Participants will work in small teams during the hackathon. In order to help you come up with or select a challenge/project/task to work on, we're suggesting the following themes: Sustainability, Security & Privacy, and Sovereignty.
You can also suggest your own project when applying or specify that you wish to work on one of the project ideas below. And please specify if you need any specific resources, such as cloud computing, data repositories etc. Access to RIPE Atlas will be provided and SIDN have .nl DNS traffic data set for you to look at also.
Sustainability
The sustainability theme focuses on the long-term ecological, economic and social impact of an activity. For this DNS Hackathon, we're looking specifically at the operation of a DNS service and how it relates to sustainability, and in particular to the environmental impact, for example: direct or indirect carbon emissions, energy consumption, etc.
How green can you make the DNS? Is it even possible to calculate the emission of a DNS query?
Project ideas:
- Compare carbon footprint of various SaaS DNS (software-as-a-service, a.k.a. "managed DNS") providers based on what can be learned about both their edge and core computing.
- Figure out potential cost/savings from moving to DNS-over-QUIC based on the extra computing required, and also the smaller need for overcapacity.
- Redesign the DNS packet format and build some benchmarks comparing performance versus existing format.
Security & Privacy
There is a long history of how DNS can be (mis)used to target victims, of which cache poisoning is one of the most well-known. DNSSEC is designed to counter this attack vector, but adoption is still relatively slow and the classic vulnerability is still exploited in attacks. With privacy enhancements to the DNS, more encryption is used in DNS transactions between client and server, making eavesdropping impossible (or more difficult). Alternatively, DNS data can be used for detection and incident analysis such as providing insight into sources and patterns of, for example, botnets used for DDoS attacks.
The security and privacy theme can examine: what the impact of these recent developments on DNS and DNSSEC security has been on robustness and the resilience against (unintentionally) participating in DDoS attacks; what new information models and analytical methods can be used to provide this kind of insight into security incidents or ongoing attacks.
Project ideas:
Sovereignty
Increasingly emerging in the context of geopolitical governance, the term sovereignty is the effort by a governing entity, such as a state, to set boundaries on a network and then exert some form of control, often in the form of law enforcement across such boundaries. Infrastructure is becoming more centralised, for convenience (outsourcing of non-core activities) and for economies of scale.
The sovereignty theme can examine whose infrastructure you rely on, under what jurisdiction a service is offered and, in the event of international tensions, whether the service could be interrupted and what the impact of that would be.
In the EU region, we should think of national and EU/regional sovereignty and their interdependence. For privacy reasons, this has also become an important topic as EU law and protection of the privacy of EU citizens requires strict monitoring of what data is collected by major technology companies and whether their (non-EU) companies comply with the GDPR.
Project ideas:
- Review of dns0.eu traffic versus their claims (https://www.dns0.eu/privacy), and how this compares with Quad9 privacy claims/guarantees (https://quad9.net/service/privacy).
- Domain name centralisation using open data such as https://data.openintel.nl/data/ and others.
- Analyse open source vulnerability history, the implications of the EU Cyber Resilience Act and the Software Bill of Materials, and propose some mechanisms that would improve vulnerability patching.
Bring your own idea!
Are you interested in other topics related to the DNS? Feel free to bring your ideas, concepts, or prototypes to the event as well!
Project ideas:
- Incremental (or complete... hey why not?) work towards resolverless DNS.
- Compare returning GeoIP-based NS vs. anycast solutions, possibly with modelling/simulation.
- Reproducing (or not) of already published DNS research.
