High‐performance extended actors

Peter A. Buhr, Colby A. Parsons, Thierry Delisle, He Nan Li
{"title":"High‐performance extended actors","authors":"Peter A. Buhr, Colby A. Parsons, Thierry Delisle, He Nan Li","doi":"10.1002/spe.3262","DOIUrl":null,"url":null,"abstract":"Abstract Actors are a popular mechanism for indirectly expressing concurrency. This article examines an implementation in the concurrent dialect of C ++, C ++, which runs actors on shared‐memory multi‐processor computers. The C ++ actor system targets 32–256+ multi‐core shared‐memory computers that form the backbone of high‐performance computing, rather than distributed actor communication or robust execution via parentage fallback used by other actor systems. Five new mechanisms are presented to achieve expressibility, robustness, high performance, and scalability of actor applications across multiple cores: explicit life time (storage management) of actors and messages, combining actors and coroutines, a forward message‐trace and backward message‐return for debugging and failures, a new promise call‐back for ask sends, and an actor implementation that inverts the actor execution‐model by decoupling actor mailboxes with high levels of sharding. Microbenchmarks compare the new actor features with CAF, Protoactor, and classic and typed Akka.","PeriodicalId":21899,"journal":{"name":"Software: Practice and Experience","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Software: Practice and Experience","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1002/spe.3262","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Abstract Actors are a popular mechanism for indirectly expressing concurrency. This article examines an implementation in the concurrent dialect of C ++, C ++, which runs actors on shared‐memory multi‐processor computers. The C ++ actor system targets 32–256+ multi‐core shared‐memory computers that form the backbone of high‐performance computing, rather than distributed actor communication or robust execution via parentage fallback used by other actor systems. Five new mechanisms are presented to achieve expressibility, robustness, high performance, and scalability of actor applications across multiple cores: explicit life time (storage management) of actors and messages, combining actors and coroutines, a forward message‐trace and backward message‐return for debugging and failures, a new promise call‐back for ask sends, and an actor implementation that inverts the actor execution‐model by decoupling actor mailboxes with high levels of sharding. Microbenchmarks compare the new actor features with CAF, Protoactor, and classic and typed Akka.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
高性能扩展演员
actor是一种流行的间接表达并发的机制。本文研究了c++并发方言的一个实现,它在共享内存多处理器计算机上运行actor。c++ actor系统的目标是32-256 +多核共享内存计算机,这些计算机构成了高性能计算的骨干,而不是其他actor系统使用的分布式actor通信或通过父级回退实现的健壮执行。提出了五种新的机制来实现跨多个核心的参与者应用程序的可表达性、健壮性、高性能和可扩展性:参与者和消息的显式生命周期(存储管理),将参与者和协程结合起来,用于调试和失败的前向消息跟踪和后向消息返回,用于请求发送的新承诺回调,以及通过将参与者邮箱与高级别分片解耦来反转参与者执行模型的参与者实现。微基准测试将新的actor特性与CAF、Protoactor以及经典和类型化的Akka进行比较。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Algorithms for generating small random samples A comprehensive survey of UPPAAL‐assisted formal modeling and verification Large scale system design aided by modelling and DES simulation: A Petri net approach Empowering software startups with agile methods and practices: A design science research Space‐efficient data structures for the inference of subsumption and disjointness relations
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1