Results 11 to 20 of about 10,499 (305)

Research on Function Vectorization Technology Based on Directive Statements [PDF]

Jisuanji kexue
With the continuous development of processor technology,SIMD(Single Instruction Multiple Data) vectorization has been widely applied in various fields.However,previous research has mainly focused on loops and basic blocks,while full-function ...
LIU Lili, SHAN Zheng, LI Yingying, WU Wenhao, LIU Wenbo
doaj   +2 more sources

A Graph-Based Learning Framework for Compiler Loop Auto-Vectorization

Intelligent Computing
The single instruction multiple data (SIMD) capability in modern processors is critical to improving the performance of current compute-intensive programs.
Yao Xiao, Nesreen K. Ahmed, Mihai Capotă, Guixiang Ma, Theodore L. Willke, Shahin Nazarian, Paul Bogdan   +6 more
doaj   +2 more sources

Vectorization Optimization of LDS Memory Access for DCU [PDF]

Jisuanji gongcheng
In a domestic general-purpose accelerator Deep Computing Unit(DCU), Local Data Shared(LDS) is a key storage component with a lower latency and higher bandwidth than global memory.
Sichi YANG, Rongcai ZHAO, Lin HAN, Hongsheng WANG
doaj   +2 more sources

Vectorization-aware loop unrolling with seed forwarding [PDF]

, 2019
Loop unrolling is a widely adopted loop transformation, commonly used for enabling subsequent optimizations. Straight-line-code vectorization (SLP) is an optimization that benefits from unrolling. SLP converts isomorphic instruction sequences into vector
Petoumenos, P, Leather, H, Wang, Zheng, Hugh Leather, Petoumenos, Pavlos; id_orcid, Cole, Murray, Murray Cole, Rocha, RCO, Leather, Hugh, Rocha, Rodrigo Rocha, Vasileios Porpodas, Zheng Wang, Wang, Z, Rocha, Rodrigo C. O., Luís F. W. Góes, Goes, Luis F.W., Góes, LFW, Cole, Murray; id_orcid, Pavlos Petoumenos, Porpodas, Vasileios, Porpodas, V, Petoumenos, Pavlos, Góes, Luís F. W., Rodrigo C. O. Rocha, Cole, M   +24 more
core   +3 more sources

Vectorization with Haswell and CilkPlus [PDF]

, 2013
Project Specification: This project concerns the parallel computing and vectorization field for Physics Computing at CERN. The document summarises the results and experience from vectorization activities and an initial evaluation of the CilkPlus ...
Juan José, Fumero Alfonso, Andrzej, Nowak   +1 more
core   +1 more source

Integer‐Grid Sketch Simplification and Vectorization [PDF]

, 2020
International audienceA major challenge in line drawing vectorization is segmenting the input bitmap into separate curves. This segmentation is especially problematic for rough sketches, where curves are depicted using multiple overdrawn strokes ...
Stanko, Tibor, Tibor Stanko, Adrien Bousseau, Bommes, David, Mikhail Bessmeltsev, Bessmeltsev, Mikhail, David Bommes, Bousseau, Adrien   +7 more
core   +1 more source

General Vectorization of Line Objects in Drawn Maps [PDF]

, 2009
Drawn maps consist of multiple object types. The most important are line objects which represent infrastructure. Attributes of these objects are essential for many tasks but in raster format they provide only low level information.
Peter Tarabek
core   +1 more source

Speculative dynamic vectorization to assist static vectorization in a HW/SW co-designed environment [PDF]

, 2013
Compiler based static vectorization is used widely to extract data level parallelism from computation intensive applications. Static vectorization is very effective in vectorizing traditional array based applications.
Rakesh Kumar, Antonio Gonzalez, Gonzalez, A., Kumar, R., Martinez, A., Alejandro Martinez   +5 more
core   +1 more source

Urban Scene Vectorized Modeling Based on Contour Deformation

ISPRS International Journal of Geo-Information, 2020
Modeling urban scenes automatically is an important problem for both GIS and nonGIS specialists with applications like urban planning, autonomous driving, and virtual reality.
Lingjie Zhu, Shuhan Shen, Xiang Gao, Zhanyi Hu   +3 more
doaj   +1 more source

Vectorization of Program Code Containing Low Probability Regions in Computational Geometry Problems

Современные информационные технологии и IT-образование, 2022
Improving application performance is an important practical task for supercomputer calculations. Along with parallelization of calculations between cluster nodes (for example, using MPI tools), as well as multithreaded programming (for example, using ...
Alexey Rybakov
doaj   +1 more source