2006Unpublished venueRequires access

Auto-vectorization of interleaved data for SIMD

Dorit Nuzman, Ira Rosen, Ayal Zaks

Open publisher page 127 citations

Abstract

Most implementations of the Single Instruction Multiple Data (SIMD) model available today require that data elements be packed in vector registers. Operations on disjoint vector elements are not supported directly and require explicit data reorganization manipulations. Computations on non-contiguous and especially interleaved data appear in important applications, which can greatly benefit from SIMD instructions once the data is reorganized properly. Vectorizing such computations efficiently is therefore an ambitious challenge for both programmers and vectorizing compilers. We demonstrate an automatic compilation scheme that supports effective vectorization in the presence of interleaved data with constant strides that are powers of 2, facilitating data reorganization. We demonstrate how our vectorization scheme applies to dominant SIMD architectures, and present experimental results on a wide range of key kernels, showing speedups in execution time up to 3.7 for interleaving levels (stride) as high as 8.

About this research paper

What this paper is about

Most implementations of the Single Instruction Multiple Data (SIMD) model available today require that data elements be packed in vector registers. Operations on disjoint vector elements are not supported directly and require explicit data reorganization manipulations. Computations on non-contiguous and especially interleaved data appear in important applications, which can greatly benefit from SIMD instructions once the data is reorganized properly. Vectorizing such computations efficiently is therefore an ambitious challenge for both programmers and vectorizing compilers. We demonstrate an automatic compilation scheme that supports effective vectorization in the presence of interleaved data with constant strides that are powers of 2, facilitating data reorganization. We demonstrate how our vectorization scheme applies to dominant SIMD architectures, and present experimental results on a wide range of key kernels, showing speedups in execution time up to 3.7 for interleaving levels (stride) as high as 8.

Why it matters

OpenAlex reports 127 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Most implementations of the Single Instruction Multiple Data (SIMD) model available today require that data elements be packed in vector registers. Operations on disjoint vector elements are not supported directly and require explicit data reorganization manipulations. Computations on non-contiguous and especially interleaved data appear in important applications, which can greatly benefit from SIMD instructions once the data is reorganized properly. Vectorizing such computations efficiently is therefore an ambitious challenge for both programmers and vectorizing compilers. We demonstrate an automatic compilation scheme that supports effective vectorization in the presence of interleaved data with constant strides that are powers of 2, facilitating data reorganization. We demonstrate how our vectorization scheme applies to dominant SIMD architectures, and present experimental results on a wide range of key kernels, showing speedups in execution time up to 3.7 for interleaving levels (stride) as high as 8.

Key concepts: SIMD, Vectorization (mathematics), Computer science, Parallel computing, Compiler, Interleaving, Computation, Scheme (mathematics)

Related papers

Back to paper searchBrowse research topicsOriginal source
Auto-vectorization of interleaved data for SIMD — Research Paper | ScholarLens