FIND: Distributed Application Management Service (DAMS)

中枢神经系统0828923伯尔曼，肯尼斯 Cornell University康奈尔大学分布式应用程序管理服务（DAMS）自诞生以来，Internet一直是一个难以支持提供安全性、可靠性或其他信任属性的应用程序的环境。 然而，我们？我们看到了一波需要维护私有数据（例如，敏感的医疗记录）或扮演正确的容错行为至关重要的角色（例如，空中交通管制）的网络应用程序。 每个开发人员都被迫发明自己的解决方案来弥补现有互联网的局限性。 这项研究的中心前提是，这代表了一种危险的趋势，研究界必须尽一切努力创造新的和更好的选择。智力成果：康奈尔大学的研究人员将开发旨在支持互联网上一种新的值得信赖的分布式计算服务的技术。 该项目将通过展示以前被视为十几个独立的，通常定义不清的功能可以统一为一个概念上优雅的抽象（分布式角色委托），并通过开发所需的科学和数学来推理这种机制，从而产生基础性的智力影响。 该项目旨在帮助创建下一代关键任务网络应用程序？系统，可以委托与敏感的角色，因为他们体现了解决方案，显示有所需的属性更广泛的影响：该项目旨在创建一个有用的解决方案，开发人员今天缺乏通过创建一个强大的，可扩展的技术，将实现理论上最佳的性能和可扩展性，符合应用程序特定的一致性要求所施加的约束。

CNS ? 0828923Birman, Kenneth Cornell UniversityFIND: Distributed Application Management Service (DAMS)Since its inception, the Internet has been a difficult environment for supporting applications offering security, reliability, or other trust properties. Yet we?re seeing a wave of network applications that need to maintain private data (for example, sensitive medical records) or play roles in which correct, fault-tolerant behavior is critical (for example, air traffic control). Each developer has been forced to invent his or her own workarounds to compensate for the limitations of the existing Internet. The central premise of this research is that this represents a dangerous trend and that the research community must do what it can to create new and better options.Intellectual Merit: Cornell University researchers will develop technologies aimed at supporting a new kind of trustworthy distributed computing service on the Internet. The project will have a foundational intellectual impact by showing that what have previously been treated as a dozen separate, often ill-defined functions can be unified into a single conceptually elegant abstraction (distributed role delegation), and by developing the needed science and mathematics to reason about this mechanism. This project aims to help create the next generation of mission-critical networked applications ? systems that can be entrusted with sensitive roles, because they embody solutions shown to have the needed propertiesBroader Impact: This project seeks to create a useful solution that developers worldwide today lack by creating a powerful, extensible technology that will achieve theoretically optimal performance and scalability, consistent with the constraints imposed by application-specific consistency requirements.