Infrastructure Migration to the Cloud: Problems and Experience

About 80% of cases of companies’ with less than 200 users migration account for the transfer of an already virtualized infrastructure from the local server to the data center, the remaining 20% ​​is the transfer directly from the users’ PC to the virtual environment plus the removal of the computing node to the cloud.

Background

The first thing that is done is an assessment of capacity and infrastructure. In fact, you need to have a list of services and know what capacities are needed for them and what needs. As a rule, there are issues of data transfer and conversion of virtual machines into the format of the proper hypervisor.

Migration process

It often happens that a customer has one virtualization platform in a server node, and the move is made to another for various reasons. For offline migration, information is most often collected using the Clonezilla product. An even simpler method is to turn off the product, load it into the file storage and then convert it.

If you need to move online, the application is most often used, which transfers only the necessary part, and not the entire operating system. There are cases when the database is of such size that even a 10 gigabit channel will be small. Then you can upload everything to your hard drive and bring it to the data center in order to upload it to the cloud. Another method of moving is through the backup and restoring on a new site, if the customer has a good backup or disaster recovery scheme.

If a company on its site used cluster when migrating to the cloud than such architecture usually requires rework. More precisely, uploading a classic cluster to the cloud is not the most rational approach. There are always difficulties with this: how exactly to do it? Cluster software itself will be an extra level. Here it is necessary to attract experts on clusters. Most often, the cluster is removed in advance on the side of the customer, because unloading it into a cloud is meaningless.