Understanding Your Current Infrastructure and Migration Approaches is Critical for Cloud Success
“The Cloud” – a phrase that still conjures thoughts of digital transformation, business acceleration, and promise for the future.
You’ve done your homework, and have decided that you want in – it’s time for YOUR cloud migration.
You understand the impact that “going cloud” will have on your business; you’ve addressed how you will overcome common security and vendor lock-in concerns; and you have a solid understanding of SaaS, PaaS, and IaaS cloud models and how they can benefit your business (if you haven’t done these things, check out this article).
So you think you’re ready for the cloud?
Think again. At this point, many organizations make the mistake of beginning their migration. But without an inventory of hardware, software, and networking infrastructure, coupled with an understanding the three approaches to workload migration, it’s unlikely that you will be able to take full advantage of all the cloud has to offer.
It is also probable that your migration will cause unnecessary downtime, consuming more resources than initially intended. What follows is intended to help you take the final steps to ensure cloud migration success.
Get a complete inventory of your hardware, software, and networking infrastructure
Approaching your cloud migration without a clear picture of the current state of your hardware, software, and networking infrastructure is like driving for hundreds of miles without a map: while you might hit your destination eventually, but you’re going to spend a lot of extra time – and a lot more money on gas – getting there.
Taking a hardware and software inventory
The main goal of taking stock of your hardware and software inventory is to ensure you understand what relies on what. This is critical for determining how you are going to migrate processes and workloads to the cloud (more on that later). A full hardware and software inventory accounts for all servers (RAM, DISK, CPU, etc.), storage, and security appliances, as well as operating systems, models, versions and releases.
Taking a network inventory
Taking a network inventory is about so much more than just your Internet connection. A proper inventory of your network includes:
- Network capacity (WAN and Internet) by location
- Technology in use such as MPLS, Ethernet, and “IP”
- Appliances, including routers, switches, and firewalls both physical and virtual, including their model and throughput capabilities
- In addition to inventory, organizations should create a topology map including IP address ranges showing WAN and Internet uplinks, as well as internal data center connectivity with bandwidth and throughput.
Understanding your network inventory is critical for a couple of reasons. First, you need to ensure that your chosen CSP can meet the network requirements of your workloads. This will help you determine which applications are the most bandwidth-intensive and may need to remain on-premises. Additionally, understanding bandwidth and throughput is necessary for correctly timing migrations. Whether migrating a workload takes a day or a week has a significant effect on your business operations, and this is largely determined by bandwidth and throughput.
Rehost, replatform, or refactor: how are you going to migrate your applications?
It’s all too common to find organizations that believe they can just “lift-and-shift” their existing workloads onto VMs in the cloud. While this is certainly possible – and even preferable – in certain situations, often it is the case that more work needs to be performed before an application can live comfortably in the cloud.
But before we determine how to assess our applications for migration requirements, let’s look at our options:
- Rehost: Otherwise known as “lift-and-shift”, rehosting involves migrating applications to the cloud without any code modification. This approach is quicker and requires less up-front resources than the next two processes we’ll discuss. However, rehosting fails to take advantage of many of the benefits of the cloud, such as elasticity. Additionally, while it may be cheaper than on-premises solutions, rehosting is usually more expensive than other approaches that optimize for the cloud.
- Replatform: Replatforming involves small upgrades to workloads for the purpose of taking better advantage of the cloud than would be the case with a rehosting approach. This might mean adding automation-enabled autoscaling, or perhaps a managed database offering. Replatforming is the happy middle ground of cloud migration, offering the benefits of cloud functionality and cost optimization without the heavy resource commitment of our next migration method.
- Refactor: The most involved approach of all, refactoring involves recoding and rearchitecting workloads to take advantage of cloud-native functionality. It is by far the most time- and resource-intensive solution, but can also be the most cost-effective in the long-term. Refactoring enables organizations to take advantage of not just cost-optimization, but full cloud functionality as well.
Understanding which option is right for your applications begins with an assessment of the role the app plays. Is it a revenue-generating application that bears investing in? If so, perform a cost-benefit analysis to determine the cost in terms of resources and downtime, and the benefits the application stands to gain from a replatform or refactor. If the application does not produce revenue and merely needs to be sustained, it makes more sense to look at the feasibility of rehosting, or replatforming where rehosting is not practical.
It’s complex, choose clarity
Choosing between rehosting, replatforming and refactoring is a complex undertaking. Fortunately, any cloud provider worth their salt will be willing to take this responsibility on for you. If you’re interested in learning more about what it takes to perform a successful cloud migration, contact the experts at Hanu today.