Multi-cloud approach, AWS–Google and the “New” Microsoft Azure Arc in 8 steps

Today, the adoption of cloud computing has quickly become a key driving force for businesses. Selecting one cloud over the other, comes down to the wants and needs of each individual customer and the workloads they are running. It is often the case that organizations will use multiple providers within different parts of their operations, or for different use cases, which is called a multi-cloud approach.

However, there are a number of differentiating factors that separate the approaches of the three most known firms (Azure, AWS & Google), which can help end users consider which is right for them. These Platform offer largely similar basic capabilities around flexible compute, storage and networking. They all share the common elements of a public cloud: self-service and instant provisioning, auto-scaling, plus security, compliance and identity management features.

Customer Needs

A lot of customers have servers running in their private data centers, in their offices, or even have other parts in the organization which they use another Cloud provider or another service provider.

One of the main challenges with all these servers they have is basically keeping control of all of them whenever they are running to make sure that are secure, that they are patch, that they have the compliance. The need is to manage these very like “Hibrid environments”. It doesn't matter which Cloud, they just need to be able to go in, and deploy things.

What are Google Anthos – AWS Outposts– Azure Arc? And how can this help to our Customer needs?

Choosing to use public, private or hybrid cloud is difficult as each platform comes with its own advantage. However, today multi-cloud is the next big thing, offering businesses multiple cloud services to get the best of every offering in one simple infrastructure. So instead of having to move between platforms, businesses can bring AWS, Azure or Google together in one system to efficiently operate their business, depending on their aims and objectives. Here′s what each of these companies offer:

AWS Outposts (details here), brings AWS infrastructure, services, and operating models to virtually any data center or on-premises facility. It is designed for connected environments and can be used to support workloads that need to remain on-premises due to low latency or local data processing needs.

Google Cloud has Anthos (details here). In Anthos, Google has a fully-managed, broadly-supported, multi-cloud solution for containers. It’s not just a product but an umbrella brand for multiple services aligned with the themes of application modernization, cloud migration, hybrid cloud, and multi-cloud management.

This month in MS Ignite 2019, Microsoft launched Azure Arc (details here), which is their new hybrid and multi-cloud platform. Azure Arc brings Azure products and management to multiple clouds, edge devices, and datacenters on any infrastructure, with the goal of unifying orchestration and governance. With Azure Arc, developers can build containerized apps with the tools of their choice and IT teams can ensure that the apps are deployed, configured, and managed uniformly using Git Ops-based configuration management

What is, and how does Azure Arc Work?

First we need to understand how Azure works. Microsoft manages Azure services trough Azure Resource Manager (ARM). Ever since they launched it, Microsoft has continuously improved it, making even easier to manage virtual machines, database instances, Hadoop clusters and Kubernetes clusters. If you want to understand the benefits of ARM read this. A lot of investment went into ARM; like identity, like RBAC, like policies.

Most importantly a lot of customers really care about compliance and also just regular management like tag them, show what are my servers are all in production, those kinds of simple things are all capable trough ARM. In addition, all the services now can be deployed onto Azure as well as on-prem in the same fashion.

As you can see on the following image, Azure Arc extends ARM capabilities to Linux and Windows servers, as well as Kubernetes clusters on any infrastructure across on-premises, multi-cloud, and edge.

From a management perspective, MS wanted all these solutions to manage the servers the same way for Azure as well as for on-prem and also they get the same ARM benefit.

Add your machines to Azure Arc in 8 steps

This is a page that we built to show all the on-prem servers that has been on-boarded to Azure. To on-board a server, we need to run a script on the server, which we generate trough a flow in Azure.

So this is option that they can click to generate the script but at the same time, it also recognize is a challenge for customers to on-board a scale if they have to connect to every single server individually to run these scripts.

1 - To do this we first go to https://portal.azure.com/ and search. Select option Machines – Azure Arc. 

2 – Once you select the Machines – Azure Arc option, click on “Create Machine – Azure Arc”.

3 –Select the option Generate Script. 

4- So we′re also trying to understand what are some common on-prem server management application so we can integrate to help customers to on-board those machines at scale. For example, here if the server is already managed by the Azure Updates Service, we build actually the script or the runbooks to actually deploy to on-board those machines onto Azure without actually customers touching all those machines.

Let me show you how to generate the script.

In this step, you can pick which subscription and resource group you want to go and here the region indicates that which Azure region is running these servers. So you can see from compliance or regulatory point perspective, we know where the metadata is stored in Azure.

Here we also can choose the operating systems Windows as well as Linux.

5 – Click “Next: Proxy Server”. For on-prem especially, if you don′t want to expose your servers to the internet directly you can put it behind a proxy server. If these servers are not connect to Azure directly, they can configure the proxy server here and then the agent will be able to communicate through the proxy server.

6 – Click “Next: Tags”.

This is just an Azure resource capability so they can tack the servers to indicate maybe who owns them or whether they are part of a team. So for example in my environment I tag resources based on production development environment, demo environments, and so on, so they can use the same tagging for their basically on-prem servers.

7 – Click “Next: Review + Generate”. In the end, we generate this script. So now you can take a copy of the script and run it on the target server.

8- Here we can see the script content. As you can see it involves three steps.

First you download the package, but if you actually already downloaded and put on a file share, you can just change that copy it off from that power share.

Second, you install that package.

The last one is the important one here which we′re actually doing the on-boarding.

This tool will actually create the ARM resource and the link back to the agent so that at the end of the on-boarding process, you will actually see these resource presenting that physical server in the Azure Portal.

So it makes it super easy to on-board the servers by basically creating them the script they need, they can also run the scripts against multiples of servers if they on-board like not just one or two servers but maybe hundreds of servers.

So we can have our server in the portal and we can see it and manage it using the Azure Resource Manager.

What about the cost?

It’s FREE , you don’t pay to onboard your machines to azure, you will only pay for the solutions that you are going to deploy onto those servers.

AWS provides a price calculator here, Microsoft here and Google here.