We live in a world of networks, where users expect instant access to online services, and where our services need to rapidly respond to changes in demand. It’s a world driven by digital transformation, but one that’s held back by the brakes of technical debt.
What is technical debt? At heart it’s a simple concept, initially from the world of software development, a way of describing the work that needs to be done before a system can be upgraded or new features can be added. As a concept, it’s one that’s moved from code to architecture to the entirety of a company’s IT environment: software, hardware, and network.
In much of our infrastructure, it’s networks that suffer from technical debt. They’re background technology that support our services, and most of the time they just work. But now, a confluence of different demands on our networks mean that there are new requirements for how we build and run them; demands that require us to overcome technical debt before we can move on.
One of the biggest brakes on improving our network management is how we manage and deploy IP addresses. In a modern network, providing customer access to wireless networks and managing BYOD can mean allocating and managing thousands of IP addresses. A small company can have more than 1,000 addresses, with larger businesses having 100,000 or more. It’s not just user devices creating the need for more and more IP addresses- growing use of virtualized infrastructures and automated IT require tools to manage IP addresses as services scale.
The old ways of managing IP addresses aren’t suitable for this new world. Building your own solutions takes time and is expensive, and any resulting home-grown automation may not have all the features you need. As networks grow, the option of manual management quickly becomes untenable.
If you’re using a home-grown IP management solution, you’re likely to be aware of its issues. It’s hard to add features to it, and what was developed in the past might not be relevant today. For example, a system designed to manage internal IP addresses may not be able to scale to supporting customer IP addresses when offering free WiFi in a store. It’s also unlikely to be able to support any migration to IPv6.
Another key issue is handling documentation. If a home-built system doesn’t handle logging and documentation, it’s almost impossible for an IT department to keep track of which addresses are in use and what they’re being used for. Similarly, an internally-developed solution might also only handle IP address provisioning, and might not keep DNS and DHCP in sync. Then there’s the problem of knowledge: who knows how the system works, and what happens if they leave or are ill? That leads into the possibility of long, drawn-out recovery from any failures.
A problem of technical debt
If you want to move on from an existing implementation, you’re faced with a build versus buy decision. Do you build a new system from scratch, aware that you may be creating fresh technical debt, or do you purchase a new tool that offers upgrade paths and a support policy?
Things are harder for businesses that have chosen to use manual management. Complex environments, especially IT systems, evolve over time, and that means both DHCP assignments and fixed IP addresses need controlling and updating, even in the smallest networks. Manual configuration has a risk of errors, especially with documenting changes, leaving you in a situation where your documentation and the network don’t match.
One option is the adoption of an integrated DDI solution, bringing DHCP, DNS, and IPAM together in one service that’s able to take advantage of automation (either using its own tools, or integrating with a change management system and configuration management tools). Deploying a new server, or a VM image, becomes a matter of defining the server, and then letting your automation handle the rest – choosing the right IP address, configuring it in a DHCP system, and placing it in your DNS.
Frequent tasks require automation
A recent survey conducted by EfficientIP shows that implementing DDI saves money: for a customer with 60,000 IP addresses, over three years’ a saving of $412,650; for one with 8,800, $56,335. Both were using home-brewed IP address management solutions that were unable to support new business cases, and that were starting to affect their infrastructure.
Technical debt is expensive and a significant issue for end users. Buying a DDI solution means you can take advantage of dedicated developers and a dedicated solution that integrates with your existing automation platform – and is ready for any changes you make to your network and management tools.