Google Cloud Migration Strategy

Assess

Assess the appropriateness of apps and workloads for the Google Cloud.

Primary factors include the following:

• Hardware configuration and performance
• Issues concerning the number of users and licencing
• Compliance
• Interdependence of applications
• Sort your applications into three categories:
• Simple to transport
• It’s difficult to relocate
• Incapable of movement

Pilot

Migrate one or two apps, ideally from the “simple to migrate” bucket. For a period of time, run the applications in production and monitor performance. Consider the licencing implications of scaling up your Google Cloud workloads or adding more apps, and always prepare for a rollback to on-premises or another cloud in the event of a migration failure.

Move Google recommends that you migrate all of your data to the cloud first, followed by the remainder of your apps. Consider Google Cloud Storage’s storage tiers (Standard, NearLine, and ColdLine), SSDs vs. traditional hard drives, and database services such as Google Cloud SQL, Datastore, and Bigtable. Prepare for the physical transmission of data — data transfer, transferring an offline disc to a Google data centre, streaming to persistent drives, and so on.

Move Applications

If possible, directly migrate your apps to Google Cloud. For instance, by building a local virtual machine that represents your workload and importing it as a Google VM, or by backing up your app to GCP and so generating a cloud replica automatically. If a straightforward solution is not feasible, try recreating apps in the cloud using a mix of bespoke virtual machines and GCP infrastructure services.

Optimize

Consider methods to improve apps now that they are operating on the cloud:

• Ensure the redundancy of your applications across GCP availability zones
• Prepare for catastrophe recovery with the help of Google Cloud Backup.
• establish flexibility via the use of autoscaling groups.
• Utilize Google Stackdriver to monitor your workloads.
• Conserve static assets by storing them in cold storage.
• Launch and scale new instances using Google’s Deployment Manager.
• Automatic Data Center Migration from On-Premises to Google Cloud
• While there are many ways to migrate to Google Cloud, a popular route is to take local VMware vSphere-managed virtual machines and migrate them to GCP in their current state. Google Cloud completely automates this process. Convert to Compute Engine.

The following are the typical procedures for automatically migrating your vSphere workloads to Google Cloud Platform:

Configure your Google Cloud Platform organisation, including your account, organisation, project, networking, roles, and service accounts (see full instructions).

Establish a secure communication route between your on-premises data centre and GCP through Google Cloud VPN (which utilises an IPsec VPN connection) or Cloud Interconnect (a highly available, low-latency connection).

Configure corresponding firewall rules on-premises and in GCP.

GCP has a management interface called Velostrata Manager that enables you to manage the migration. You must install and configure the Velostrata Manager through Google Marketplace.

GCP offers an on-premises virtual appliance that handles the migration process from inside your own data centre. Create a role in VMware vSphere and follow the steps below to setup the Velostrata On-Premises Backend.

Create Cloud Extensions for your migration now — these are used to move virtual machine storage between your on-premises environment and GCP. This article will demonstrate how to do this with the vSphere vCenter Web Client.

If you have Linux virtual machines, learn how to migrate them.

On the vSphere Console, you can now run any VM in the cloud.