[Podcast] IP networking in deep space
Marc Blanchet discusses modelling the delay of a deep space IP stack using Linux virtual network methods, and the suitability of QUIC as a transport for applications in space.
Category:
Tech matters
Exploring the blind spot of IXPs Route Servers
Guest Post: IRR-based filtering at IXPs often breaks the link between prefixes and their legitimate AS, allowing invalid announcements to slip through. This study measures the issue across many IXPs and offers actionable fixes.
Why are ‘we’ building all these data centres again?
An article in The Conversation argues that governments should ensure data centre developments serve the public interest.
NIST updates its secure DNS deployment guide: What operators need to know
NIST’s updated DNS deployment guide treats DNS as a core security control, offering practical guidance on protective DNS, encryption, DNSSEC, and both authoritative and recursive operations to help operators strengthen resilience, visibility, and policy enforcement.
CDNs, optical communications, and AI at NZNOG 2026
Geoff Huston attended NZNOG 2026, and shares some of what he found interesting.
Setting up Akvorado: A NetFlow analyser for your IPv6-first network
How to deploy Akvorado in a SOHO network to gain real-time visibility into traffic flows and improve IPv6 performance.
A day in the life of RIPE Atlas
On any given day, RIPE Atlas generates billions of measurement results, offering a comprehensive view of connectivity and reachability worldwide. Here, we focus on a single day to show how RIPE Atlas operates in practice.
Project IPv6-first: A case study in achieving an 80% native IPv6 SOHO network
Guest Post: How a modern SOHO network can achieve an 80% native IPv6 ratio by leveraging NetFlow insights to identify residual IPv4 paths and eliminate them.
[Podcast] What does ‘BCP’ really mean?
A conversation with Geoff Huston on what ‘best’ and ‘current’ mean in BCP. Is it a reflection of what is deployed today, or an expression of what the community aims to achieve?
The potential of erroneous outbound traffic
Guest Post: Network monitoring has traditionally been built around a simple assumption: Most problems come from outside. What can we learn by observing the traffic that leaves our networks — especially when it does not receive a reply?