How to get cloud migration right
If you want to get cloud migration right, you must deal with an inconvenient truth: Cloud or hybrid cloud environments lower the drawbridge between your data center and the internet, and that creates opportunity as well as security risk.
Many enterprises that migrate to hybrid or multi-cloud environments gain a network that is more resilient, automated, cost-effective, and secure. But there’s also plenty to lose: Unsuccessful migrations reduce network control and visibility, exacerbate routing issues, hamper digital transformation, and jeopardize security.
Want to improve your chances of a successful migration? Don’t downplay the importance of network architecture.
Many enterprises overlook the essential infrastructure and tools (DDI services) that enable network communications. DDI includes Domain Name System (DNS), Dynamic Host Configuration Protocol (DHCP), and IP address management (IPAM). DDI services are essential to routing and establishing network connections — on-premises or in the cloud. Cloud migration cannot succeed without reevaluating the role of DDI services in your enterprise.
Analysis is the bedrock
A successful migration — like a house renovation — begins with an analysis of your current environment. Knowing how DNS/DHCP functions in your environment, as well as identifying adjacent technologies and integrations, security posture, and business processes is a necessary step. It won’t prevent all surprises during migration, but it can help.
Paint me a portrait of your ideal network
Next, outline and explore the challenges related to your current network architecture. Stakeholders should arrive with a vision of their ideal infrastructure. What things do they not want to see in their new network? What do they want to prevent, improve, and optimize — and how do they expect the cloud to help?
Resilience drives many enterprises to cloud migration. This might occur after crippling outages that disrupt user experiences and business operations. But the hunt for efficiency and new IT initiatives that can reduce service level agreements are also factors.
Inclusion matters for successful migrations
There’s another often-ignored factor that can derail cloud migrations: not including the right stakeholders. In an on-premises environment, the main stakeholders were the data center or network team. Successful cloud migrations demand inclusion. Other business units, application owners, dev ops, compliance and risk teams must align and get involved. Do it right and each stakeholder gets what they need. Do it wrong and you have a migration that stalls or becomes never ending.
Adapting to the cloud
All this preparation and analysis is done to avoid one worst-case scenario: replicating your existing environment in the cloud. Ditch the notion of a lift-and-shift approach. It won’t deliver.
Why is that? Because some cloud services also require cloud native DNS such as AWS Route 53 or Azure DNS. Deploying DNS servers in cloud environments won’t help. DNS works differently in the cloud than on-premises. There are multiple regions, and each VPC/VNET in that region has its own DNS. Sometimes those domains overlap, which leads to more complexity. A complexity that increases when you add private links and endpoints.
The ideal approach would be to use cloud native DNS for cloud services and DNS servers deployed to the cloud for serving branches and data centers. But the end goal is a unified resolution layer for the whole environment. That’s how you create a single source of truth and manage the network with full visibility.
So how do you get there?
Here’s where you need creative minds or — as we call them — solutions architects. They engineer a bridge from your present to your future. It starts with high-level architecture. This blueprint outlines what infrastructure should be moved into the cloud, what stays on-premises, and how everything will work together (while keeping DNS and DHCP in mind). Then it’s time to collect feedback from stakeholders, adjustments, and the creation of a low-level architecture that’s ready for inspection.
Planning for data migration
There are a lot of migration risks. That’s why collecting and consolidating data — everything from IP models to DNS Domains and Records, DHCP Scopes and Reservations and much more — is so important. But so is scrubbing the data so that it’s optimized for use in the new environment.
Many migrations tackle the backend first. A better approach is to start by taking over client-facing DNS within the existing infrastructure. This makes it easier to switch from the legacy environment, minimizing disruption for applications and users.
The new environment must be monitored to locate devices or users stuck on the legacy network. Things may have been missed and it’s crucial to have support to troubleshoot, increase your knowledge of DDI, and enable your network to support the business.