SIMD programming illustrated using Intel's MMX instruction set
S. Krishnaprasad
Abstract
S. Krishnaprasad
Abstract
Parallel and vector processor architectures support SIMD-style (Single-Instruction, Multiple-Data) programming where a common operation is performed simultaneously on multiple data elements to improve performance. Intel Architecture's MMX technology, designed and optimized for multimedia applications, supports SIMD-style programming. In this paper we illustrate the use of Intel's MMX instruction set with simple SIMD-style coding examples. These codes are embedded in Visual C++ programs so that interested readers can quickly verify the logic and develop their own applications. The paper is also intended to help students and professionals to acquire the necessary skills to appreciate the advanced processor technology.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Parallel and vector processor architectures support SIMD-style (Single-Instruction, Multiple-Data) programming where a common operation is performed simultaneously on multiple data elements to improve performance. Intel Architecture's MMX technology, designed and optimized for multimedia applications, supports SIMD-style programming. In this paper we illustrate the use of Intel's MMX instruction set with simple SIMD-style coding examples. These codes are embedded in Visual C++ programs so that interested readers can quickly verify the logic and develop their own applications. The paper is also intended to help students and professionals to acquire the necessary skills to appreciate the advanced processor technology.
Key concepts: MMX, SIMD, Computer science, Instruction set, Parallel computing, Programming style, Coding (social sciences), Set (abstract data type)