A Workshop on Extensible Distributed Systems

可扩展分布式系统研讨会

• 批准号: 1450502
• 负责人: Robbert VanRenesse
• 金额: $ 6.08万
• 依托单位: Cornell University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1450502&HistoricalAwards=false
• 关键词: Workshop; Extensible; Distributed; Systems

A Distributed System is a system consisting of multiple computers communicating through message passing. In the past 50 years, distributed systems have evolved from being a novelty to a fact of life---a very large fraction of computers today are part of a distributed system. With computers, sensors, and actuators becoming embedded in virtually every device, users are themselves becoming, as it were, embedded in a ubiquitous computing fabric. An Extensible Distributed System is a distributed system that is pervasively integrated into people's environments. The distributed systems research community has developed models and techniques that have supported this remarkable success story and enabled technologies, such as cloud computing, that have the potential to fundamentally change the way in which individuals, businesses, and government access computing resources. The pervasive or ubiquitous systems community in the meantime has leveraged distributed systems and networking technology, as well as artificial intelligence, to integrate devices even closer into our experience. And in the post Dennard-Scaling era, computer architecture is undergoing a profound transformation, and, in an effort to address the conflicting goals of delivering performance and saving energy, it is exploring designs that are exporting weaker abstractions to software developers.There are, however, gaps between the foundations built by the various research communities and the opportunities offered by today's and tomorrow's applications and infrastructure. To respond to these challenges, the field not only needs to profoundly retool today's techniques and abstractions, but also to develop a new vision, designed to address the opportunities offered by new hardware, new networking, and new applications. This project supports a workshop that will serve as a catalyst for triggering the process of reflection and innovation necessary to develop this new vision and the research agenda that will bring it about.Extensible Distributed Systems are the cornerstone of, to name a few, cloud computing, sensor/actuator networks, and industrial control systems, and their impact and ubiquity is likely to only increase in the next decade. The in-depth discussions about the state of field and research agenda that this workshop aims to develop are thus likely to have tremendous impact on society. In addition to helping define a research agenda for the extensible distributed systems community, this workshop will help crystallize the shape of future graduate and undergraduate education in distributed computing. The workshop will balance academic and industrial participation, will engage leaders in the field, and will actively seek participation from underrepresented groups.

分布式系统是由多台计算机通过消息传递进行通信组成的系统。在过去的50年里，分布式系统已经从一个新奇事物发展成为现实——今天很大一部分计算机都是分布式系统的一部分。随着计算机、传感器和执行器几乎嵌入到每个设备中，用户本身也被嵌入到无处不在的计算结构中。可扩展分布式系统是一种广泛集成到人们环境中的分布式系统。分布式系统研究社区已经开发出了支持这一非凡成功故事的模型和技术，并启用了云计算等技术，这些技术有可能从根本上改变个人、企业和政府访问计算资源的方式。与此同时，无处不在的系统社区利用分布式系统和网络技术，以及人工智能，将设备更紧密地集成到我们的体验中。在后Dennard-Scaling时代，计算机体系结构正在经历一场深刻的变革，并且，为了解决交付性能和节省能源的冲突目标，它正在探索向软件开发人员输出较弱抽象的设计。然而，在各种研究团体建立的基础与今天和明天的应用和基础设施提供的机会之间存在差距。为了应对这些挑战，该领域不仅需要深刻地重组当今的技术和抽象，而且还需要开发一个新的愿景，旨在解决新硬件、新网络和新应用程序提供的机会。该项目支持一个研讨会，该研讨会将作为催化剂，触发发展这一新愿景和实现这一愿景的研究议程所必需的反思和创新过程。可扩展分布式系统是云计算、传感器/执行器网络和工业控制系统的基石，它们的影响和无处不在可能只会在未来十年中增加。因此，本次研讨会旨在开展的关于领域现状和研究议程的深入讨论可能会对社会产生巨大影响。除了帮助定义可扩展分布式系统社区的研究议程之外，本次研讨会还将帮助明确分布式计算领域未来研究生和本科生教育的形态。讲习班将平衡学术界和工业界的参与，将吸引该领域的领导人，并将积极寻求代表性不足的群体的参与。