首页> 外文期刊>Future generation computer systems >Cloud forward: From distributed to complete computing
【24h】

Cloud forward: From distributed to complete computing

机译:云向前:从分布式到完整的计算

获取原文
获取原文并翻译 | 示例
       

摘要

Cloud computing matures and is increasingly being considered a baseline technology/enabler/utility [1], while being "extended" to address the requirements of emerging applications (e.g. mobility through fog computing), incorporate the advancements of other technologies (e.g. computations on the network through edge computing) or even become the enabler for service-orientation of different assets through advanced marketplaces [2]. The same applies for communications - enabling SDN/NFV 5G networking - and data-oriented technologies - enabling real-time big data analytics. Nonetheless, the advancements and the corresponding optimizations achieved in the aforementioned cases are (in their majority) "isolated" in each technological domain (i.e. computation, communications, data) and mainly utilize each other in a "black-box" way. While they all aim at serving the applications, the tools and frameworks in each domain exploit the information from the application space (i.e. following a top-down approach) to self-optimize the operation and provision of the corresponding underlying services without respecting the adaptations in or impact of the respective other domains. Representative examples are orchestration/choreography services in clouds, SDNs in networks and incremental queries in data management, which all regard the application as a behavioural model isolated within their domain - disrespecting thereby that e.g. changes on the network layer will affect the data availability and therefore expected locality. Moreover, the future application landscape is expected to be comprised of Internet of Things (IoT), Cyber-Physical Systems (CPS) as well as Factories of the Future (FoF) and mobile applications, thus posing even more requirements on computation, communication and data spaces: IoT and CPS are mainly characterised by complex dynamics and emerging behaviours in response to real-world events; FoF aim at digital automation incorporating unconventional technologies; while mobile applications provide media-rich interactive services. Are the highly-advanced but yet isolated computation, communication and data technologies able to address the requirements of the application plane in an efficient way?
机译:云计算已经日趋成熟,并被越来越多地视为基础技术/使能器/实用程序[1],同时被“扩展”以满足新兴应用程序的需求(例如通过雾计算的移动性),并结合了其他技术的进步(例如基于云计算的计算)。网络(通过边缘计算)),甚至通过高级市场成为不同资产面向服务的促成因素[2]。这同样适用于通信-支持SDN / NFV 5G网络-和面向数据的技术-支持实时大数据分析。但是,在上述情况下取得的进展和相应的优化(在大多数情况下)在每个技术领域(即计算,通信,数据)都是“隔离的”,并且主要以“黑匣子”的方式相互利用。尽管它们都旨在为应用程序提供服务,但是每个域中的工具和框架都利用了来自应用程序空间的信息(即遵循自上而下的方法),以自优化操作和提供相应基础服务,而无需考虑应用程序中的修改。或其他各个域的影响。代表性示例包括云中的编排/编排服务,网络中的SDN和数据管理中的增量查询,这些都将应用程序视为其域内孤立的行为模型-因此不尊重例如网络层的更改将影响数据可用性,并因此影响预期的位置。此外,预计未来的应用领域将由物联网(IoT),网络物理系统(CPS)以及未来工厂(FoF)和移动应用组成,因此对计算,通信和通信提出了更高的要求。数据空间:物联网和CPS的主要特点是复杂的动力学和响应现实事件的新出现的行为; FoF致力于结合非常规技术的数字自动化;而移动应用程序提供了媒体丰富的交互式服务。先进但隔离的计算,通信和数据技术是否能够以有效的方式满足应用程序平面的需求?

著录项

  • 来源
    《Future generation computer systems》 |2018年第1期|87-88|共2页
  • 作者单位

    Department of Digital Systems, University of Piraeus, 80, Karaoli & Dimitriou Str, 18534 Piraeus, Greece;

    euroCRIS, 51, Anna van Saksenlaan Str, 2593 HW Den Haag, The Netherlands;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号