• Home
  • Search
  • A DCU code generation and optimization method based on polyhedral model
  • Cite Icon2
  • https://doi.org/10.1117/12.2678907Copy DOI Icon

A DCU code generation and optimization method based on polyhedral model

  • May 25, 2023
  • Jingyu Niu +4 more
Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

The domestic processor DCU(Deep Computing Unit) is a GPU-like accelerator. To address the problem that DCU lacks more effective compilation and optimization tools, a DCU code generation and optimization method based on polyhedral model is proposed. First, a source-to-source compiler C2HIPC is implemented based on the polyhedral model. The input of C2HIPC is a serial C program, and the output is a HIP (Heterogeneous-compute Interface for Portability) executable on the DCU; secondly, to further improve the heterogeneous performance of the program, a kernel partitioning method based on computing unit occupancy is proposed, which completes the kernel partitioning by limiting loop merging in the polyhedral schedule calculation; at the same time, a global memory access optimization method based on data reuse is proposed, By modeling the two kinds of data reuse and global memory access costs existing in the DCU kernel, the global memory access costs in different loop orders are calculated. Using the PolyBench test set to verify the effectiveness of C2HIPC, the performance of the automatically generated HIP code after C2HIPC optimization reaches 92.4% of that of PolyBench-GPU.

Similar Papers
  • Conference Article

Research on of Heterogeneous Parallel Algorithm for High-Dimensional FFT on GPGPU

  • Sep 19, 2025
  • Ruihan Sun +3
  • Research Article
  • Citations5

Low occupancy high performance elemental products in assembly free FEM on GPU

  • Mar 22, 2021
  • Engineering with Computers
  • Nileshchandra K Pikle +2
  • Conference Article
  • Citations8

State-Driven Energy Optimization in Wireless Sensor Networks

  • Aug 14, 2005
  • R Jurdak +2
  • Research Article

Accelerating content-based image-patch retrieval (CBIR) algorithms using HPC architectures and code optimization methods

  • Jun 25, 2016
  • Figshare
  • Re Silvina
  • Research Article
  • Citations17

The Polyhedral Model of Nonlinear Loops

  • Dec 08, 2015
  • ACM Transactions on Architecture and Code Optimization
  • Aravind Sukumaran-Rajam +1
  • Research Article
  • Citations25

Tology-Aware Optimal Data Placement Algorithm for Network Traffic Optimization

  • Aug 01, 2016
  • IEEE Transactions on Computers
  • Wuhui Chen +2
  • Conference Article
  • Citations15

HPP: A High Performance PRAM

  • Aug 26, 1996
  • Publications of the UdS (Saarland University)
  • Arno Formella +2
  • Conference Article

FractalCloud: A Fractal-Inspired Architecture for Efficient Large-Scale Point Cloud Processing

  • Jan 31, 2026
  • Yuzhe Fu +8
  • Conference Article
  • Citations23

PRO: Progress Aware GPU Warp Scheduling Algorithm

  • May 01, 2015
  • Jayvant Anantpur +1
  • PDF
  • Research Article
  • Citations1

Directional Analytic Discrete Cosine Frames

  • Jan 01, 2022
  • IEEE Access
  • Seisuke Kyochi +2
  • Research Article
  • Citations7

Global optimality conditions and optimization methods for quadratic integer programming problems

  • Jan 25, 2011
  • Journal of Global Optimization
  • Z Y Wu +2
  • Conference Article
  • Citations11

Instruction Selection and Scheduling for DSP Kernels on Custom Architectures

  • Sep 01, 2013
  • Mehmet Ali Arslan +1
  • Book Chapter
  • Citations1

Automatic Partitioner for Behavior Level Distributed Logic Simulation

  • Jan 01, 2005
  • Kai-Hui Chang +5
  • Research Article
  • Citations293

Using profile information to assist classic code optimizations

  • Dec 01, 1991
  • Software: Practice and Experience
  • Pohua P Chang +2
  • Research Article
  • Citations44

A tool path generation strategy for three-axis ball-end milling of free-form surfaces

  • Jan 20, 2008
  • Journal of Materials Processing Technology
  • Tao Chen +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.