Last Friday, our central IT infrastructure, mostly running on Microsoft Windows, got infected by ransomware. All central systems were shutdown and a process started to reset them to the last (presumably) safe backup. Three days have passed since then and systems providing mail service (the communication service I rely on) are still not operational. It is amazing that it takes several days to restore regular service after such an attack.
The Network Management Research Group of the Internet Research Task Force got 20 years old, it was approved on March 14th, 1999. We had a short retrospective at the 104th meeting of the Internet Engineering Task Force in Prague today. It is interesting to reflect on which problems we made progress and which problems are still largely unsolved. We surely have much better network management technologies in place today compared to what we had 20 years back. If we, however, look at high level abstractions to express how a network should provide its services, we still seem to be searching for the holy grail.
The network management datastore architecture (NMDA) has been published about a year ago and this work has now been completed with the publication of the NMDA extensions for the NETCONF protocol and the NMDA extensions for the RESTCONF protocol. The work on the protocol extensions also required an update of the YANG Library. The definition of a common model describing how configuration data is conceptually processed to finally determine the behaviour of devices took a bit longer than expected but it feels good to have this piece of work finished (and we will soon forget about some of the roadblocks we had to deal with).
Below is a collection of ideas for student projects. Some are half-backed, some are not even written down. If you are interested in systems-oriented computer science, talk to me in person. In general I expect that students have a solid understanding of operating systems and computer networks and that they are able to handle programming tasks well.
This is a topic for someone interested in computer generated sounds. I am interested in algorithms that convert status information obtained by monitoring systems (say a monitor of computer network) into audible sounds that (i) are not intrusive but (ii) can signal significant changes in the conditions. There is work in this space. Some people recently wrote special programming languages that allow to describe sound generations as programs.
Good bye Wordpress. We have moved (back) to static web pages, thanks to the wonderful Hugo web site generator. The move to Hugo was a bit of an effort but now we have (almost) all content back in plain files, a big advantage if you ask me. And the best of all, Hugo allows us to use org-mode to edit our content very conveniently in emacs.
So what are the benefits from the user's perspective? The biggest advantage is good support for mobile devices via responsive themes. And of course faster page delivery - serving static files is as fast as it can possibly be. And serving static files is very robust against attacks since there is not much left to attack.
The definition of a network management datastore architecture (NMDA) has been published [RFC 8342]. We believe this is one of the more important RFCs we have worked on since it tells us how to think about configurations and their relationship to operational state. A few more RFCs will appear during the coming weeks providing the technology extensions that allow us to use the new framework in practice. Work on this document started with a trip to Stockholm in May 2016 but the discussions have a much longer history. We have run repeatedly into problems because the values of configuration objects may mean different things at different processing stages and we hope that the new network management datastore architecture allows us to be much clearer about the semantics of configuration objects in the future.
The EU funded LEONE project (2012-2015) contributed to the development of standards for large-scale measurements of Intern et performance, produced by the LMAP working group of the IETF. The key specifications, the LMAP information model [RFC8193] and the LMAP YANG data model [RFC8194], have been published as RFCs just a few days ago. It took roughly five years from the start of the formation of the working group until the publication of the core specifications. The first version of the information model was posted roughly 2.5 years before publication of the final document. This example shows that initiating and completing standardization work within the lifetime of EU projects is just not realistic.
Vaibhav Bajpai presented a paper summarizing his PhD thesis at the IFIP/IEEE International Symposium on Integrated Network Management (IM 2017) and he received the best dissertation paper award.
We helped to organize a Dagstuhl seminar on Using Networks to Teach About Networks, which took place March 12–15, 2017. During the seminar, we discussed the different approaches to teach computer networking and how to best use of online resources to better educate students. For details, see the seminar report.