Posted on: 18th August 2026, by Magrathea
By Martyn Iles, Tech Director
As we mark 25 years of Magrathea, it feels appropriate to reflect not just on the services we provide, but on how the industry has evolved, as well as the role we have played -and continue to play – in shaping it.
Much of that work is invisible. It happens as networks are upgraded and improvements are designed, tested and implemented without anyone noticing (hopefully). It also happens in working groups, technical forums and long-running industry discussions where progress is often incremental rather than immediate.
I have been fortunate to be closely involved in much of this behind the scenes work for over two decades, developing Magrathea’s core systems while also contributing to key initiatives through NICC Standards Ltd (NICC).
Virtualisation
The network started with large, individual servers, often housed in industrially telecoms chassis that were custom built (and therefore hideously expensive). Having always enjoyed messing about with new technologies, I started playing with virtual PCs.
When I first started work, many years ago, we had big mainframes running lots of individual tasks, which quickly gave way to much smaller commodity servers and PCs each serving a single purpose. The ‘new’ server virtualisation idea was once again putting lots of different tasks into a bigger, highly redundant single server, something of a full-circle moment.
As a company, we took a risk on what was still an emerging technology. The promise was simple: deploy multiple virtual machines on a single host, which could – in theory – drastically reduce the time, effort and cost of deploying new servers.
Initially this was limited, but by 2010 live migration had arrived. Being able to move a running computer from one rack in our Telehouse suite to another without anything stopping working was amazing – even if the first time we tried it was at 3am, just in case.
Eventually this got past the ‘testing’ phase, and we moved to planning a proper deployment. At this point I found that our own ideas and requirements for resilience and redundancy were somewhat more stringent than many, even having to take the unusual step of going back to a vendor and saying they’ve under specified how much kit they need to sell us – not a conversation vendors have very often.
However, this stems from a long-held principle at Magrathea: we do not rely on supplier promises or ‘magic’. We always test systems thoroughly, often to destruction, invariably finding issues, edge cases and holes that others haven’t found. It’s also why our Development team know when I am presented with a system, I’ll probably find a way to break it within minutes due to ‘doing weird stuff no-one else would bother trying’.
Wake Me Up (before you go boom)
Another part of the network that I spent a lot of time working on in the early years was the monitoring and alerting of the core network, which has evolved significantly.
Initially, this was simply running scripts at predefined times which were then hooked into a rudimentary SMS gateway with zero intelligence that sent an SMS message to everyone, once a minute, 24/7 when issues were detected.
As a heavy sleeper, this was not especially popular when the clocks went forward, and the entire company (other than me) was woken up because I forgot to account for the hour which hadn’t existed in which no calls occurred!
The evolution from that to today’s monitoring platform is massive. We now have a fully automated assessment system, which determines the severity of issues and triggers the alerts to the correct team appropriately.
This all comes with huge amounts of stats, graphs and other metrics, because sometimes you know there is something wrong, but don’t know what. We always found that paging through graphs in this instance could often provide ‘oh, yeah’ moments and increasingly much of this analysis is now becoming more automated.
We need to have some Standards
I have also spent a huge amount of time working on developing standards and guidance across many of the NICC Technical Working Groups.
I first got involved with NICC as part of the Emergency Location Task Group (TG) EMLOC now Emergency Communications (EC), as it worked on trying to ensure that accurate location information is available when an emergency call is made. Knowing exactly where an IP connection originates can be surprisingly difficult because of the different technologies involved in the connection, as well as the privacy implications that people are now, quite rightly, much more aware of than they were fifteen years ago.
A lot of time was spent on the assessment of a very trendy new technology: Blockchain. This was seen as a possible solution to issues of number management and number portability. However, like many emerging technologies which generate a huge amount of interest (I suspect the early meetings were some of the best attended in NICC history) the realities of implementation meant that it wasn’t pursued.
The problems it was trying to solve didn’t disappear and so, somewhat reluctantly, I agreed to chair the NICC CDB TG which looked at how a Common Number Database could address many of the issues around number portability, routeing, CLI fraud and many other challenges. During this time, I gained a huge amount of respect for the other chairs who made it look effortless (it is not!).
I’ve also been involved and continue to contribute to the groups looking at CLI, Fraud and Security.
One positive to come out of the Covid pandemic was that the NICC, a body comprised entirely of modern, well-resourced Telecommunications companies, finally decided that meetings could be held using a new-fangled invention, that of IP Telecoms! This saved many of us hours of waiting on stations for trains to London that never arrived and finally made a two-hour meeting take two hours instead of an entire day.
Looking ahead
Whilst many of the problems I’ve spent many years working on aren’t solved, it feels like there has been progress in so many areas. There will always be a huge amount to work on and I feel quietly proud that the letters ‘CDB’ are often slipped into comments in chats and NICC meetings to this day. It’s a reminder that even the work happening quietly behind the scenes can leave a lasting mark.