A Multi-Objective Optimization Mapping Algorithm for Coarse Grained Reconfigurable Architectures

CHEN Nai-jin; JIANG Jian-hui

doi:10.3969/j.issn.0372-2112.2015.11.003

您当前的位置：

首页 >

文章列表页 >

A Multi-Objective Optimization Mapping Algorithm for Coarse Grained Reconfigurable Architectures

更新时间：2025-07-16

- A Multi-Objective Optimization Mapping Algorithm for Coarse Grained Reconfigurable Architectures
- Acta Electronica Sinica Vol. 43, Issue 11, Pages: 2151-2160(2015)
- 作者机构：
  
  1. 安徽工程大学计算机与信息学院,安徽,芜湖,241000
  2. 同济大学软件学院,上海,201804
  3. 天津大学计算机科学与技术学院,天津,300072
  4. 安徽工程大学计算机与信息学院,安徽,芜湖,241000
  5. 同济大学软件学院,上海,201804
  6. 天津大学计算机科学与技术学院,天津,300072
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.0372-2112.2015.11.003
  CLC： TP302
- Published：2015
- 稿件说明：
移动端阅览
CHEN Nai-jin, JIANG Jian-hui. A Multi-Objective Optimization Mapping Algorithm for Coarse Grained Reconfigurable Architectures[J]. Acta Electronica Sinica, 2015, 43(11): 2151-2160.
DOI：

CHEN Nai-jin, JIANG Jian-hui. A Multi-Objective Optimization Mapping Algorithm for Coarse Grained Reconfigurable Architectures[J]. Acta Electronica Sinica, 2015, 43(11): 2151-2160. DOI： 10.3969/j.issn.0372-2112.2015.11.003.

摘要

针对多约束下的行流水粗粒度可重构体系结构的硬件任务划分映射问题

提出了一种多目标优化映射算法.该算法根据运算节点执行时延、依赖度等因素构造了累加概率权值函数

在满足可重构单元面积和互连等约束下

通过该函数值动态调整就绪节点的映射调度次序

当一块可重构单元阵列当前行映射完毕后

就自动换行

当一块阵列被填满

就切换到下一块

当一个数据流图映射完毕后

就自动计算划分块数等参数.实验结果表明

与层贪婪映射算法相比

文中算法平均执行总周期降低了8.4%(RCA

)和5.3%(RCA

)

与分裂压缩内核映射算法相比

文中算法平均执行总周期降低了20.6%(RCA

)和21.0%(RCA

)

从而验证了文中提出算法的有效性.

Abstract

Based on row pipelining coarse grained reconfigurable architecture(CGRA)

we presented MOM(multi-objective optimization mapping) algorithm to solve the multi-constraints hardware task partitioning-mapping problem.The cumulative probability weight function was constructed by the execution delay of computing nodes and the dependence between two nodes

etc.With the constraints of reconfigurable cell area and interconnection

the proposed algorithm could adjust dynamically the scheduling order of the ready nodes by thefunction values.When a row of the RCA was mapped completely

MOM began on a new row.When the RCA was filled

MOM switched t

o the next one.When a DFG(data flow graph) was mapped completely

the number of modules and etc were calculated automatically in MOM.Experiment results show that the average execution total cycles of MOM decrease by 8.4%(RCA

) and 5.3%(RCA

) comparing with LBGM(level based greedy mapping) algorithm.Comparing with SPKM(split-push kernel mapping) algorithm

the average execution total cycles of MOM decrease by 20.6%(RCA

) and 21%(RCA

).Experimental evaluation confirms the efficiency of our approach.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Modeling and Verification of Workflow Based on Resource Constraint

Related Author

PANG Shan-chen

Related Institution

山东科技大学信息科学与工程学院

清华大学计算机科学系

⁰