Parametric architecture for implementing multimedia algorithms

N. Kavvadias, S. Nikolaidis
{"title":"Parametric architecture for implementing multimedia algorithms","authors":"N. Kavvadias, S. Nikolaidis","doi":"10.1109/ICDSP.2002.1028322","DOIUrl":null,"url":null,"abstract":"Multimedia applications are characterized by high computational demands related to data transfer and storage operations. Multimedia algorithms in their majority consist of regular repetitive loop constructs. In this paper, a novel control architecture for implementing such loop intensive algorithms is described. The proposed control unit takes advantage of the regularity of computations in order to serve as high performance parametric controller of multimedia datapaths. The control unit cooperates with datapath modules and their corresponding controlling FSM. Algorithmic flow dependencies which determine the appropriate loop sequencing are mapped on a LUT. For another algorithm to execute, LUT context and FSM configurations only have to be reprogrammed. Thus, partial reconfiguration possibilities for implementing multimedia algorithms on programmable platforms can be exploited. For demonstration purposes, a matrix multiply algorithm implementation case is investigated. Compared to a software realization on ARM7 processor, significant performance improvements are reported.","PeriodicalId":88900,"journal":{"name":"International Conference on Digital Signal Processing proceedings : DSP. International Conference on Digital Signal Processing","volume":"108 1","pages":"1261-1264"},"PeriodicalIF":0.0000,"publicationDate":"2002-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Digital Signal Processing proceedings : DSP. International Conference on Digital Signal Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDSP.2002.1028322","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

Multimedia applications are characterized by high computational demands related to data transfer and storage operations. Multimedia algorithms in their majority consist of regular repetitive loop constructs. In this paper, a novel control architecture for implementing such loop intensive algorithms is described. The proposed control unit takes advantage of the regularity of computations in order to serve as high performance parametric controller of multimedia datapaths. The control unit cooperates with datapath modules and their corresponding controlling FSM. Algorithmic flow dependencies which determine the appropriate loop sequencing are mapped on a LUT. For another algorithm to execute, LUT context and FSM configurations only have to be reprogrammed. Thus, partial reconfiguration possibilities for implementing multimedia algorithms on programmable platforms can be exploited. For demonstration purposes, a matrix multiply algorithm implementation case is investigated. Compared to a software realization on ARM7 processor, significant performance improvements are reported.
实现多媒体算法的参数化架构
多媒体应用的特点是与数据传输和存储操作相关的高计算需求。大多数多媒体算法由规则的重复循环结构组成。本文描述了一种实现这种循环密集算法的新型控制体系结构。该控制单元利用计算的规律性作为多媒体数据路径的高性能参数控制器。控制单元与数据路径模块及其对应的控制FSM协同工作。确定适当循环排序的算法流依赖关系映射到LUT上。要执行另一种算法,只需重新编程LUT上下文和FSM配置。因此,可以利用在可编程平台上实现多媒体算法的部分重新配置可能性。为了演示目的,研究了一个矩阵乘法算法的实现案例。与ARM7处理器上的软件实现相比,报告了显着的性能改进。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信