Your infrastructure, our operations. Set up and handed over, or maintained for you under a monthly service level agreement.
Scientific and business computing has a failure mode that is easy to recognise and expensive to live with: a system that works, but only one person understands it. When that person leaves or gets busy, progress stops. We build the opposite: infrastructure that is documented, reproducible and operable by your own team.
Two ways to work with us
Set up and hand over
We design the architecture, provision it as code, harden it, verify the backups actually restore, document the result, and train your team to run it. Then it is yours. Tiers are sized by how many environments and accounts you need, whether existing workloads have to be migrated, and how much compliance surface is involved, not by hours booked.
At the larger end this covers multi-account landing zones, high-performance and research computing, hybrid setups, and architectures that have to produce an audit trail.
Support and maintain
Managed operations under Care: monitoring, patching, backups, and incident response to an agreed response time. Service levels run from business-hours support on a single environment set, through multi-environment cover with quarterly disaster-recovery testing, to 24/7 on-call for priority incidents with monthly reporting.
Where it runs, and who owns it
We work on AWS, Google Cloud, Azure and Hetzner (EU). For organizations that need European data residency, Hetzner is the option we lead with, and the data processing agreement and subprocessor list are on the table in the first meeting rather than after signature.
The cloud account stays yours. Cloud invoices run on your own account and we add no resale margin on them: your infrastructure, our operations. That keeps the bill transparent, and it means making your setup more efficient is not something that costs us money.
On compliance claims
We document readiness, not certification. Where a tier includes compliance work, what you receive is the architecture, the logging and the evidence pack that support an audit. Novodan does not certify you as compliant, and you should be careful of any supplier who says they can.
Who this is for
Research institutions running shared compute for several groups. Companies whose workload has outgrown the machine under someone’s desk. Organizations with a residency or audit requirement they cannot currently evidence. Teams that inherited an environment nobody documented and need it made safe to change.
