Edora Cloud, 1 month later
A little more than one month ago we launched Edora Cloud. A Danish, sovereign, public cloud. Our approach is unconventional compared to other cloud providers in that we publish our infrastructure orchestration software as open source, and offer the posibility of running this on-prem. Compared to the very few and scattered, previous Nordic attempts at building public cloud offerings, we have purposely chosen not to rely on traditional enterprise tech, but are instead relying on our own orchestration stack, built around open source building blocks like Linux and KVM. We offer both a full-featured IaaS offering for running virtual servers, as well as various services built on top op this, including a Kubernetes service.
The launch
The launch attracted some attention, with the Danish Minister of Digitization attending and performing the official opening of the cloud. But how has the market reception since been? There has been a bunch of sign-ups from people just looking around and starting a few servers, using a few services. But for now, we only accept sign-ups from Danish clients, which of course limits the addressable market. So far – no sign-ups from large, Danish outfits looking to run large volumes of servers or other infrastructure. This is in line with expectations. We do not expect large private or public clients to just sign up to a new public cloud, and start running loads of VMs or provision large Kubernetes clusters. We are very much aware that this will take time, and we are in this for the long haul. On the other hand, general interest from the Danish public sector has been intense. This is most likely because, when it comes to the public sector, we have a very strong value proposition, which for now just keeps getting stronger every day with Trump’s posturing and Putin’s disruptions: We can actually offer a solution that is safe from both.
What’s ahead?
Even thousands of self-service clients who sign up and run a few servers are simply not enough to sustain a viable public cloud offering. Also, it is not why we launched Edora Cloud. Developers, small clients and community-building is vital to the long-term health of our offering. But, our overarching vision is to move Danish and European IT infrastructure away from the enterprise tech stack (hardware and software), that have defined (and shackled) Danish and European IT for decades, into an open source and cloud native tech stack. To move the market in a meaningful way in this direction, we believe we must win large public IT tenders. Price, and the promise of an escape route from the enterprise incumbents, are probably our most effective tools here. Since launching Edora Cloud we have seen unusually many large, public tenders in the IT operations space, all requiring that stuff NOT run in foreign clouds, and all mentioning Kubernetes as their infrastructure platform of choice going forward. In other words, large public clients are now pushing for their infrastructure to not be run by big tech, but by local outfits, and for their tech stack to be based on open source. For a guy like me, who has followed this space for many years, this is highly unusual (to put it mildly). To be fair, this is probably not some collective epiphany, but driven by the current Trump/Putin squeeze (i.e. fear), but we’ll take it. We are currently involved in 4 large tenders, all involving critical Danish IT infrastructure. We hope to win at least one. Regardless we will be investing heavily in Edora Cloud in 2025. A public cloud should be highly automated on all levels, and hardware will be a big part of 2025. We already have automated handling of regular compute and storage capacity in Edora Cloud, but we are itching to add new capabilities like GPUs and ARM. So stay tuned!
