Table of Contents
Why it may not be a good solution for you
Most of our clients are not the ones who start afresh – they all have existing systems running at their premises or in some data centre somewhere, happily munching data of whatever they’re supposed to do. Only now they want to migrate those systems to the cloud, expecting to reap all the benefits of such a move as soon as possible. Preferably, with no downtime and with minimal costs. Hm…
Well, it can be done, usually. The answer is something that has the moniker ‘lift and shift’. Basically, this approach will replicate your existing infrastructure, servers and data to the cloud. At the end you’ll be getting the same thing that you already had, only now running in the cloud.
This approach is multi-phased and well supported by different migration tools most of the cloud providers readily offer.
Lift and shift – Assessment phase
The whole cloud migration process starts with the assessment phase. This is where we’ll analyze your current system and processes and determine if there’s a possibility of lift-and-shift migrating in the first place. Then, we will identify elements of your system and determine what are your good and bad areas and what should be done to improve those bad areas and fill the gaps.
Big cloud providers (AWS, Azure, GCP) offer tools and services that can help to accomplish this phase. There are also third-party tools and services that can help, like TSO Logic.
Lift and shift – Planning phase
After we assess your system, we will start the next phase – planning the migration. This will start with defining a clear business case. Next, we will perform a discovery process to identify all the elements of your system, including servers, applications, databases and storage elements. This will enable us to start planning the cloud migration process itself. This will include not just the technical side of the job but also the business side, including the organization and governance of such a process.
Security and compliance assessment of the new system has to be incorporated in the cloud migration planning phase as this is essential for the business continuity process. Last, but not the least important is planning the training and education of your ops teams.
Tools like AWS Migration Hub or Azure Migrate and services like AWS Application Discovery Service will help in this phase.
Lift and shift – Migration phase
After all the planning is done and signed off, it’s time for the actual migration. Several tools can help you in this phase. The goal is to automate the migration process as much as possible.
First, you have to prepare your cloud environment, setup the accounts, add users and organizational units, setup permissions. Tools like AWS Landing Zone can automate this process.
Next, you have to prepare your current system for migration. This usually includes installing software agents on servers and connecting them with some command and control centre that will manage the migration process.
Finally, the migration process itself. AWS CloudEndure Migration can perform a migration process, including rehosting the servers and migrating data automatically, and with no or minimal downtime.
Migrating to the cloud? Don’t miss the complete guide to cloud enablement >>
Lift and shift – Finalization phase
The new infrastructure has to be tested, secured and checked for compliance to required standards. Various tools in tool chests of your cloud providers can help you with that.
And now, you’re migrated to the cloud. Enjoy!
Why this approach may not be for you
You did all of the above and in record time. Now you’re on the cloud. Then, first bills arrive in your mailbox and your beancounters are…well, not that enthusiastic. The bills are larger than expected and the promised savings are not materializing. Your cloud infrastructure still struggles to cope with peak usage and short downtime of one of your cloud provider’s availability zones caused your services to be unavailable. What went wrong?
The point is – you’re not utilizing the best features of the cloud. Remember, what you did is just replicating your existing setup to a cloud. Your on-premises systems are sized to endure the highest loads which may occur only sporadically and disks have so much free space as they are planned to hold data in the next 5-10 years. Your infrastructure is not designed for resiliency because you only had one datacenter. Furthermore, you’re not using the IaaS products that your cloud provider is offering. Only if you embrace the entire philosophy of the cloud you will reap the benefits of it.
And that means that you need to re-architect your whole solution. But that’s the story for another blog post.
