An Out-of-Order Load-Store Queue for Spatial Computing

Lana Josipović, P. Brisk, P. Ienne
{"title":"An Out-of-Order Load-Store Queue for Spatial Computing","authors":"Lana Josipović, P. Brisk, P. Ienne","doi":"10.1145/3126525","DOIUrl":null,"url":null,"abstract":"The efficiency of spatial computing depends onthe ability to achieve maximal parallelism. This needs memoryinterfaces that can correctly handle memory accesses arrivingin arbitrary order while still respecting data dependencies andensuring appropriate ordering for semantic correctness. However, a typical memory interface for out-of-order processors (i.e., aload-store queue) cannot immediately fulfill these requirements:a different allocation policy is needed to achieve out-of-orderexecution in a spatial system. We show a practical way toorganize the allocation for an out-of-order load-store queue forspatial computing by dynamically allocating groups of memoryaccesses, where the access order within the group is staticallypredetermined (for instance by a high-level synthesis tool).","PeriodicalId":124631,"journal":{"name":"2017 IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"31","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3126525","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 31

Abstract

The efficiency of spatial computing depends onthe ability to achieve maximal parallelism. This needs memoryinterfaces that can correctly handle memory accesses arrivingin arbitrary order while still respecting data dependencies andensuring appropriate ordering for semantic correctness. However, a typical memory interface for out-of-order processors (i.e., aload-store queue) cannot immediately fulfill these requirements:a different allocation policy is needed to achieve out-of-orderexecution in a spatial system. We show a practical way toorganize the allocation for an out-of-order load-store queue forspatial computing by dynamically allocating groups of memoryaccesses, where the access order within the group is staticallypredetermined (for instance by a high-level synthesis tool).
空间计算中的乱序负载存储队列
空间计算的效率取决于实现最大并行性的能力。这需要内存接口能够正确处理以任意顺序到达的内存访问,同时仍然尊重数据依赖性并确保语义正确性的适当顺序。然而,乱序处理器(即负载存储队列)的典型内存接口不能立即满足这些需求:需要不同的分配策略来实现空间系统中的乱序执行。我们展示了一种实用的方法,通过动态分配内存访问组来组织无序负载存储队列的分配,其中组内的访问顺序是静态预定的(例如通过高级合成工具)。
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:604180095
Book学术官方微信