4. Conclusion
单词学习
翻译
This paper has offered an intellectual history of programmable networks. The idea of a programmable network initially took shape as active networking, which espoused many of the same visions as SDN, but lacked both a clear use case and an incremental deployment path. After the era of active networking research projects, the pendulum swung from vision to pragmatism, in the form of separating the data and control plane to make the network easier to manage. This work focused primarily on better ways to route network traffic—a much narrower vision than previous work on active networking
本文介绍了可编程网络的发展历史——可编程网络的概念最初形成为主动网络，支持许多与SDN相同的愿景，但缺乏明确的用例和增量部署路径。在主动网络研究项目时代之后，钟摆从视觉转向实用主义，以数据和控制平面分离的形式，使网络更易于管理，这项工作主要集中在路由网络流量的更好的方法上，这比以前关于主动网络的工作要细得多。
Ultimately, the work on OpenFlow and network operating systems struck the right balance between vision and pragmatism. This work advocated network-wide control for a wide range of applications, yet relied only on the existing capabilities of switch chipsets. Backwards compatibility with existing switch hardware appealed to many equipment vendors clamoring to compete in the growing market in data-center networks. The balance of a broad, clear vision with a pragmatic strategy for widespread adoption gained traction when SDN found a compelling use case in network virtualization.
最终，关于OpenFlow和网络操作系统的工作在愿景和实用主义之间取得了正确的平衡，这项工作提倡对广泛的应用进行网络范围的控制，但仅依赖于交换芯片组的现有能力。现有交换机硬件的向后兼容性吸引了许多设备供应商，他们要求在不断增长的数据中心网络市场上展开竞争，当SDN在网络虚拟化中发现了一个引人注目的用例时，广泛、清晰的愿景与广泛采用的实用策略之间的平衡得到了重视
As SDN continues to develop, we believe that history has important lessons to tell. First, SDN technologies will live or die based on “use pulls”. Although SDN is often heralded as the solution to all networking problems, it is worth remembering that SDN is just a tool for solving network-management problems more easily. SDN merely places the power in our hands to develop new applications and solutions to longstanding problems. In this respect, our work is just beginning. If the past is any indication, the development of these new technologies will require innovation on multiple timescales, from long-term bold visions (such as active networking) to nearterm creative problem solving (such as the operationally focused work on separating the control and data planes)
随着可持续发展网络的不断发展，我们相信历史有重要的教训要讲。首先，SDN技术将在“使用拉动”的基础上生存或消亡。尽管SDN通常被认为是所有网络问题的解决方案，但值得记住的是，SDN只是一个更容易解决网络管理问题的工具，SDN仅仅是把开发新应用程序和解决长期问题的能力交给我们，在这方面，我们的工作才刚刚开始。如果过去有任何迹象表明，这些新技术的发展将需要在多个时间尺度上进行创新，从长期大胆的愿景（如积极的网络）到近期创造性的问题解决（如专注于分离控制和数据平面的工作）
Second, we caution that the balance between vision and pragmatism remains tenuous. The bold vision of SDN advocates a wide variety of control applications; yet, OpenFlow’s control over the data plane is confined to primitive matchaction operations on packet-header fields. We should remember that the initial design of OpenFlow was driven by the desire for rapid adoption, not first principles. Supporting a wide range of network services would require much more sophisticated ways to analyze and manipulate traffic (e.g., deep-packet inspection, and compression, encryption, and transcoding of packets), using commodity servers (e.g., x86 machines) or programmable hardware (e.g., FPGAs, network processors, and GPUs), or both. Interestingly, the renewed interest in more sophisticated data-plane functionality, such as Network Functions Virtualization, harkens back to the earlier work on active networking, bringing our story full circle
第二，远见和实用主义之间的平衡仍然脆弱。SDN的大胆设想提倡各种各样的控制应用程序，然而OpenFlow对数据平面的控制仅限于包头字段上的原始匹配操作。我们应该记住，OpenFlow的最初设计是由快速采用的愿望驱动的，而不是最初的原则。支持广泛的网络服务需要更复杂的方法来分析和操作流量（例如，深度数据包检查，数据包的压缩、加密和转码），使用商品服务器（例如，x86机器）或可编程硬件（例如，FPGAS，网络处理器和/或GPU）。趣的是，对更复杂的数据平面功能（如网络功能虚拟化）的重新兴趣，让哈肯回到了先前关于主动网络的工作中，使我们的故事更为全面。
Maintaining SDN’s bold vision requires us to continue thinking “out of the box” about the best ways to program the network, without being constrained by the limitations of current technologies. Rather than simply designing SDN applications with the current OpenFlow protocols in mind, we should think about what kind of control we want to have over the data plane, and balance that vision with a pragmatic strategy for deployment
保持SDN的大胆设想需要我们继续“开箱即用”地思考网络编程的最佳方法，而不受当前技术的限制。与其简单地按照当前的OpenFlow协议设计SDN应用程序，我们应该考虑我们希望对数据平面进行什么样的控制，并用实际的部署策略来平衡这一愿景。