IETF 92 was a unique experience compared to the Association for Computing Machinery (ACM) and USENIX conferences I regularly attend. Naturally, IETF is more focused on concrete solutions and detailed specifications for working systems, as opposed to conceptual research. This practical focus appealed to my interest in “systems building” research.
By attending working and research group sessions closely related to my research–in particular, SFC, NFVRG, and SDNRG–I gained a better understanding of: (1) what problems are currently in need of solutions, (2) what problems will need to be solved in the near future, and (3) what constraints shape the space of possible solutions. For example, the SFC session had a presentation on dealing with legacy network functions, which is a problem I have attempted to address in some of my past research. This presentation affirmed the relevancy of this problem. Moreover, discussions during this session made me realize that the solution I had originally proposed–re-purposing some field in the Ethernet or IP header to serve as a tag–is not well suited for an actual deployment. It seems that making it easier to retrofit legacy functions with support for new SFC standards–e.g., using program analysis techniques–may be a more viable approach.
In the NFVRG session, there were several presentations on open source virtual network function management and orchestration (MANO) frameworks. These frameworks address some of the practical issues I have encountered in my research–e.g., high speed forwarding of packets to network function virtual machines. I plan to use some of these frameworks to conduct more realistic evaluations of the solutions/systems I have developed.
One of my favorite sessions was the plenary presentation on security in the Internet-of-Things. There has been little discussion on this topic at the networking conferences I regularly attend, so this presentation provided a great introduction to this emerging area.
In summary, attending IETF has given me new research problems to think about, and helped me identify better ways to evaluate my research. It’s also improved my teaching: I am now better equipped to teach students about Internet standards and the Internet-of-Things.