Consulting

Network consultancy

Before our company founder set up Opteamax GmbH, he worked for a major German hosting provider, where his responsibilities included, amongst other things, the overall network planning and the operation of the network.

In the field of routing, our focus is on Juniper, but we have always also looked at products from other manufacturers.

We can now look back on many years of experience and a wide range of projects in the field of network planning and operations. Our remit includes connections to Internet exchanges – including peering evaluation and negotiations – as well as the planning of data centre and campus networks.

We support our customers every step of the way, from the initial idea through to requirements analysis, hardware selection and procurement, right through to long-term operation. Thanks to our strong partnerships with various distributors and hardware manufacturers, we are also able to present our customers with cost-effective options.

Infrastructure as Code and Automation

A network that has grown over the years is rarely documented in terms of how it actually operates. Configurations are maintained manually, changes can only be understood by a select few, and no one can say with any certainty whether two sites with identical hardware are actually configured in the same way.

We move networks like that onto an Infrastructure-as-Code footing: the configuration becomes versioned source, every change is traceable, reviewable and repeatable — and a new site is built from the same template as the existing one.

Our automation is based on Ansible, with NetBox as the system of record for inventory and address data. That combination lets roll-outs, changes and recurring work run reliably — on Juniper equipment as on other vendors’.

How we migrate

A migration on a production network has to happen without a maintenance window the customer would notice. So we work in stages:

• Taking stock: We extract the actual configurations and compare them with the documentation. What this reveals – special cases that have developed over time, orphaned rules, discrepancies between supposedly identical devices – is often the most valuable part of the project.
• Mapping: The existing configuration is transferred into roles and templates, initially in such a way that it exactly replicates the current setup.
• Dry-run comparison: Nothing is rolled out until the generated configuration matches the live configuration bit for bit. Only when the comparison shows no further differences is the mapping complete.
• Handover: From this point on, every change is handled via automation. Manual access remains possible, but becomes the exception.
• Building out: Only then come the things that were impractical before — automated configuration checks, reproducible test environments, new sites from the template.

The benefits do not become apparent on the day of the changeover, but afterwards: changes can be made more quickly and with less risk; knowledge is no longer confined to individual minds; and the question ‘Why is this configured this way?’ can be answered by consulting the version history.