SBIR Phase I: Stateless - Taking from the Lab to a Production Environment

此小型企业创新研究(SBIR)第一阶段项目的更广泛影响/商业潜力是为数据中心网络中的网络基础设施带来敏捷性。虽然现代数据中心利用服务器虚拟化快速、轻松地部署应用程序，以及抽象化硬盘驱动器并使应用程序能够随时从任何服务器访问任何数据的存储系统，但保护网络并提高网络效率的网络基础设施仍主要是硬件设备。根据思科的说法，未来几年云流量预计将翻两番，设备数量快速增长(随着物联网的出现)，安全威胁的复杂性和规模不断增加，该项目将为数据中心运营商提供快速轻松地部署基础设施的灵活性，以跟上这些趋势和威胁。这一小型企业创新研究(SBIR)第一阶段项目将为科罗拉多大学博尔德分校发明的一种新的网络功能虚拟化方法提供技术验证，在该方法中，网络功能被分解为单独的处理和状态存储组件。实验室中的实验已显示出无缝的可扩展性、无中断的故障管理和与其他软件解决方案一致的处理速度。该项目的目标是在生产环境(具有真实流量的数据中心)中运行，以充分量化面对生产条件时的吞吐量、延迟、可扩展性、弹性和成本效益。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is to bring agility to the network infrastructure in data center networks. While modern data centers leverage server virtualization to quickly and easily deploy applications, and storage systems that abstract away the hard drives and enable applications to access any data from any server at any time, the network infrastructure that protects and makes the network more efficient is still largely hardware appliances. With cloud traffic expected to quadruple over the next several years (according to Cisco), a rapid growth in the number of devices (with the Internet of Things emerging), and an ever increasing complexity and scale of security threats, this project will provide data center operators with the agility to quickly and easily deploy infrastructure to keep up with these trends and threats. In doing so, society will experience more reliable networked applications with reduced data breaches.This Small Business Innovation Research (SBIR) Phase I project will provide technical validation for a new approach to network function virtualization, invented at the University of Colorado, Boulder, in which network functions are disaggregated into separate processing and state storage components. Experiments in the lab have shown seamless scalability, disruptionless failure management, and processing rates in line with other software solutions. The goal of this project is to run in a production environment (a data center with real traffic) to fully quantify the throughput, latency, scalability, resilience, and cost effectiveness when faced with production conditions.