APCer: An Agile Physical Compiler for Multi-Port Register File

This paper proposes a novel Agile Physical Compiler (APCer) for multi-port register files based on a standard cell library. It employs a Performance, Power and Area (PPA)-driven genetic algorithm to facilitate agile iterations and design generation. APCer can automatically iterate, generate, and optimize floorplan, rapidly creating efficient register file circuits and layouts with specified capacity and port numbers. Additionally, APCer responds to user-defined PPA requirements, enabling targeted design optimizations. It accelerates the register file design duration by over 40× and 20× compared with the full custom approach and existing compilers, respectively. Compared to existing compilers, APCer improves performance by 20.7% and reduces power consumption by 33.9%. Compared to the full custom approach, APCer improves performance by 31.5% and reduces power consumption by 28.8%.

Authors

Institutions

Publication Details

Published
2025-05-25
DOI
https://doi.org/10.1109/iscas56072.2025.11043756
Primary Topic
Parallel Computing and Optimization Techniques
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

APCer: An Agile Physical Compiler for Multi-Port Register File

Biwei Liu, Yaohua Xu, Yi Wang, Na Bai et al.
Parallel Computing and Optimization Techniques
article

APCer: An Agile Physical Compiler for Multi-Port Register File

Biwei Liu, Yaohua Xu, Yi Wang, Na Bai, Tianbo Ming
article en

Abstract

This paper proposes a novel Agile Physical Compiler (APCer) for multi-port register files based on a standard cell library. It employs a Performance, Power and Area (PPA)-driven genetic algorithm to facilitate agile iterations and design generation. APCer can automatically iterate, generate, and optimize floorplan, rapidly creating efficient register file circuits and layouts with specified capacity and port numbers. Additionally, APCer responds to user-defined PPA requirements, enabling targeted design optimizations. It accelerates the register file design duration by over 40× and 20× compared with the full custom approach and existing compilers, respectively. Compared to existing compilers, APCer improves performance by 20.7% and reduces power consumption by 33.9%. Compared to the full custom approach, APCer improves performance by 31.5% and reduces power consumption by 28.8%.

Anhui University (CN), National University of Defense Technology (CN)
Natural Science Foundation of Anhui Province
Openalex Percentile: Top 84%
Parallel Computing and Optimization Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.