What is a Cloud Architect?

Consider Instagram: It’s a social network designed to connect millions of users worldwide, which enables individuals to create personal accounts and post photos and videos on their stories; like and comment on posts; add friends; and directly message other users. How does something so simple require hundreds of employees and hundreds of millions of dollars in expenditures to manage?

As you can imagine, each one of these business features is an entire system architecture on the backend. 
Creating a personal account and managing the entire user login flow is a system architecture. 
Directly messaging other users is a system architecture.
Serving a live feed to your users is a system architecture. 
And each of these elements of Instagram requires a cloud architect to understand the business objectives and requirements, to assess the feasibility of various architectures, and to recommend a design solution.

Example: Instagram Posts feature. 
Instagram’s core feature as a business is to enable a user to post a photo or video directly to a personalized feed with a caption, a geolocation, and hashtags and to tag other users in the post. All users worldwide should be able to access this post in near real time, with 99.999 percent availability, and the post should be tamper-proof to maintain its integrity. If the post is archived, it should be invisible to other users.
There are more than 1 billion monthly active users on Instagram posting more than 80 million posts a day, so the solution needs to be cost-effective, because Instagram already manages a storage system of more than 100 petabytes of data. For a rough idea of how much this would cost on Google Cloud Storage, this translates to roughly $31 million per year in at-rest storage fees plus network egress and operations fees. At $0.12 per gigabyte of network egress and $0.05 per 10,000 operations, you could imagine this would raise the cost for Instagram significantly, as the entire business is about users accessing posts globally. As an architect, you need to design a system that can support serving these posts to more than a billion users worldwide in the most cost-effective manner. 

As a cloud architect, your goal is to assess a business objective at hand and distil the needs into business requirements and technical requirements. You’ll then use those requirements to architect a solution. You won’t always get all the requirements laid out so easily, and oftentimes you’ll spend hours upon hours diving into meetings with various stakeholders to gather pieces of data that start to formulate the building blocks of the solution. You’ll also need to be aware of future technical and strategic implications, ensuring that you’re building a solution that won’t encounter any major roadblocks.

Prerequisite Knowledge 
At this point in your career, if you’re pursuing the Professional Cloud Architect certification, you’re most likely not a high school senior who just nailed AP History and has a side hobby of learning GCP. This certification is geared toward individuals with multiple years of industry experience. Google Cloud certification does not have hard requirements like Certified Information Systems Security Professional (CISSP) certification does, however, where you have to prove five years of domain-related experience before you can be formally awarded the certification. Anyone can take a Google Cloud certification exam, but Google recommends that you have several years of industry experience, with more than a year designing and managing solutions using GCP, before you do so. (See https://cloud.google.com/certification/cloud-architect for more information about the exam.)

Some of the prerequisite knowledge that the Professional Cloud Architect should possess. 
A Google Cloud Certified Professional Cloud Architect enables organizations to leverage Google Cloud technologies. Through an understanding of cloud architecture and Google technology, this individual designs, develops, and manages robust, secure, scalable, highly available, and dynamic solutions to drive business objectives. The Cloud Architect should be proficient in all aspects of enterprise cloud strategy, solution design, and architectural best practices. The Cloud Architect should also be experienced in software development methodologies and approaches including multitiered distributed applications which span multi-cloud or hybrid environments.

- Enterprise cloud strategy
- Solution design
- Architectural best practices
- Software development methodologies and approaches including multitiered distributed applications which span multi-cloud or hybrid environments A

As a certified Cloud Architect, you’ll most likely be interfacing with senior business and technology leaders. For the C-suite, the key focus is setting the technology strategy, aligning this strategy to the business goals, and leading the charge. You will focus on turning that executive business strategy into tangible solutions by designing a technology architecture that enables the business needs.
Executives don’t have enough time to know all the specifics to help in deciding between one cloud program over another, or what technologies to leverage to develop a new core business offering. They need problem-solvers with foresight, who can help distill a business objective into incredibly complex technology, bundle it back up into a clear and concise solution, and present it back to the leadership in laymen’s language. As a cloud architect, you have the ability to speak both languages, the language of the business and the language of the technology, as you’ll be interfacing with stakeholders across both worlds.
Beyond the C-suite and leadership, you’ll be in the weeds with senior technology leaders and senior engineers. These are the stakeholders who will hold the knowledge of existing systems, team dynamics, and engineering expertise. These are all valid points to consider when you’re building a solution with key milestones and timeline considerations.

Suppose, for example, that you’ve decided to use Kubernetes as your application layer to eliminate the gripes of manual deployment and scaling. The product, however, needs to launch by Q3, and the engineering team does not have any experience in Kubernetes. So how does that affect your solution? Technical stakeholders hold the key to the other half of the requirements, the technical requirements, and you’ve got to understand how those factor into the situation. A cloud architect must have a strong understanding of software development methodologies, IT/technology systems, and architectural best practices, and the architect must use a methodological approach to designing an architecture.

Think about some of the key elements that software architects leverage, such as gathering functional requirements and nonfunctional requirements:
- Functional requirements The “what”—What is the system supposed to do? For example, my system needs to extract data from this API and load it into a storage bucket.
- Nonfunctional requirements The “How”—How should my system perform? What are the constraints? For example, my system needs to transform the data to a certain format before it’s loaded, or it needs to process at least X amount of data objects per second.

A cloud architect would approach designing a solution in a very similar way that a software architect would: Schedule your scoping meetings. Gather your requirements—business requirements and technical requirements. Do some research for more nuanced requirements and understand the constraints of the system. Put together a high-level design diagram. Work through the high-level design and then break it down into deeper design diagrams. Ensure that you’re aligned with business and technology stakeholders along the way. Work toward a final draft and get the necessary approvals if you’re not the accountable stakeholder.

Then you’ve got a reference architecture that your development teams can use to begin building. You’ll probably want to reuse this pattern for similar use cases and continually refine it. You should also have a solid understanding of Agile best practices and the overall software development life cycle, understanding how code gets built and the various deployment methods (such as blue-green deployments, canary deployments, continuous integration/continuous delivery [CI/CD] deployment).
Aside from all that, by this point in your career you’re assumed to have a good understanding of foundational technology and cloud concepts. 

Here are a bunch of concepts that you’re expected to know ahead of this exam:

- Identity and access management concepts such as LDAP, SSO, RBAC, and Active Directory
- Networking concepts such as DNS, DHCP, firewalls, and CIDR
- Computing concepts such as VMs and containers
- Data concepts such as object storage, block storage, and network file systems
- Engineering concepts such as DevOps, CI/CD, and IaC
- Security concepts such as least privileges, segregation of duties, and defense in depth
- General concepts such as multitiered architectures, service-oriented architecture, microservices, hybrid cloud, multi-cloud, and so on