Google Cloud Professional Architect: Get Certified 2022

Think like an architect by translating business needs into a technical framework across planning, design, and construction, bridging sponsors and domain experts with engineers through collaboration.

Examine how the wheel of architecture maps Google Cloud domains—compute, storage, networking, cost controls, security, and monitoring—across projects from batch processing to IoT and ML.

Identify the six domains of the Google Cloud Professional Architect exam, from designing cloud solution architecture to operations reliability, including security, compliance, and ci/cd practices.

Create a billing account in the Google Cloud console, link it to your resource hierarchy, and set up a payment profile with a credit card to bill projects.

Enable required Google Cloud APIs via the Google Cloud console's APIs and services page. Select the Vision API, click enable, and you'll be prompted to enable it if needed.

Create a Google Cloud Compute Engine virtual machine and evaluate when to use general purpose, high memory, or high cpu configurations, with labels, region, boot disk, and service accounts.

Learn about Google cloud machine families, from low-cost E2 and balanced N2/N1 to memory optimized M2/M1 and compute optimized C2, plus GPU-accelerated A2 for CUDA workloads.

Compare boot disk types from standard to extreme persistent disks, and explain how zonal versus regional replication affects cost, performance, and data redundancy, including local solid-state drives.

Use Google Cloud Compute Engine images to preinstall software and data, including deep learning and Debian options, enabling rapid replication of regional custom images from disks.

Explore Google Cloud tensor processing units, custom deep learning accelerators designed for deep learning model training, offering higher performance but lower precision than GPUs, with optional preemptible pricing.

Learn to create instance templates that describe a virtual machine's configuration, including operating system, disk, security, networking, and service accounts, enabling identical VM copies and use with managed instance groups.

Create custom machine types in Google Cloud by configuring exact cores and memory in the N2 family to fit workloads, including extended memory for higher memory per CPU.

Explore managed and unmanaged instance groups in Google Cloud, learn how autoscaling, load balancing, autohealing, and regional deployments boost availability, and distinguish stateless from stateful configurations.

Create a stateless managed instance group in the Google Cloud console, choose single-zone or multi-zone deployment, apply an instance template, and configure autoscaling with CPU utilization and health checks.

Create a stateful managed instance group in a single zone, mark the boot disk as stateful and detach on deletion, and set a fixed size with no autoscaling.

Discover Google Cloud's Kubernetes Engine (GKE) as a managed container orchestration service with clusters, node pools, auto healing, load balancing, and built-in monitoring; compare standard mode with autopilot mode.

Learn to create a Google Cloud GKE autopilot cluster with name, region, and public or private networking, then configure pod and service address ranges, maintenance windows, and exclusions.

Understand how Kubernetes scales architecture from the api server gateway to the control plane, then compare infrastructure and workload scaling via horizontal pod autoscaler and vertical pod scaler.

Kubernetes deployments manage replicas of stateless pods and replace failed ones via a deployment manifest, while replica sets maintain stable pod groups overseen by deployments.

Learn how stateful sets provide unique pod identities, persistent storage, and ordered deployment for stateful applications, while daemon sets ensure every node runs a pod for monitoring and logging.

Explore Kubernetes pod storage options in GKE, including cloud storage, cloud sql, cloud spanner, cloud file store, and persistent volumes backed by Compute Engine disks via the Container Storage Interface.

Learn how to separate configuration from code using config maps and secrets in Kubernetes, and securely expose environment variables to applications without hardcoding confidential data.

Understand Kubernetes service types—from cluster IP and node port to load balancer and external name—and how kube-proxy, ingress, and network policies route and secure traffic inside and outside the cluster.

Monitor Google Kubernetes Engine with cloud operations for GCC, using native logging and monitoring dashboards and Prometheus integration for external metrics.

Kubernetes health checks include readiness and liveliness probes that determine when a pod is ready to serve traffic and when to restart or replace a failing pod.

Kubernetes logging integrates with cloud logging and monitoring by default in GKE, enabling automatic collection of application logs and exporting to BigQuery, Cloud Storage, or Cloud Pub/Sub.

Leverage Anthos multi-cluster ingress, a Google-managed cloud-based ingress controller for GKE, with a config cluster, fleet, and member clusters to enable shared resources, single IP, multi-regional availability, and proximity-based routing.

Enable consistent communication and data sharing in microservices with the Anthos service mesh, built on Istio, using sidecar proxies to enforce security, authentication, logging, and observability through standardized policies.

Learn how Cloud Run, a stateless container compute service, offers a managed pay-per-use option with gvisor sandbox isolation, autoscaling, and seamless deployment via revisions, including deployment in Anthos.