CAREER: Enabling Futuristic Distributed Applications With Integrative Multistream Networking

实现信息基础设施的全部潜力需要构建分布式应用程序，在这些应用程序中，这些无数设备的功能被一起利用。今天的互联网应用程序连接设备（例如，连接我的浏览器和CNN服务器）。未来的互联网应用将连接环境（例如，提案者的办公室与CNN新闻编辑室）。高性能的未来分布式应用程序将使用大量的媒体流和其他类型的信息（例如，3D几何图形、控制信息、结构化XML文档等）。这些应用程序所带来的网络挑战与传统的客户机/服务器应用程序有着根本的不同。它们的特点是：多个语义相关的信息流。分布式应用程序架构。复杂的适应行为。可扩展性和可组合性。大带宽延迟产品。未来的分布式应用程序很难构建，因为当前的网络基础设施不支持其特征的多流架构。这样做很重要，因为当环境中的所有设备都有助于丰富和沉浸式体验时，这些类型的应用程序（例如，远程学习，远程医疗等）最有效。提出的研究解决了未来分布式多流应用所带来的网络挑战。研究将集中在管理应用程序中多个语义相关的数据流时出现的问题。利用流之间的高级语义关系对于实现应用程序级目标和适当分配有限的网络资源非常重要。我们解决这个问题的方法包括：构建由聚合应用程序行为和相关对等流状态通知的协议。尝试跨网络堆栈级别集成的网络机制。专门针对应用程序跨多个设备实现的分布式体系结构。启用分布式应用程序驱动的适应，利用流之间的高级语义关系。开发与底层协议动态更一致的数据类型（如视频、音频、几何等）表示。演示多流的有效性（即，在可能采用不同协议的多个流中传输复杂的媒体类型）。

Realizing the full potential of the information infrastructure requires building distributed applications in which the capabilities of these myriad devices are harnessed together. The Internet applications of today connect devices (i.e., connecting my browser with the CNN server). The Internet applications of the future will connect environments (i.e., proposer's office with the CNN newsroom). High-performance futuristic distributed applications will employ scores of media streams in conjunction with other types of information (e.g., 3D geometry, control information, structured XML documents, etc.). The networking challenges presented by these applications are fundamentally different than traditional client/server applications. They are characterized by: Multiple, semantically related flows of information.Distributed application architectures.Complex adaptive behaviors. Extensibility and composability. Large bandwidth-delay products. Futuristic distributed applications are difficult to build because current networking infrastructures do not support their characteristic multistream architecture. Doing so is important because these types of applications (e.g., distance learning, tele-medicine, etc.) are most effective when all of the devices in an environment contribute to a rich and immersive experience. The proposed research addresses the networking challenges presented by futuristic distributed multiflow applications. The research will focus on the problems that arise when managing multiple semantically related flows of data within an application. Exploiting high-level semantic relationships between flows is important for achieving application-level goals and appropriately allocating limited network resources. Components of our approach to this problem include:Building protocols informed by both aggregate application behavior and the state of related peer flows.Experimenting with network mechanisms integrated across levels of the networking stack. Specifically targeting distributed architectures in which applications are implemented across multiple devices. Enabling distributed application-driven adaptation which exploits high-level semantic relationships between flows. Developing data type (i.e. video, audio, geometry, etc.) representations that are more congruent with underlying protocol dynamics. Demonstrating the effectiveness of multistreaming (i.e., transmitting complex media types in multiple flows possibly employing different protocols).