Ant Colony Optimization based Module Footprint Selection and Placement for Lowering Power in Large FPGA Designs

被引:0
|
作者
Herath, Kalindu [1 ]
Prakash, Alok [1 ]
Jiang Guiyuan [1 ]
Srikanthan, Thambipillai [1 ]
机构
[1] Nanyang Technol Univ, Singapore, Singapore
来源
2018 INTERNATIONAL CONFERENCE ON RECONFIGURABLE COMPUTING AND FPGAS (RECONFIG) | 2018年
基金
新加坡国家研究基金会;
关键词
fpga; low-power; large designs; placement; modular design methodology; communication-aware;
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Technology scaling has led FPGA manufactures to produce FPGAs with millions of logic cells. It has motivated hardware designers to map increasingly large applications into FPGAs. However, state-of-the-art FPGA CAD tools consume longer runtime to compile such applications affecting the design productivity. Modular design methodology, which allows to partition a large application into smaller modules, compile and verify individually and subsequently combine all modules to get the final design, has been introduced in the past to improve the design productivity. To reduce the dominant power dissipation component in FPGAs, the routing power, existing workflows have considered data communication between design modules during module formation, as well as during placing them on FPGA space in the final solution. But, these workflows have not discussed a methodology to select the best shape and the region (footprint) on FPGA space for each module. Selecting proper footprints, on the other hand, is becoming a critical process on modern FPGAs because of their heterogeneity in resources and column arrangement. In this work we propose a communication-aware module placement strategy which selects the best footprint for each module. Our methodology has shown about 16% routing power reduction with respect to commercial CAD flow and about 11% with respect to a communication-aware design partitioning workflow that does not consider placement of modules on FPGA.
引用
收藏
页数:8
相关论文
共 6 条
  • [1] Communication-aware Module Placement for Power Reduction in Large FPGA Designs
    Herath, Kalindu
    Prakash, Alok
    Kanewala, Udaree
    Srikanthan, Thambipillai
    2018 IEEE COMPUTER SOCIETY ANNUAL SYMPOSIUM ON VLSI (ISVLSI), 2018, : 209 - 214
  • [2] FPGA implementation of population-based ant colony optimization
    Scheuermann, B
    So, K
    Guntsch, M
    Middendorf, M
    Diessel, O
    ElGindy, H
    Schmeck, H
    APPLIED SOFT COMPUTING, 2004, 4 (03) : 303 - 322
  • [3] FPGA-Based Path Planning Using Improved Ant Colony Optimization Algorithm
    Hsu, Chen-Chien
    Hou, Ru-Yu
    Kao, Wen-Chung
    Li, Shih-An
    2015 IEEE 5TH INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS - BERLIN (ICCE-BERLIN), 2015, : 443 - 444
  • [4] ANT3D: Simultaneous Partitioning and Placement for 3-D FPGAs based on Ant Colony Optimization
    Danassis, Panayiotis
    Siozios, Kostas
    Soudris, Dimitrios
    IEEE EMBEDDED SYSTEMS LETTERS, 2016, 8 (02) : 41 - 44
  • [5] Imputation of Discrete and Continuous Missing Values in Large Datasets Using Bayesian Based Ant Colony Optimization
    Priya, R. Devi
    Sivaraj, R.
    ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, 2016, 41 (12) : 4981 - 4993
  • [6] FPGA Implementation of an Ant Colony Optimization Based SVM Algorithm for State of Charge Estimation in Li-Ion Batteries
    Stighezza, Mattia
    Bianchi, Valentina
    De Munari, Ilaria
    ENERGIES, 2021, 14 (21)