There are many different cloud services, and each cloud service is made up of a number of building blocks. The cloud architect’s tasks range from designing a cloud infrastructure to selecting the right cloud services, from implementing the cloud infrastructure to the configuration on-site. In this article we’re going to have a look at what it takes to be a cloud architect.

Before we get into that, though, let’s first define what cloud architects do: they design and implement infrastructure as code for an organization on top of clouds like Microsoft Azure or Amazon Web Services (AWS). They provide guidelines for teams in their organizations related, but not limited to, these technical details.

To sum up: Cloud architects design and implement cloud infrastructures, provide guidelines about cloud technologies and cloud usage, and educate the people in their organizations about cloud related topics.

A cloud architect’s work can be quite different from one organization to another. As cloud computing is still a relatively new resource class, cloud architects need to ensure that their cloud designs are aligned with organizational needs. To do so they take several factors into account:

– Security and Compliance (ISO 27001 or other security standards)

– Privacy (e.g., GDPR compliance)

– Availability (99.99%+)

– Flexibility (ability to scale up and out; rolling upgrades; patching without downtime; etc.)

– Cost Efficiency (cost effective as compared to on-premise infrastructure or cloud competitors)

– Performance (speed and throughput; latency; resource guarantees; etc.)

– Availability of cloud services (e.g., Azure Active Directory)

These cloud design principles are the basis for cloud architects to decide which cloud services they’re going to use in their on-premise cloud infrastructure. Cloud technologies can be categorized into several types: IaaS, PaaS, FaaS, CaaS and XaaS. Within these categories cloud architectures can be distinguished by several factors like availability, cost efficiency, performance or flexibility.

Read more:- AWS certification path

As an example let’s have a closer look at different cloud services within the Infrastructure as a Service category. For each service there are different design considerations that need to be considered when building cloud infrastructures.

Azure Virtual Machines, for example, provide cloud services that can be used to create cloud compute nodes or cloud storage nodes. Nodes are part of a cloud infrastructure provided by cloud providers like Microsoft Azure or Amazon Web Services (AWS). It includes the underlying hardware resources necessary for running applications and/or processing data.

Azure VMs offer different compute sizes, based on the number of CPU cores available. These VM sizes range from small (1 core), medium (2 cores) up to extra large (8 cores). Depending on how complex an application is, you might need more or less CPU power. If your application needs more CPU power – say 8 cores instead of 4 – you’ll have to scale out your cloud infrastructure and provision more cloud compute nodes with the same cloud service.

Azure Virtual Machines provide different storage types: attached disks, cloud disks and managed disks. Depending on the use case (e.g., high-performance cloud storage, cloud backup, etc.) you’re going to use one of these storage types.

Access to Azure VMs can be done via RDP for Windows or SSH for Linux (which requires a Linux endpoint as well). When accessing an Azure VM this way, network traffic has to traverse from your cloud infrastructure through the internet before it gets routed over your local network to your cloud compute node(s) where your VM is hosted. This can lead to network performance problems if the latency between cloud infrastructure and cloud compute nodes is high.

Azure VMs are provisioned with a public IPv4 address. This cloud service helps you to create cloud compute nodes in your cloud infrastructure, but doesn’t help with network security or cloud compliance factors like DNS records for each cloud compute node.

Microsoft Azure Cloud Service Providers (CSPs) are organizations that have signed up to offer their cloud services through Microsoft’s billing and support systems. Some examples of CSPs include Heroku (PaaS), Twilio (XaaS), Engine Yard (PaaS), etc. The most important factor when deciding which CSP to choose is the integration between the on-prem cloud infrastructure and the cloud provider’s cloud service.

If you’re using Microsoft Azure as a cloud provider, then cloud services can be integrated with the on-premise cloud infrastructure via different connectors like ExpressRoute or VPN . When considering cloud service providers it’s important to consider which connector is available.

At the end of the day an architect has to make design decisions about all these things, because there are many cloud technology options available for creating cloud infrastructures that can be used in combination with each other. For example, you could use IaaS compute nodes running Linux and connect them together over a virtual network with PaaS workloads running on top of those VMs. This would require designing your own cloud storage solution since managed disks aren’t available for Linux cloud compute nodes.

Categories: Education