Virtual Machines in 2023 and Google

I wanted to take some time to talk to you about Virtual Machines… I know, I know, they’re not as sexy as running things in the latest serverless iteration, or as modern as containers. But they still have their place, not to mention the hundreds or even thousands of them you may have running in your data centre(s), with this being especially true if your organisation has been around over 10 years or more.

There will be some of you who will be thinking, why would anyone still be running VMs in 2023? But for the majority of organisations who have them, the answer is because they work: because they’re currently running our critical SAP or line of business application; because their vendor can’t or won’t modernise them — the list goes on.

Now don’t get me wrong, I’m not saying organisations shouldn’t be modernising the workloads that currently depend on these VMs, but sometimes it might be out of their control, there may be licensing commitments, or maybe they just don’t have the cash to invest in such a modernisation just yet. So what then? How can they take advantage of the cloud? What happens when Facilities tell IT, “we’re not paying for the ‘leccy bill of those data centres any more?” — those same data centres that are running those virtual machines, which the running of the company depends on.

Now, nothing in the blog post will set the world alight, these things aren’t new, but there are different considerations when you’re migrating on prem to cloud vs migrating to the latest iteration of ESXi.

To start the process off, I want to give you an overview of the tools available in Google Cloud to migrate existing Virtual Machines.

Migrate to Virtual Machines

Migrate for Virtual Machines Dashboard

This product has had a couple of names, originally Velostrata, was acquired by Google. Its previous name was Migrate for Compute. I will admit the latest name change makes it unambiguous as to the products function. (Yes, it does actually migrate virtual machines). It’s primarily a V2V (virtual to virtual) tool for migrating VMs from VMware, Azure, EC2 and some physical servers (P2V).

The latest version of the product makes light work of getting set up, with the only hard requirement being a HTTPS/443 connection to Google Cloud from an virtual appliance which is deployed on premise on a supported environment. Migrations can be set up in a matter of clicks — and I’m not exaggerating!

See https://cloud.google.com/architecture/migrating-vms-migrate-for-compute-engine-getting-started for more information.

Migrate to Containers (Migrate for Anthos)

This product is still fairly new, but has had a lot of Google love already. It allows for quick, and automated containerisation of the popular and well known stacks, and supports both Linux and Windows. Millage may vary with this, as when I last tested the Windows variant couldn’t handle dependent services, however I’m certain this will come with time. Linux is handled much better, and has a wider range of supported configurations including Apache, Tomcat and JBoss servers.

For the full details please see https://cloud.google.com/migrate/containers/docs

Google Cloud Vmware Engine (with HCX)

HCX, whilst having more prerequisites than Migrate to VMs, can make the process of migration as quick, as simple and as seamless as if you were performing a vMotion (host-to-host migration) within the same data centre. Once connectivity is established between Google Cloud and your on premise, it allows Google Cloud to act effectively as an extension to your Data Centre.

Features include:

• Bulk migration of virtual machines, with scheduled cutovers and low downtime (less than a minute typically)
• HCX vMotion — migrate virtual machines with no downtime
• Stretch your layer two network for the duration of the migration (no re-IP required)

So those prerequisites? Other than making sure your VMware on-premise versions are supported, VPN connectivity is a hard minimum requirement.

Now don’t be fooled into thinking this tool is just a way to migrate VMware virtual machines to Google VMware Engine, because it’s not! Support for migration from Hyper-V and KVM is available with HXC OS Assisted migration. VMware’s Rajeev Chawla published an article on the very topic back in 2019 https://blogs.vmware.com/cloud/2019/08/08/vmware-hcx-enterprise-non-vsphere-vsphere-workload-migration/

For further details on VMware HCX and Google VMware Engine please see https://docs.vmware.com/en/VMware-HCX/4.5/hcx-user-guide/GUID-A7E39202-11FA-476A-A795-AB70BA821BD3.html and https://cloud.google.com/vmware-engine

What tools are there for Non-VMware VMs?

That is an interesting question, and unfortunately all I can say is, this is something that is currently lacking in the product set. I’m not trying to put you off on your quest to migrate to Google, but unfortunately it’s going to be a somewhat more bumpy ride. Depending on the target within Google Cloud (i.e. Compute Engine vs VMware Engine), a typical migration path would be to import disk images or migrate to VMware Engine using HCX as mentioned above.

Now I’m in Google Cloud, now what?

So enough of the tools, what do I gain from running Virtual Machines on Google Cloud?

• Sustainability, for one. Google has been carbon-neutral since 2007 and has a goal to be running on carbon-free energy by 2030
• Google looks after the underlying infrastructure (as detailed in Google Cloud’s shared responsibility model), so no more worrying about those power supply failures at 1am
• Direct access to native Google products and tools, at high speed — now this is where I think the real gains can be. Why not make more use of that data stuck in that SQL database?
• Not to mention all the elasticity you can want. Need your scale your VM instance based on variable loads? You can. Need to spin up a new copy of your application to service that new office in Sydney, just a few clicks or lines of code away.

Now at the start of the post, I went into why businesses might not be able to modernise some of their VMs — but what I didn’t say, was that once the Virtual Machines are operating in Google Cloud, it makes it so much easier to chip away at those applications, modernising and improving in place.

You can spin up additional environments, test configurations, try new things and tear them down at will. You can start looking at modernising workload RDBMS backends with CloudSQL, strategically protecting them with Cloud Armor or an API Gateway (APIGee), or just the ability to “let loose” your engineers to fully test containerisation in a risk free way using a segregated project and GKE or Cloud Run.

The wrap-up

It’s worth remembering that moving to cloud and modernising your workloads to fully exploit the true power of the cloud is a journey — but using the information above, the first steps might be easier and more impactful than you think!

CTS is the largest dedicated Google Cloud practice in Europe and one of the world’s leading Google Cloud experts, winning 2020 Google Partner of the Year Awards for both Workspace and GCP.

We offer a unique full stack Google Cloud solution for businesses, encompassing cloud migration and infrastructure modernisation. Our data practice focuses on analysis and visualisation, providing industry specific solutions for; Retail, Financial Services, Media and Entertainment.

We’re building talented teams ready to change the world using Google technologies. So if you’re passionate, curious and keen to get stuck in — take a look at our Careers Page and join us for the ride!