How we route-leaked everything to everyone due to a fun Arista bug.
A post-mortem kind of story about us changing Arista RCF function names, resulting in a global route leak. The talk includes the history of why we intended to make a change, how we rollout configurations, why it resulted in an unexpected behavior and how it was fixed.
PeeringDB is a freely available, user-maintained, database of networks, and the go-to location for interconnection data. The database facilitates the global interconnection of networks at Internet Exchange Points (IXPs), data centers, and other interconnection facilities, and is the first stop in making interconnection decisions.
We give regular updates at Internet related events to make...
AI represents one of the biggest inflection points in the modern world of technology. Hyperscalers and other companies see a once-in-a-lifetime opportunity to leverage AI to enhance existing business models and create new innovations. Today the only way we know of to enhance AI performance is to use more and more compute power to train Generative AI models. This in turn seems to require...
According to a number of cloud vendor usage studies, cross-availability zone (AZ) data transfers regularly account for at least 25% of the public cloud users’ production cost. Cutting down these costs can affect your bottom line and your application affordability. All major cloud vendors provide daily aggregated cost metrics for the cross-zone network traffic, however oftentimes, these reports...
A long format talk on handling extremes of weather when your network is mostly outdoors, covering the basics of wind, snow, heat and cold, and then focusing on the events of October 23 and Storm Babet and how we managed to keep our network running with our office and primary POP under 4ft of floodwater, and the lessons we learned form it
Draft Powerpoint presentation attached
This presentation investigates the proximity to a low Signal-to-Noise Ratio (SNR) threshold that can still maintain a tolerable Bit Error Rate (BER) in 100G / 400G / 800G network links. Additionally, we account for factors such as temperature and cable length to predict the duration for which a reliable network connection can be sustained between transceivers. The analysis, based on data...
This presentation explores the impact of recent submarine cable outages in the Baltic Sea and how the Internet adapted to maintain connectivity. Using RIPE Atlas measurements, we analyze latency shifts, packet loss, and rerouting patterns following incidents affecting the BSC East-West and C-LION1 cables.
I would like to give a very short, approximately 2 minute, lightning talk to let the NetUK audience know of a relevant event coming to the United Kingdom next year. I was recently successful in my proposal for one of the upcoming RIPE Meetings to come to the United Kingdom with the UK being selected over other countries in a competitive process. RIPE92 in May 2026 is coming to Edinburgh,...
Every click, swipe and scroll leaves an impact on the environment. As we rely more on the Internet in many sectors and for many applications, the carbon footprint of the internet is rising, and it is unclear if we can achieve the net zero goals by 2050. Introducing carbon awareness to computer networks is one promising solution, yet with many challenges. The work presented in this talk is in...
BCS, the Chartered Institute of IT talking about the use of career frameworks for self development
In response to increasing regulatory pressures, major telecommunications providers have begun to measure and report their carbon footprints. However, this initial step is just the beginning of a complex journey toward achieving sustainability. One of the most significant challenges these companies face is addressing Scope 3 emissions, which are generated by their supply chains and lie outside...
Update on the state of NetUK
How peeringdb is localised in LACNIC region. Going through why localisation is important and data about how much the LAC community is using peeringdb and how much peeringdb related activities are in the region.
ISPs may notice that traffic from certain sources is entering their network at an unexpected location, but it is hard to know if this represents a problem or is just normal spoofed background noise. If such traffic is not spoofed, it would be useful to generate alerts, but alerting on background noise is not useful.
We describe Penny, a robust, reliable, and practical traffic checker that...
