2018 · 22 citations · 19 references
ComputingComputer EngineeringComputer ScienceCompilersComputer ArchitectureParallel ComputingCode GenerationCompiler ConstructionData ParallelismHardware AccelerationInstruction Set ArchitectureVector DesignSimd ArchitecturesThread ParallelismThread-based ParallelismThread-level ParallelismVector InstructionsVector ProcessingVector Processing UnitsVectorization
Data-level parallelism is frequently ignored or underutilized. Achieved through vector/SIMD capabilities, it can provide substantial performance improvements on top of widely used techniques such as thread-level parallelism. However, manual vectorization is a tedious and costly process that needs to be repeated for each specific instruction set or register size. In addition, automatic compiler vectorization is susceptible to code complexity, and usually limited due to data and control dependencies. To address some these issues, Arm recently released a new vector ISA, the Scalable Vector Extension (SVE), which is Vector-Length Agnostic (VLA). VLA enables the generation of binary files that run regardless of the physical vector register length.
19
Nathan Binkert, Bradford M. Beckmann, Gabriel Black et al. · ACM SIGARCH Computer Architecture News · 2011 · 4.9K citations
Richard M. Russell · Communications of the ACM · 1978 · 697 citations · Full text
Yuan Tang, Rezaul Chowdhury, Bradley C. Kuszmaul et al. · 2011 · 338 citations · Full text