Migrating your company’s infrastructure to the cloud is becoming a surefire way of getting the speed, capacity, and access your business, clients, and employees need to keep pace with the demands of the modern enterprise.
However, before you begin, there are certain considerations to take into account. Here are eight ways to kickstart your cloud migration:
The very first thing you should do is to consider why your company is moving its infrastructure to the cloud. This means asking the right questions in order to define the business objectives underpinning the proposed move. Is it to save money? To improve scalability, flexibility, or speed of deployment? Defining what you want out of moving to the cloud will help you to design what you need for the future.
Next, you need to determine the scope and gather all details relating to the scale of your proposed migration. This involves collecting information on what you currently have deployed – servers, applications, databases, storage, etc. including corresponding compute resources, performance requirements, network connections and ports, and application mapping.
Additionally, consider the types of data you’re moving and their classification. Are you moving secure data, health information, personal data? Or just public or generic data? Be aware of what you’re migrating in order to identify applicable security and compliance requirements.
Running an assessment tool will help you conduct some discovery of your environment. For Azure, we suggest using Azure Migrate to discover your workloads- virtual or physical servers – to recommend sizing and cloud compatibility. Use Data Migration Assistant to help with SQL database migrations to Azure by assessing cloud-readiness.
Consider networking connectivity and bandwidth requirements for your environment – taking a look at both internal and external connections. Do you need global or just regional network coverage? Do external parties need to connect to your environment, such as a client-based VPN, site-to-site tunnel, or web interface.
For internal connectivity, how are office locations going to connect? Do you have remote users? What is your network landscape? What are your bandwidth and latency requirements for anything running in the cloud? How will you optimize user experience in a cloud-first environment?
Establish your needs for real-time communication between employees. Are you rolling out UCaaS (Unified Communications as a Service)? How can you optimally peer with the UCaaS cloud?
If you can define those needs up front, then based on what your requirements are, you can, for example, determine if you need an ExpressRoute connection to Azure or determine your point-to-site VPN solution to access cloud resources.
The same enterprise security policies that you have for your on-premises environments need to be applied to cloud environments as well, because it’s really just an extension of your enterprise network.
It is crucial to understand what is needed so when you design what your cloud environment is going to look like, all of these requirements can be met. For example, this means understanding if your data-at-rest needs to be encrypted, which source IP addresses can access your application, or what next-generation firewall services should be implemented. Should your security be implemented with a cloud-based model?
It also means knowing how to manage access and data in the cloud environment; which role-based access control (RBAC) policies should be utilized, what data protection policies you need to comply with, and how long you’ll need to retain your data.
It’s important to understand performance, size, and availability requirements when moving to the cloud. They all impact the resourcing and sizing you’ll deploy when you migrate to the cloud. And because everything equates to a cost where the cloud’s pay-per-use model is concerned, getting requirements wrong can really have an impact– an expensive one!
In addition to cost impact, getting the size or SKU wrong can have a big impact on performance, accessibility, or availability. If your application running on IaaS virtual machines requires 99.9% availability or greater, it can impact the number of VMs required and/or the type of disks attached.
Without knowing the requirements upfront, the success of the migration will be impacted as well as have an impact on operations down the line.
Understand who should be involved in the project; who’s going to be actually doing the design, deployment and migration, as well as the end users, the business owners of the applications and resources that are going to move.
You need to have a full understanding of who’s involved to make it a success; know the key players in the entire project: from end users, to the actual people doing the deployment.
You also need a communication plan for end users. If you don’t communicate well, the entire process may look like a failure- even though it was technically successful- because there’s disruption to the way applications or services are used, such as a new URL, may be used to access a particular application.
Do this before you actually migrate any data or any workloads. Take into account everything else referred to above: your security policies, goals, availability and performance requirements, network connectivity, and inventory. If you don’t, you run the risk of running up your costs unexpectedly, thus negatively impacting end users, and overlooking major security holes.
The support of an Expert Managed Services Provider can give your business the advice and support it needs to move forward in a way that’ll work right now and in the future. Talk to our team today for your personalized cloud migration advice.