computing Computing is any goal-oriented activity requiring, benefiting from, or creating computing machinery. It includes the study and experimentation of algorithmic processes, and development of both hardware and software. Computing has scientific, e ...

, a compute kernel is a routine compiled for high throughput accelerators (such as

graphics processing units A graphics processing unit (GPU) is a specialized electronic circuit designed to manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. GPUs are used in embedded systems, mobil ...

(GPUs),

digital signal processors A digital signal processor (DSP) is a specialized microprocessor chip, with its architecture optimized for the operational needs of digital signal processing. DSPs are fabricated on MOS integrated circuit chips. They are widely used in audio sig ...

(DSPs) or

field-programmable gate arrays A field-programmable gate array (FPGA) is an integrated circuit designed to be configured by a customer or a designer after manufacturinghence the term '' field-programmable''. The FPGA configuration is generally specified using a hardware de ...

(FPGAs)), separate from but used by a main program (typically running on a

central processing unit A central processing unit (CPU), also called a central processor, main processor or just processor, is the electronic circuitry that executes instructions comprising a computer program. The CPU performs basic arithmetic, logic, controlling, an ...

). They are sometimes called compute shaders, sharing

execution unit In computer engineering, an execution unit (E-unit or EU) is a part of the central processing unit (CPU) that performs the operations and calculations as instructed by the computer program. It may have its own internal control sequence unit (not ...

s with

vertex shaders In computer graphics, a shader is a computer program that calculates the appropriate levels of light, darkness, and color during the rendering of a 3D scene - a process known as ''shading''. Shaders have evolved to perform a variety of speci ...

and

pixel shaders In computer graphics, a shader is a computer program that calculates the appropriate levels of light, darkness, and color during the rendering of a 3D scene - a process known as ''shading''. Shaders have evolved to perform a variety of speci ...

on GPUs, but are not limited to execution on one class of device, or

graphics API A graphics library is a program library designed to aid in rendering computer graphics to a monitor. This typically involves providing optimized versions of functions that handle common rendering tasks. This can be done purely in software and runn ...

Description

Compute kernels roughly correspond to

inner loop Inner loop may refer to: *Inner loop in computer programs *Inner Loop (Phoenix), a section of Interstate 10 in downtown Phoenix, Arizona, United States *Inner Loop (Rochester), an expressway around downtown Rochester, New York, United States * Inne ...

s when implementing algorithms in traditional languages (except there is no implied sequential operation), or to code passed to internal iterators. They may be specified by a separate

programming language A programming language is a system of notation for writing computer programs. Most programming languages are text-based formal languages, but they may also be graphical. They are a kind of computer language. The description of a programming ...

such as "

OpenCL C OpenCL (Open Computing Language) is a framework for writing programs that execute across heterogeneous platforms consisting of central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), field-pr ...

" (managed by the

OpenCL OpenCL (Open Computing Language) is a framework for writing programs that execute across heterogeneous platforms consisting of central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), field-progra ...

API), as "compute

shader In computer graphics, a shader is a computer program that calculates the appropriate levels of light, darkness, and color during the rendering of a 3D scene - a process known as ''shading''. Shaders have evolved to perform a variety of spec ...

s" written in a

shading language A shading language is a graphics programming language adapted to programming shader effects (characterizing surfaces, volumes, and objects). Such language forms usually consist of special data types, like "vector", "matrix", "color" and "normal". ...

(managed by a graphics API such as

OpenGL OpenGL (Open Graphics Library) is a cross-language, cross-platform application programming interface (API) for rendering 2D and 3D vector graphics. The API is typically used to interact with a graphics processing unit (GPU), to achieve hardwa ...

), or embedded directly in

application code This glossary of computer software terms lists the general terms related to computer software, and related fields, as commonly used in Wikipedia articles. Glossary See also * Outline of computer programming * Outline of soft ...

written in a

high level language In computer science, a high-level programming language is a programming language with strong Abstraction (computer science), abstraction from the details of the computer. In contrast to low-level programming languages, it may use natural language ...

, as in the case of C++AMP.

Vector processing

This

programming paradigm Programming paradigms are a way to classify programming languages based on their features. Languages can be classified into multiple paradigms. Some paradigms are concerned mainly with implications for the execution model of the language, suc ...

maps well to

vector processor In computing, a vector processor or array processor is a central processing unit (CPU) that implements an instruction set where its instructions are designed to operate efficiently and effectively on large one-dimensional arrays of data called ...

s: there is an assumption that each invocation of a kernel within a batch is independent, allowing for

data parallel Data parallelism is parallelization across multiple processors in parallel computing environments. It focuses on distributing the data across different nodes, which operate on the data in parallel. It can be applied on regular data structures lik ...

execution. However,

atomic operations In concurrent programming, an operation (or set of operations) is linearizable if it consists of an ordered list of invocation and response events (event), that may be extended by adding response events such that: # The extended list can be re-e ...

may sometimes be used for

synchronization Synchronization is the coordination of events to operate a system in unison. For example, the conductor of an orchestra keeps the orchestra synchronized or ''in time''. Systems that operate with all parts in synchrony are said to be synchronou ...

between elements (for interdependent work), in some scenarios. Individual invocations are given indices (in 1 or more dimensions) from which arbitrary addressing of buffer data may be performed (including scatter gather operations), so long as the non-overlapping assumption is respected.

Vulkan API

The

Vulkan API Vulkan is a low- overhead, cross-platform API, open standard for 3D graphics and computing. Vulkan targets high-performance real-time 3D graphics applications, such as video games and interactive media. Vulkan is intended to offer higher perform ...

provides the intermediate

SPIR-V Standard Portable Intermediate Representation (SPIR) is an intermediate language for parallel compute and graphics by Khronos Group. It is used in multiple execution environments, including the Vulkan graphics API and the OpenCL compute API, to re ...

representation to describe ''both'' Graphical Shaders, and Compute Kernels, in a language independent and

machine independent In computing, cross-platform software (also called multi-platform software, platform-agnostic software, or platform-independent software) is computer software that is designed to work in several computing platforms. Some cross-platform software r ...

manner. The intention is to facilitate language evolution and provide a more natural ability to leverage GPU compute capabilities, in line with hardware developments such as

Unified Memory Architecture Unified may refer to: * The Unified, a wine symposium held in Sacramento, California, USA * ''Unified'', the official student newspaper of Canterbury Christ Church University * UNFD, an Australian record label * ''Unified'' (Sweet & Lynch album) ...

and

Heterogeneous System Architecture Heterogeneous System Architecture (HSA) is a cross-vendor set of specifications that allow for the integration of central processing units and graphics processors on the same bus, with shared memory and tasks. The HSA is being developed by the HSA ...

. This allows closer cooperation between a CPU and GPU.

References

GPGPU Parallel computing {{Graphics Processing Unit