Google Professional Cloud Architect - The Complete Guide

Explore what the cloud is and why it matters, contrasting on-premises servers and IT maintenance with cloud scalability and cost efficiency, illustrated by a Black Friday surge.

The cloud provides compute, networking, storage, and services managed by others in global data centers, enabling you to spin up servers in minutes and pay only for what you use.

Explore the five cloud characteristics: on demand self-service, broad network access, resource pooling, rapid elasticity, and measured service, and see how automated provisioning and usage-based pricing empower scalable resources.

Explore CapEx vs OpEx to understand the financial models for cloud adoption; compare upfront investments and ongoing usage costs, highlighting the cloud's flexible OpEx advantage.

Explain IaaS, PaaS, and SaaS by detailing compute, networking, storage, and runtime responsibilities, from virtual machines to web apps and cloud software like Office 365.

Explore the three cloud types—public cloud, private cloud, and hybrid cloud—and understand their access, management, and infrastructure.

Explore cloud providers and services such as IaaS, PaaS, and SaaS, focusing on AWS, Azure, and Google, and note Google Cloud’s 22% growth in Q1 2023.

Explore Google Cloud Platform, or GCP, Google's public cloud announced in 2008 and released in 2011, evolving from App Engine serverless PaaS to broader IaaS and services.

Map regions and zones in Google Cloud Platform, explaining data centers, availability, proximity to audience, and pricing considerations, with global, regional, and zonal service examples.

Explore Google Cloud resource hierarchy from organization to folders to projects, where resources reside in a project and inherit policies; in the free tier, organization, folders, and groups are unavailable.

Explore Google Cloud Platform services and their core compute and database offerings, including App Engine, Cloud Run, Compute Engine, Alloydb, Bigtable, Cloud SQL, Memorystore, and Spanner.

Create a Google Cloud free account to get $300 credit for 90 days and learn its limits. Understand constraints like no Windows virtual machines on Linux or GPU virtual machines.

Explore the Google Cloud Console, from sign-in and project ID visibility to the left navigation and the search bar, and access resources like IAM, Compute Engine, Kubernetes, and Cloud Storage.

Learn how to create resources in Google Cloud Platform by navigating the left menu to Cloud Storage, selecting buckets, naming your bucket uniquely, and confirming creation.

Copy the bucket name from the created resource, paste it into the Google Cloud console search, and open the bucket from the resources list.

Learn how to remove a bucket in Google Cloud by selecting it, choosing delete, confirming with the delete text, and reviewing the deletion notification.

Master Google Cloud resources with Cloud Shell and the gcloud command, learning to create and delete secrets, manage projects, and compare console versus shell workflows.

SLA stands for service level agreement and defines yearly uptime and downtime; compare App Engine and Cloud Spanner SLAs, including multi-regional vs regional, and compute SLA by multiplying component SLAs.

Explore cloud cost concepts, including per resource, per consumption, and reservations, and learn to use the Google Cloud Pricing Calculator to estimate compute and storage expenses before provisioning.

Create a specified amount monthly budget in Google Cloud to monitor costs and trigger alerts at 90% and 100% thresholds, with email notifications to billing admins.

This lecture explains how cloud architects must blend traditional architecture skills with infrastructure knowledge, including networking, firewalls, VPNs, load balancers, and cloud security, to design secure, hands-on cloud infrastructure.

Build a real-world cloud bookshop app with four services: catalog, shopping cart, inventory, and order engine, while deploying locally and to the cloud and exploring compute, networking, security, and cost.

Install the dotnet six SDK to prepare your computer for developing the Reddit app, download the SDK from dotnet.microsoft.com/download/dotnet/6.0, and install for your operating system.

Install and launch Visual Studio Code to serve as the development tool for building the Reddit app, selecting the OS-compatible version and pinning the icon for quick access.

Install the C sharp extension and the Google Cloud Code extension in VSCode to support dotnet core and cloud work; the C sharp dev kit adds solution explorer.

Set up and run the Reddit app locally by loading the catalog, opening the project in VS Code, and pressing F5 to render the catalog with four books in-memory cache.

Open the catalog project in VS Code and inspect index.cshtml and index.css to see how the books table renders from view data books, driven by an in-memory cache.

Explore Compute Engine, Google Cloud service for creating and managing virtual machines, an IaaS offering, and review cloud service models (IaaS, PaaS, SaaS) and control levels.

Understand how virtual machines in Compute Engine deliver servers via host operating systems, hypervisors, and guests, with a 99.5% sla, VM density, and creation steps like location, image, and size.

Explore Google Cloud VM families across general purpose, compute, memory, and accelerator workloads, with series like e2, n2, c2d, h3, m2, a2, and options for GPUs and custom configurations.

Learn to create a vm with Google Cloud Compute Engine, estimate costs via the pricing calculator, and deploy an E2 small in Europe West, within the free tier.

Connect to the Google Cloud VM in the browser via SSH and view details. Note internal vs external IP addresses and the security risk of exposing the VM.

Discover techniques to reduce virtual machine costs, including scheduled shutdown, spot instances, committed use discounts, and disk type choices, and learn how to combine them for optimal savings.

Learn to automate cost savings with instance schedules that start and stop VMs on a daily cycle, attach schedules to instances, and resolve permissions using Cloud Shell.

Spot instances save up to 91% and can be stopped or removed by Google, ideal for batch processing and data analytics, with run-time limits and shutdown behavior configured.

Learn how committed use discounts save up to 70% by paying the committed price for 1 or 3 years, with Compute Engine instances by region, cores, memory, and disks.

Choose the disk type for your vm instance to balance performance and cost; compare local ssd (ephemeral) with persistent disks (balanced, performance, standard, extreme) across regional-zonal options and iops.

Learn how to delete a VM instance and its boot disk to save costs by removing unnecessary resources in the cloud console.

Create a catalog VM for the Reddit app by selecting a nearby region, using an E2 medium standard instance, and configuring a standard persistent disk and http and https firewall.

Learn how to access and customize VM metadata in Google Cloud via HTTP requests to the metadata server, using the Google header to retrieve instance, project, and maintenance data.

Explore sole tenancy, using sole tenant nodes for your VMs, a costly option. Estimate costs with the pricing calculator and learn to create a node group by region and template.

Discover how instance templates enable consistent creation of virtual machines with identical configurations, including global, regional, or zonal settings, while templates are immutable and you pay only for deployed machines.

Create and deploy identical VM instances in GCP using instance templates, configure ARM-based templates, choose T2-a, SSD persistent disks, and deploy from Europe West four; templates are immutable and cost-saving.

Publish the catalog application, start the catalog vm in the cloud, and deploy files via putty using ssh keys, then install the dotnet runtime on Debian to run the app.

Publish the catalog app to a Google Cloud VM by uploading the publish folder, setting production, running dotnet catalog dll, and enabling a startup script for port 8080 access.

Master cloud backup with snapshots by understanding exact disk copies at a point in time, incremental saves, and scheduling across standard, archive, and instant types for cost-effective, reliable recovery.

Learn to create and manage snapshots in Google Compute Engine, including standard, instant, and archive types, schedule daily backups, and use snapshots to boot new instances or create disks.

Learn how Google Cloud images in a cloud storage bucket enable backup, maintenance, and cloning of machines, including machine and disk image types, with costs based on size.

Create a machine image from a catalog VM to duplicate and scale. Launch a new VM from that image with the same boot disk, network tags, and startup script.

Learn to use managed instance groups to deploy scalable vm fleets with autohealing, health checks, regional deployment, load balancing, auto scaling, and automatic updates.

Deploy the catalog app on a regional managed instance group by creating a catalog disk image and template, then enable auto scaling and health checks.

Configure a weather API for the catalog by provisioning a weather VM in the same region, disabling external firewall access, and using internal IP access from the catalog VM.

Understand VM manager in Google Cloud, handling OS patch management, OS inventory management, and OS configuration management for VMs. Enable per project or per VM, free up to 100 VMs.

Learn how to enable VM Manager for a project or a specific VM, view OS info, and monitor catalog VM instances through VM Manager and OS Config API.

Design GCP architecture diagrams using official icons, cheat sheets, and slides; download SVG/PNG icons and use the diagramming tool to illustrate services like Compute Engine with best practices.

Explore Google Cloud Platform's App Engine, a managed runtime that scales automatically and handles infrastructure, letting you deploy code while you control only your application code and data.

Explore the App Engine architecture, including a single project region, multiple services and versions, and traffic routing. Understand how Google manages instances, scaling, and the default service.

Compare standard and flexible App Engine types, including sandboxed runtimes, rapid scaling, background processing, web sockets, and pricing models for instance hours versus vcpu memory and disks.

Create and configure the App Engine application, add app.yaml, and deploy the service with gcloud, optionally specifying a version to manage traffic; access via default and service URLs.

Explore Cloud Build, a continuous integration and deployment service that builds and deploys code automatically or manually, integrating with App Engine, Cloud Run, Google Kubernetes Engine, Cloud Functions, and Firebase.

Deploy a standard App Engine service by creating an application, selecting a region, and deploying a Python 3.11 standard environment with gcloud app deploy; manage versions and traffic.

Deploy the inventory service as a flexible App Engine app using gcloud, an ASP.NET Core runtime, and app.yaml, with automatic scaling from one to five instances and a version v1.

Learn how App Engine versions handle traffic, with default routing to the new version and access to older versions via a version-prefixed URL; master gradual traffic split for deployments.

Deploy a new App Engine version (v2), verify it locally, and split traffic 50/50 between v1 and v2 using random, cookie, or IP address methods for gradual updates.

Explore traffic splitting and deployment types—basic, rolling, blue-green, and canary—in App Engine. Learn the pros and cons and how to implement each with traffic control.

Compare cron jobs and cloud scheduler for scheduling App Engine, noting cron's get-only schedules and free usage versus cloud scheduler's pricing ($0.10 per job, five free) and complex options.

Create and configure a cloud scheduler job by defining a cron expression for timing, selecting the App Engine Http target with a specific service and version, and setting retry options.

Analyze the current cloud architecture with two compute engine vm instances for catalog and weather api. Add an app engine inventory service to manage the Reddit bookstore inventory.

Explore how containers enable isolation, higher host density, and a single code base, and learn about Cloud Run and Google Kubernetes Engine for running containers.

Containers bundle software, dependencies, and configuration into portable units that run on the host operating system, delivering predictable, fast startup and higher density than virtual machines.

Discover how Docker powers containers by explaining the Docker daemon, images, container registries, containers, and the Docker client, and learn how to customize images with a Dockerfile.

Discover Artifact Registry, a regional or multi-regional registry for docker images and packages (java, python, go) plus Debian packages, with integrations to Compute Engine, Cloud Run, and GKE.

Navigate to artifact registry in the console and create a docker format repository named 'my repo' in the Europe West region using standard mode.

Build a Docker image of the cart and publish it to the artifact registry using Cloud Build. Then deploy the image to Cloud Run.

Explore Cloud Run, a managed container platform offering autoscaling, traffic revision management, and public or private deployments, with service and job workloads and CPU allocation options.

Deploy a Cloud Run service from an artifact registry image, choose a near region, enable a minimum of one instance with unauthenticated access, and review revisions and logs.

Deploy from source enables one-step deployment from VS Code to Cloud Run, pushing the image to Artifact Registry’s Cloud Run Source Deploy repo, with traffic routed to the new revision.

Learn how to create cloud run jobs, a unit of work that can be scheduled, configuring image, name, region, and resource options including memory, cpus, timeouts, retries, and parallelism.

Explore containers management strategies to deploy, scale, monitor, and route multiple container instances, ensuring high availability and reliable operation for cloud architectures.

Discover how Kubernetes, the de facto standard and leading container management platform, uses pods and services to provide routing, scaling, high availability, automated deployment, and configuration management.

Explore GKE, Google Kubernetes Engine, a fully managed Kubernetes environment in Google Cloud Platform that handles cluster nodes and pods, with autoscaling, upgrades, security auto repair, monitoring, and 99% SLA.

Explore GKE types: standard and enterprise editions with single or multi-cluster deployments, autopilot automation, service mesh, policy and team management, and Anthos pricing across clouds.

Deploy the cart container to GKE autopilot, create a cluster, deploy the artifact registry image as a public service with a load balancer, and access the cart by its IP.

Analyze the current architecture featuring five components: inventory app engine, catalog VM with weather API on a VM, and two shopping carts—one in GKE, one in Cloud Run.

Explore cloud functions, a serverless, event-driven environment that runs single-purpose code when triggered by events or http requests, with multi-language support, cost efficiency, and a 99.95% sla.

explore cloud functions, a serverless, auto-scaling environment that runs code in Node.js, Python, Go, Java, Ruby, PHP, and DotNet, with v2 concurrency and traffic splitting.

Learn to create and test a .NET Google Cloud Function locally, triggered by an http request, to process orders and validate the flow using postman.

Learn how to deploy a dotnet six cloud function to Google Cloud using the gcloud CLI, with a Gen2 runtime, an http trigger, and unauthenticated access.

An updated architecture introduces a new order function, temporarily wired from users via the shopping cart, with plans to add databases and networking for a real-world app.