基于相似性的粗粒度可重构指令压缩

李锦超; 赵仲元; 绳伟光

Chinese

您当前的位置：

首页 >

文章列表页 >

基于相似性的粗粒度可重构指令压缩

更新时间：2026-09-04

- 基于相似性的粗粒度可重构指令压缩
- Similarity aware context reduction for coarse grained reconfigurable architectures
- 微电子学与计算机 2020年37卷第8期页码：21-26
- 作者机构：
  
  上海交通大学电子信息与电气工程学院,上海,200240
- 作者简介：
- 基金信息：
  
  国家自然科学基金项目 61176037国家自然科学基金项目 61201059
- DOI：
  中图分类号：
- 纸质出版：2020
- 稿件说明：
移动端阅览
李锦超, 赵仲元, 绳伟光. 基于相似性的粗粒度可重构指令压缩[J]. 微电子学与计算机, 2020, 37(8): 21-26.

LI Jin-chao, ZHAO Zhong-yuan, SHENG Wei-guang. Similarity aware context reduction for coarse grained reconfigurable architectures[J]. 2020, 37(8): 21-26.
李锦超, 赵仲元, 绳伟光. 基于相似性的粗粒度可重构指令压缩[J]. 微电子学与计算机, 2020, 37(8): 21-26. DOI：

LI Jin-chao, ZHAO Zhong-yuan, SHENG Wei-guang. Similarity aware context reduction for coarse grained reconfigurable architectures[J]. 2020, 37(8): 21-26. DOI：

摘要

粗粒度可重构架构在能效比方面具有明显优势，然而其指令存储与传输过程的功耗代价过高.实验发现指令间具有明显的相似性，由此本文提出一种基于指令相似性的压缩技术，通过对指令的压缩、传输与解压，可以在不降低性能的前提下，优化架构的功耗和面积.针对同构和异构平台分别提出了指令分发模型和指令寄存器模型的解决方案，结合编译策略优化，最终与两种传统结构相比，面积效率比分别提升36%和181%，功耗效率比分别提升33%和118%.

Abstract

Coarse-Grained Reconfigurable Architectures (CGRAs) are promising computing platform with better efficiency. However

too much cost is wasted on the path of context storage and transmitting. With statistics of real case

there is obvious similarity between context. We propose one kind of context reduction method based on similarity of context. By making context compress

transmit

and decompress

this method can reduce power and area of CGRA obviously

with little overhead on performance. We propose context dispensing and register model for homogeneous heterogeneous platform. With the optimize of compiler

we get 33% and 118% promotion on power efficiency to tradition architecture

36% and 181% on area efficiency.

关键词

Keywords

references

KARUNARATNE M, TAN C, KULKARNI A, et al. Dnestmap: mapping deeply-nested loops on ultra-low power CGRAs[C]//2018 55th ACM/ESDA/IEEE Design Automation Conference (DAC). IEEE, 2018: 1-6.

KIM Y, MAHAPATRA R N. Dynamic Context Compression for Low-Power Coarse-Grained Reconfigurable Architecture[J]. IEEE Transactions on Very Large Scale Integration Systems, 2009, 18(1):15-28. DOI: 15

EGGER B, LEE H, KANG D, et al. A space- and energy-efficient code compression/decompression technique for coarse-grained reconfigurable architectures[C] // International Symposium on Code Generation % Optimization. IEEE Press, 2017.

LIU D, YIN S, LIU L, et al. Polyhedral model based mapping optimization of loop nests for CGRAs[C]//Proceedings of the 50th Annual Design Automation Conference. ACM, 2013: 19.

ASLAM, NAZISH, MILWARD, et al. Code Compression and Decompression for Coarse-Grain Reconfigurable Architectures[J]. IEEE Transactions on Very Large Scale Integration Systems, 16(12):1596-1608. DOI: 1596

KIM Y, KIEMB M, PARK C, et al. Resource sharing and pipelining in coarse-grained reconfigurable architecture for domain-specific optimization[C] // Proceedings of the conference on Design, Automation and Test in Europe-Volume 1. IEEE Computer Society, 2005: 12-17.

MEI B, VERNALDE S, VERKEST D, et al. ADRES: An architecture with tightly coupled VLIW processor and coarse-grained reconfigurable matrix[C] // International Conference on Field Programmable Logic and Applications. Springer, Berlin, Heidelberg, 2003: 61-70.

YIN W Z, ZHAO Z Y, MAO Z G, et al.A fast and efficient compiler frame work for CGRAs[J].Microelectronics % Computer, 2019, 36(8):45-48. DOI: 45

KIM Y, MAHAPATRA R N, PARK I, et al. Low Power Reconfiguration Technique for Coarse-Grained Reconfigurable Architecture[J]. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2009, 17(5):0-603. DOI: 0

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

超低功耗实时时钟电路的设计

基于监控指标关联性的微服务异常检测

基于MSS-YOLO-MPHands融合架构的装配手部动作识别研究

低功耗设计的双链电源关断策略的实现与验证

一种基于特征上下文理解与切片分析的肺结核检测算法