; Smith, Jeffrey; Daniele Paolo Scarpazza (2019). "Dissecting the NVidia Turing T4 GPU via Microbenchmarking". arXiv:1903.07486 .
114. "Dissecting the Ampere GPU Architecture through Microbenchmarking".
115. Note that Jia, Zhe; Maggioni, Marco; Smith, Jeffrey; Daniele Paolo Scarpazza (2019). "Dissecting the NVidia Turing T4 GPU via Microbenchmarking". arXiv:1903.07486 . disagrees and states 2 KiB L0 instruction cache per SM partition and 16 KiB L1 instruction cache per SM
116. Jia, Zhe; Maggioni, Marco; Staiger, Benjamin; Scarpazza, Daniele P. (2018). "Dissecting the NVIDIA Volta GPU Architecture via Microbenchmarking". arXiv:1804.06826 .
117. "asfermi Opcode". GitHub.
118. ^ for access with texture engine only
119. 25% disabled on RTX 4090
120. "I.7. Compute Capability 8.x". docs.nvidia.com. Retrieved 2022-10-12.
121. "Appendix F. Features and Technical Specifications" (PDF). (3.2 MiB), page 148 of 175 (Version 5.0 October 2012).
122. "nVidia CUDA Bioinformatics: BarraCUDA". BioCentric. 2019-07-19. Retrieved 2019-10-15.
123. "Part V: Physics Simulation". NVIDIA Developer. Retrieved 2020-09-11.
124. "oneAPI Programming Model". oneAPI.io. Retrieved 2024-07-27.
125. "Specifications | oneAPI". oneAPI.io. Retrieved 2024-07-27.
126. "oneAPI Specification — oneAPI Specification 1.3-rev-1 documentation". oneapi-spec.uxlfoundation.org. Retrieved 2024-07-27.
127. "Exclusive: Behind the plot to break Nvidia's grip on AI by targeting software". Reuters. Retrieved 2024-04-05.
128. "Question: What does ROCm stand for? · Issue #1628 · RadeonOpenCompute/ROCm". Github.com. Retrieved January 18, 2022.

Further reading

• Buck, Ian; Foley, Tim; Horn, Daniel; Sugerman, Jeremy; Fatahalian, Kayvon; Houston, Mike; Hanrahan, Pat (2004-08-01). "Brook for GPUs: stream computing on graphics hardware". ACM Transactions on Graphics. 23 (3): 777–786. doi:10.1145/1015706.1015800. ISSN 0730-0301.
• Nickolls, John; Buck, Ian; Garland, Michael; Skadron, Kevin (2008-03-01). "Scalable Parallel Programming with CUDA: Is CUDA the parallel programming model that application developers have been waiting for?". Queue. 6 (2): 40–53. doi:10.1145/1365490.1365500. ISSN 1542-7730.

External links

• Official website

• Computer physics engines
• GPGPU
• GPGPU libraries
• Graphics hardware
• Nvidia software
• Parallel computing
• Graphics cards
• Video game hardware