CUDA ("Compute Unified Device Architecture"), is a GPGPU technology that allows a programmer to use the C programming language to code algorithms for execution on the GPU. CUDA has been developed by NVIDIA and to use this architecture requires an NVIDIA GPU and special stream processing drivers. CUDA only works with the new GeForce 8 Series, featuring G8X GPUs; NVIDIA guarantees that programs developed for the GeForce 8 series will also work without modification on all future NVIDIA video cards. CUDA gives developers unfettered access to the native instruction set and memory of the massively parallel computational elements in CUDA GPUs. Using CUDA, NVIDIA GeForce-based GPUs effectively become powerful, programmable open architectures like today’s CPUs (Central Processing Units). By opening up the architecture, CUDA provides developers both with the low-level, deterministic, and for repeatable access to hardware that is necessary API to develop essential high-level programming tools such as compilers, debuggers, math libraries, and application platforms. General-purpose computing on graphics processing units (GPGPU, also referred to as GPGP and to a lesser extent GP²) is a recent trend focused on using GPUs to perform computations rather than the CPU. The addition of programmable stages and higher precision arithmetic to the rendering pipelines allowed software developers... Wikibooks has a book on the topic of C Programming The C programming language (often, just C) is a general-purpose, procedural, imperative computer programming language developed in the early 1970s by Dennis Ritchie for use on the Unix operating system. ... GPU may stand for: Graphics processing unit, a special stream processor used in computer graphics hardware Gosudarstvennoye Politicheskoye Upravlenie (Главное Политическое Управление, or Main Political Directorate) of the Red Army, responsible for troops morale and propaganda. ... NVIDIA Corporation (NASDAQ: NVDA) (pronounced ) is an American corporation specializing in the manufacture of GPU technologies for video cards, graphics cards, workstations, desktop computers, handhelds and more. ... It has been suggested that GPGPU be merged into this article or section. ... The Geforce 8 Series is the eighth generation of NVIDIAs GeForce graphics cards. ... Parallel computing is the simultaneous execution of the same task (split up and specially adapted) on multiple processors in order to obtain results faster. ... The new GeForce logo. ... CPU can stand for: in computing: Central processing unit in journalism: Commonwealth Press Union in law enforcement: Crime prevention unit in software: Critical patch update, a type of software patch distributed by Oracle Corporation in Macleans College is often known as Ash Lim. ... “CPU” redirects here. ... API may refer to: In computing, application programming interface In petroleum industry, American Petroleum Institute In education, Academic Performance Index This page concerning a three-letter acronym or abbreviation is a disambiguation page — a navigational aid which lists other pages that might otherwise share the same title. ... A compiler is a computer program that translates a computer program written in one computer language (called the source language) into an equivalent program written in another computer language (called the output or the target language). ... A debugger is a computer program that is used to debug (and sometimes test or optimize) other programs. ... Alternative meanings: Library (computer science), Library (biology) Modern-style library In its traditional sense, a library is a collection of books and periodicals. ...

The initial CUDA SDK was made public 15th February 2007.^[1] The compiler in CUDA is based on Open64. Open64 is an open source, state-of-art, optimizing compiler for Intel Itanium platform. ...

Hardware

The 8-Series (G8X) GeForce-based GPU from NVIDIA is the first series of GPU to support the CUDA SDK. The 8-Series (G8X) GPUs features hardware support for 32-bit (single precision) floating point vector processors, using the CUDA SDK as API. (CUDA supports the C "double" data type, however on G8X series GPUs these types will get demoted to 32-bit floats.). Due to the highly parallel nature of vector processors, GPU assisted hardware stream processing can have a huge impact in specific data processing applications. It is anticipated in the computer gaming industry that graphics cards may be used in future game physics calculations (physical effects like debris, smoke, fire, fluids). The new GeForce logo. ... GPU may stand for: Graphics processing unit, a special stream processor used in computer graphics hardware Gosudarstvennoye Politicheskoye Upravlenie (Главное Политическое Управление, or Main Political Directorate) of the Red Army, responsible for troops morale and propaganda. ... NVIDIA Corporation (NASDAQ: NVDA) (pronounced ) is an American corporation specializing in the manufacture of GPU technologies for video cards, graphics cards, workstations, desktop computers, handhelds and more. ... GPU may stand for: Graphics processing unit, a special stream processor used in computer graphics hardware Gosudarstvennoye Politicheskoye Upravlenie (Главное Политическое Управление, or Main Political Directorate) of the Red Army, responsible for troops morale and propaganda. ... A floating-point number is a digital representation for a number in a certain subset of the rational numbers, and is often used to approximate an arbitrary real number on a computer. ... A vector processor, or array processor, is a CPU design that is able to run mathematical operations on a large number of data elements very quickly. ... API may refer to: In computing, application programming interface In petroleum industry, American Petroleum Institute In education, Academic Performance Index This page concerning a three-letter acronym or abbreviation is a disambiguation page — a navigational aid which lists other pages that might otherwise share the same title. ... It has been suggested that GPGPU be merged into this article or section. ... ... A Physics Processing Unit (PPU) is a dedicated microprocessor designed to handle the calculations of physics, especially in the physics engine of video games. ...

Advantages

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

• It uses the standard C language, with some simple extensions.
• Scattered writes - code can write to arbitrary addresses in memory.
• Shared memory - CUDA exposes a fast shared memory region (16KB in size) 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

There are very few or no other articles that link to this one. ...

Limitations

• Compared to traditional floating point accelerators such as the 64-bit floating point (FP64) CSX600 boards from ClearSpeed that is used in today's supercomputers, the current GPUs from NVIDIA (and AMD/ATI) are only running on 32-bit, providing only single precision data capability^[2] – instead of double precision (64-bit) capability of supercomputers.^[3] NVIDIA stated in the CUDA Release Notes Version 0.8 file that NVIDIA GPUs supporting (64-bit) Double Precision Floating Point arithmetic in hardware will become available in late 2007.^[4]
• Only bilinear texture filtering is supported - mipmapped textures and anisotropic filtering are not supported at this time.
• Recursive functions are not supported.
• Various deviations from the IEEE 754 standard. Denormals 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.
• The bus bandwidth and latency between the CPU and the GPU may be a bottleneck.
• Threads must run in groups of at least 32 threads that execute identical instructions simultaneously. Branching does not impact performance significantly, provided most threads take the same code path most of the time; but the SIMD execution model becomes a significant limitation for any inherently divergent task (e.g., traversing a ray tracing acceleration data structure).
• CUDA-enabled GPUs are available only from NVIDIA (GeForce, Quadro and Tesla)

A floating-point number is a digital representation for a number in a certain subset of the rational numbers, and is often used to approximate an arbitrary real number on a computer. ... ClearSpeed is a company that develops processors for supercomputers. ... A supercomputer is a device for turning compute-bound problems into I/O-bound problems. ... GPU may stand for: Graphics processing unit, a special stream processor used in computer graphics hardware Gosudarstvennoye Politicheskoye Upravlenie (Главное Политическое Управление, or Main Political Directorate) of the Red Army, responsible for troops morale and propaganda. ... NVIDIA Corporation (NASDAQ: NVDA) (pronounced ) is an American corporation specializing in the manufacture of GPU technologies for video cards, graphics cards, workstations, desktop computers, handhelds and more. ... Advanced Micro Devices, Inc. ... ATI may stand for: ATI Technologies Inc. ... In computing, single precision is a computer numbering format that occupies one storage locations in computer memory at address. ... In computing, double precision is a computer numbering format that occupies two storage locations in computer memory at address and address+1. ... A supercomputer is a device for turning compute-bound problems into I/O-bound problems. ... The IEEE Standard for Binary Floating-Point Arithmetic (IEEE 754) is the most widely-used standard for floating-point computation, and is followed by many CPU and FPU implementations. ... In computer science, denormal numbers or denormalized numbers (now often called subnormal numbers) fill the gap around zero in floating point arithmetic: any non-zero number which is smaller than the smallest normal number is sub-normal. For example, if the smallest positive normal number is 1×β-n (where β is... In computing, NaN (Not a Number) is a value or symbol that is usually produced as the result of an operation on invalid input operands, especially in floating-point calculations. ... Rounding is the process of reducing the number of significant digits in a number. ...

See also

• NVIDIA Corporation
• CTM (Close To Metal), AMD/ATI's competing GPGPU technology for AMD Stream Processors and FireGL workstation cards
• Graphics Processing Unit (GPU)
• GPGPU (General-Purpose Computing on Graphics Processing Units)
• Stream Processor
• Stream programming
• Shader
• Sh, a GPGPU library for C++
• Graphics processing unit
  • Comparison of ATI Graphics Processing Units
  • Comparison of NVIDIA Graphics Processing Units
  • Graphics card
• Physics Processing Unit (PPU)

NVIDIA Corporation (NASDAQ: NVDA) is a major supplier of graphics processors (graphics processing units, GPUs), graphics cards, and media and communications devices for PCs and game consoles (Xbox). ... This articles Limitations section does not cite any references or sources. ... Advanced Micro Devices, Inc. ... ATI may stand for: ATI Technologies Inc. ... General-purpose computing on graphics processing units (GPGPU, also referred to as GPGP and to a lesser extent GP²) is a recent trend focused on using GPUs to perform computations rather than the CPU. The addition of programmable stages and higher precision arithmetic to the rendering pipelines allowed software developers... The AMD Stream Processor (previously branded as ATI FireStream) is a stream processor developed by ATI Technologies. ... The FireGL range of graphics cards is a series fabricated by ATI for use with CAD (Computer Aided Design) and DCC (Digital Content Creation) programs, usually found in workstations. ... GeForce 6600GT (NV43) GPU Radeon 9800 Pro (R350) GPU Intel GMA X3000 IGP “GPU” redirects here. ... General-purpose computing on graphics processing units (GPGPU, also referred to as GPGP and to a lesser extent GP²) is a recent trend focused on using GPUs to perform computations rather than the CPU. The addition of programmable stages and higher precision arithmetic to the rendering pipelines allowed software developers... A relatively new, yet very successful paradigm to allow parallel processing at never-seen-before efficiency with minimal effort. ... It has been suggested that this article or section be merged with stream processing. ... Shaders are a set of different technologies. ... Lib Sh is a metaprogramming language for programmable GPUs. ... GeForce 6600GT (NV43) GPU Radeon 9800 Pro (R350) GPU Intel GMA X3000 IGP “GPU” redirects here. ... This table contains general information about ATIs GPUs and videocards based on official ATI specifications. ... This table contains general information about NVIDIAs GPUs and videocards based on official NVIDIA specifications. ... A graphics/video/display card/board/adapter is a computer component designed to convert the logical representation of visual information into a signal that can be used as input for a display medium. ... A Physics Processing Unit (PPU) is a dedicated microprocessor designed to handle the calculations of physics, especially in the physics engine of video games. ...

References

February 16 is the 47th day of the year in the Gregorian calendar. ... Year 2007 (MMVII) is the current year, a common year starting on Monday of the Gregorian calendar and the AD/CE era. ... Year 2007 (MMVII) is the current year, a common year starting on Monday of the Gregorian calendar and the AD/CE era. ... is the 213th day of the year (214th in leap years) in the Gregorian calendar. ...

External links

• NVIDIA CUDA Homepage
• NVIDIA CUDA GPU Computing developer forums
• NVIDIA CUDA developer registration for professional developers and researchers
• Beyond3D - Introducing CUDA NVIDIA's Vision for GPU Computing (16th February 2007)
• University of Illinois NVIDIA CUDA Course taught by Wen-mei Hwu and David Kirk (9th March 2007)