Prepare and test your skills

Prepare and test your skills

Cloud revolution means moving from managing physical servers to using automation, managed services, and global infrastructure on demand. Google Cloud changes daily operations, team roles, and spending so businesses can scale faster and try new ideas like machine learning and artificial intelligence (AI).
The four phases of cloud migration are the ordered steps that move workloads from review to ongoing improvement. The journey starts with Assess to inventory apps, dependencies, costs, and performance benchmarks. Then Plan builds the foundation with identity management, organization and project structure, and networking, plus a prioritized migration strategy. Next Deploy moves workloads and data to Google Cloud. Finally Optimize improves performance, scalability, disaster recovery, and cost and opens the door to AI. A weak Plan phase can cause delays, confusion, and downtime.
A horizontal timeline showing the four ordered cloud migration phases: Assess, Plan, Deploy, and Optimize, with key tasks for each phase.
Users send traffic through Cloud Load Balancing to three autoscaling platforms — Compute Engine MIG, GKE and Cloud Run — each emitting CPU, memory and traffic metrics to Cloud Monitoring.
Right-sizing keeps resources matched to demand. Cloud Monitoring shows real-time utilization, Recommender suggests better sizes, and Cloud Build and Cloud Deploy create CI/CD pipelines where automated tests validate changes before production.
Economic and organizational change means spending shifts from upfront hardware to usage-based payment, while teams learn new roles. Consumption-based pricing lets spending rise and fall with demand instead of paying for idle servers. Cloud Billing (opens in a new tab) budgets, labels, and exports to BigQuery help teams track who spends what and forecast costs. Teams also build new skills, use a Cloud Center of Excellence (COE) to standardize practices, and use governance with naming, labels, and tags to prevent sprawl.
Exam tip: In public cloud, Google manages physical security, power, cabling, hardware, and virtualization, while your organization remains responsible for its app resources.
Continue with this study guide: Explain why and how the cloud is revolutionizing businesses.
Core cloud terms explain the shift from owning servers to using on-demand services and data to change the business. Cloud technology runs workloads on public cloud infrastructure, data is the measured facts used for decisions, and digital transformation aligns modernization with business priorities over several years.
Cloud technology is the use of on-demand public cloud services instead of only on-premises servers. A traditional legacy app is often a monolith on Windows Server with SQL Server that is hard to upgrade and scale. A cloud architecture uses managed services for less overhead, containers for mobility, microservices for fast upgrades, and serverless for scaling agility. A common start is cloud-first, where new workloads go to public cloud and existing workloads stay put. When some systems stay on premises, a hybrid architecture (opens in a new tab) connects both sides with VPN or Interconnect so legacy data can support new cloud apps through APIs managed by Apigee.
Cloud-native design builds apps from small, independent parts that scale and recover quickly. A decoupled architecture lets each service upgrade and scale on its own, while a stateless design uses shared storage instead of local data so new instances start fast and survive restarts. Autoscaling policies watch CPU, memory, and queue depth to add or remove resources. Data flows from services into Cloud Monitoring, which collects real-time metrics, logs, and traces, then into BigQuery, which stores and analyzes trends to guide tuning like caching, indexing, and code fixes.
Open approaches reduce dependence on one vendor but solve different problems. Open source software shares code the community can use and improve, while open standards share public rules many vendors follow for interoperability.
| Aspect | Open source software | Open standards |
|---|---|---|
| What it is | Code available for community use | Shared rules followed by multiple vendors |
| Example | Kubernetes | Common specification for data exchange |
| Main benefit | Community innovation and easier hybrid use | Products from different vendors work together |
Exam tip: Open source software is code you can inspect and modify; open standards are rules that different products follow so they can work together.
The difference between traditional and cloud technology is who buys hardware, who maintains it, and how capacity grows. Traditional IT uses fixed upfront spending and manual management, while cloud uses variable spending and elastic, managed services.
The financial shift moves from capital expenditure (CapEx) to operating expenditure (OpEx). In CapEx, the organization buys servers, storage, licenses, power, cooling, and rack space before apps run and must buy for peak demand, so much hardware sits idle. In OpEx, the organization pays for consumed VMs, containers, and storage and releases them when demand drops. Total cost of ownership (TCO) compares both models over one and five years, including maintenance, support, migration, and network transfer costs.
| Model | When cost is paid | How capacity is handled |
|---|---|---|
| Traditional CapEx | Upfront plus ongoing facility costs | Buy for peak and keep idle servers |
| Cloud OpEx | Ongoing based on measured use | Scale up and down with demand |
The shared responsibility model divides work between provider and customer. Google manages physical infrastructure, cooling, and hardware maintenance. The customer still owns operating system patches, drivers, Identity and Access Management (IAM), backup and recovery, high availability, and disaster recovery. A legacy application (opens in a new tab) keeps all layers with the organization, while cloud moves physical work to Google. Cloud Logging and Cloud Monitoring track metrics, logs, and traces to show dependencies and catch failures early.
Elastic scaling means cloud resources grow and shrink automatically without buying servers. On premises, adding power means ordering, delivering, and installing hardware by hand, so teams over-provision as a safety net. In cloud, autoscaling watches CPU, memory, and queue depth and adds instances during peaks and removes them when quiet. Usage data flows from workloads into Cloud Monitoring and BigQuery for forecasting seasonal spikes, product launches, and failover needs for primary and disaster recovery systems.
Exam tip: CapEx means fixed upfront investment in owned data centers, while OpEx means variable payment for cloud resources actually used.
Cloud benefits for transformation are scalability to grow, flexibility to run anywhere, agility to release fast, security to lower risk, cost efficiency to avoid waste, and strategic value to compete faster. Google Cloud provides these together through its global network, managed services, and pricing choices.
Scalability, flexibility, and agility describe how fast a business gets capacity, where workloads run, and how quickly teams change ideas. Scalability lets Cloud Run, Compute Engine, and Firestore grow during peaks and shrink when quiet instead of paying for idle servers. Flexibility lets each workload run on premises, in private hosting, in public cloud, or in hybrid and multicloud combinations, with Kubernetes as a common runtime to reduce lock-in. Agility uses low-risk proofs of concept (PoCs) with clear scope and measurable impact, plus CI/CD with Cloud Build and Cloud Deploy where unit, integration, and end-to-end tests validate changes before release.
Cost efficiency means paying for what is used and lowering the rate with commitments. Consumption-based cost lets capacity follow workload changes, so unused on-premises space, power, and upkeep are avoided. Discounts lower the billing rate further:
| Discount option | What is committed | Best for |
|---|---|---|
| Enterprise discounts | Minimum total spend | Large organization-wide use |
| Committed use discounts (CUDs) | Quantity or spend for 1 or 3 years | Predictable steady use |
| Spot VMs | None, but can be preempted | Fault-tolerant batch jobs |
Spend visibility comes from tags and labels by project or team, budgets and alerts, and dashboards on BigQuery exports. Savings and freed time then fund modernization, better decisions through data, and global reach with automatic mirroring and scaling.
Security and reliability in Google Cloud protect transformation because customers inherit Google-scale controls. The shared model leaves the customer always responsible for data and its access, while Google secures hardware, storage, and network in infrastructure as a service (IaaS) and almost everything except data in software as a service (SaaS). Protection layers include default encryption at rest, TLS in transit to the Google Front End (GFE), Confidential Computing for data in use, defense in depth against DDoS, redundant servers and fiber across regions, and independent audits like ISO/IEC 27001 and SOC 2. Disaster recovery (opens in a new tab) benefits from geographic distribution that limits the impact of local outages.
Exam tip: Choose Spot VMs for interruptible work with no commitment, while CUDs require a one-year or three-year commitment for a lower rate.
Infrastructure models differ by who owns the resources, how they scale, and how they connect. On-premises keeps control, public cloud adds speed and elasticity, private cloud dedicates resources to one org, hybrid cloud (opens in a new tab) links private and public, and multicloud uses more than one provider.
On-premises infrastructure is capacity owned in a private data center where the organization manages hardware, self-hosted databases like MySQL, and hardware load balancing. It fits legacy systems with critical data or workloads that cannot move for legal or technical reasons. Public cloud infrastructure is on-demand capacity from a provider with quick setup and high capacity. It fits new workloads, research, temporary projects, and elastic global apps. Public cloud replaces manual work with Cloud SQL, GKE Autopilot, Cloud Load Balancing, and Google Cloud Armor, and uses autoscaling on CPU, memory, and queue depth to match demand.
A private cloud gives one organization exclusive use of compute, storage, and networking inside its trust boundary. It supports strict privacy, compliance, and data sovereignty for personally identifiable information (PII) but requires buying capacity in advance. A hybrid cloud distributes app components across private and public environments, where network connectivity carries requests and data both ways so the two sides work as one system. Common uses are training AI on on-premises data, adding a cloud disaster recovery site, building cloud dev and test environments, and renting high compute for three to six months to avoid CAPEX.
Multicloud uses more than one cloud service provider (CSP) for sovereignty, unique capabilities, lock-in avoidance, and resilience. If one CSP region fails, traffic flows from users to the second CSP in the same region, but only when both support the required capabilities there. The tradeoff is more complexity:
| Approach | Main strengths | Main tradeoffs |
|---|---|---|
| Single cloud | Less complexity, built-in integrations, CUDs | Lock-in risk, fewer regional options |
| Multicloud | Sovereignty, best services, outage resilience | Harder security, integration, skills, and cost control |
Exam tip: Multicloud resilience depends on capability parity: both CSPs in the region must offer what the app needs.
Google Cloud transformation benefits group into intelligence from data, freedom to choose tools, collaboration across teams, trust in security, and sustainability (opens in a new tab) for lower environmental impact. Together they help organizations decide faster, build faster, and operate responsibly.
Intelligence turns large datasets into actions that improve operations and growth. Data flows from sources like geospatial, weather, supply chain, and logistics systems into AI and data platforms, which produce insights for inventory, compliance, energy use, and new revenue. BigQuery and Looker analyze large datasets for decisions, Gemini Cloud Assist helps across design to troubleshooting, and specialized tools add location insight with Google Earth AI, WeatherNext for forecasts, and Maps route optimization to cut fuel and emissions.
Freedom gives choice over infrastructure and migration pace, while collaboration connects people and workflows. Freedom comes from resizable VMs, live migration and auto-restart in Compute Engine, a private fiber network with redundancy across regions, and migration tools like Migration Center, Migrate to Virtual Machines, Storage Transfer Service, and Transfer Appliance for up to 1 petabyte. Migration follows Assess, then Plan for identity and networking, then Deploy, then Optimize for AI and cost. Collaboration comes from Developer Connect linking source code systems to Google Cloud, plus Google Cloud Professional Services, partners, and system integrators guiding setup and sprints.
Trust protects data through encryption, reliability, and audits, while sustainability lowers energy and emissions. Trust uses default encryption at rest, TLS in transit to GFE, Confidential Computing in a Trusted Execution Environment (TEE), firewalls and GFE filtering for DDoS, and redundant data centers. Compliance is verified by ISO/IEC 27001, 27017, 27018, 27701, SOC 2 and SOC 3, FedRAMP, BSI C5, and MTCS, with Assured Workloads for regulated needs. Sustainability is shared: Google owns sustainability of the cloud with efficient data centers and renewable energy, while the customer owns sustainability in the cloud by right-sizing VMs, using serverless scale-to-zero, and archiving old data.
Exam tip: Google is responsible for sustainability of the cloud, while the customer is responsible for sustainability in the cloud.
Risks of not adopting new technology include slower innovation, weaker customer experience, higher security exposure, and costly fixed operations. Without cloud practices, organizations stay reactive and competitors with automation and analytics move faster.
Competitive loss means delayed releases, missed AI and analytics, and poorer service. Cloud speeds time-to-market with automated rollouts, high-level APIs, fast provisioning, serverless elasticity, and CI/CD with Cloud Build and Cloud Deploy. Without it, manual releases and slow provisioning increase risk per release. Customer experience also suffers without feedback loops from surveys and interviews plus real-time insight from Cloud Monitoring, BigQuery, and Looker. The Google Cloud Adoption Framework rates Learn, Lead, Scale, and Secure from Tactical with no coherent plan, to Strategic, to Transformational where data drives innovation. Delay tends to trap orgs in Tactical.
Outdated systems increase attack surface, complicate audits, and fail more often. In the shared model, Google secures cloud infrastructure while the customer secures data, access, and configurations, with the split varying by IaaS, PaaS, and SaaS. Tools that reduce risk include Cloud VPN and Interconnect for secure hybrid links, Security Command Center for misconfigurations, and Cloud Audit Logs for admin activity. A typical compliance journey runs in order:
Legacy limits come from monolithic .NET apps on Windows and SQL Server that are hard to upgrade and tied to licenses and fixed servers. Capacity must be bought in advance, so spikes overwhelm the system while normal periods waste money. Modern cloud apps use managed services for minimal overhead and independently deployable microservices for fast upgrades and serverless scaling. Automated testing in CI/CD further cuts errors and downtime.
Exam tip: In the shared responsibility model, Google secures the cloud infrastructure, while the customer secures what it puts in the cloud, including data, access, and configurations.
Transformation drivers are business needs for speed, quality, and new capabilities, while challenges are legacy complexity, resistance to change, and skill gaps. Organizations modernize to respond to markets and customers, but must plan for dependencies and culture.
The shift from traditional IT to cloud replaces high upfront CAPEX with fast operational spending that scales automatically. A three-to-six-month research environment with high compute and quick setup shows the pattern: on-demand pricing keeps CAPEX low instead of buying servers for short use. Managed Instance Groups (MIGs), GKE autoscalers, and Cloud Run adjust capacity on CPU, Pod demand, and traffic so teams avoid over-provisioning. A migration strategy (opens in a new tab) then pairs CUDs for steady use, Spot VMs for flexible use, and billing budgets and alerts to control variable spend.
Primary business drivers are market speed, customer quality, competitive data use, and balanced cost. Agility means provisioning quickly, automating rollouts, and using global infrastructure for multi-regional resilience. Quality means faster fixes with cloud dev and test environments and higher availability with a cloud disaster recovery site or multicloud routing when one provider fails. Advantage comes from analytics and AI trained on existing data, plus sovereignty choices like using a provider with a local region for PII. FinOps and unit economics link spending to revenue per customer to separate healthy growth from waste.
Systemic challenges come from app dependencies on specific operating systems, databases, and licenses that hurt integration, performance consistency, and cost visibility across clouds. Cultural challenges come from reluctance to adopt Site Reliability Engineering (SRE) and DevOps, which need leadership buy-in and a Cloud Center of Excellence (COE) to standardize. Maturity moves in order from Tactical with quick returns and little disruption, to Strategic with planned scaling, to Transformational where operations run smoothly and data drives innovation. Closing skill gaps requires inventorying expertise, then training with courses and certifications.
Exam tip: Consider business drivers together, not in isolation, because the final decision depends on the balance of business priorities.
Transformation cloud is Google Cloud’s approach that links modern apps, shared data, connected people, and protected transactions into one model. Modern infrastructure lets teams release quickly, democratized data lets everyone decide from the same truth, and central identity with encryption lets people collaborate safely.
App modernization moves from fixed servers to autoscaled, managed platforms. Legacy monoliths are hard to upgrade and offer little budget agility, while modern apps use containers for portability and microservices for independent upgrades with serverless scaling. The path runs in order from Rehost as-is to Compute Engine VMs, to Replatform with Cloud SQL for SQL Server and containers, to Re-architect into .NET and Linux microservices on managed platforms. Cloud Run autoscales containers by default, bills per request duration, and splits traffic between revisions while receiving events from Pub/Sub. Cloud Build and Cloud Deploy automate build, test, and delivery with vulnerability checks, while Observability with Logging, Monitoring, and Trace tunes performance.
Data democratization makes trusted, current data accessible beyond specialists through one governed platform. BigQuery (opens in a new tab) is the serverless enterprise warehouse where data flows from many sources into tables for fast analysis without managing servers. BigQuery Omni queries data in Amazon S3 or Azure Blob Storage without moving it, while sharing and Sensitive Data Protection govern discovery and masking. Cloud Data Fusion builds code-free pipelines to cleanse and prepare data, and Looker, Connected Sheets, and Data Studio visualize the same trusted version for dashboards and stories.
| Tool | Primary user | Choose when |
|---|---|---|
| Cloud Data Fusion | Business users and developers | Work is integrating and preparing data |
| BigQuery | Data analysts and teams | Work is storing and querying large datasets |
| Looker | Analysts and business teams | Work needs governed dashboards and apps |
People connections let employees and customers reach apps from anywhere with consistent identity, while trusted transactions keep every interaction verifiable. Google Workspace with Cloud Identity centrally controls who can use collaboration and cloud services, then IAM grants least-privilege access to specific resources. Identity-Aware Proxy (IAP) protects internal apps and Identity Platform manages customer sign-on. The customer always secures data and its use, while Google secures hardware, storage, and network for IaaS and almost everything except data for SaaS. Verified access uses OS Login tied to employee lifecycle, organization policies that block external IPs or restrict regions, Access Approval for Google admin access, and encryption with CMEKs in Cloud KMS plus Confidential Computing for data in use.
Exam tip: Rehost moves an application as-is to Compute Engine VMs, while re-architect breaks a monolith into microservices on containers and managed services.
Transformation succeeds when financial flexibility, elastic operations, and governed data reinforce each other. On-demand pricing and autoscaling remove the need to buy for peaks, open tools and phased migration lower the risk of moving, and shared security with democratized analytics lets teams innovate quickly without losing control of cost, compliance, or trust.
Traditional IT uses capital expenditure (CapEx) with upfront payment for owned servers, storage, licenses, power, cooling, and rack space bought for peak demand, while cloud uses operating expenditure (OpEx) with ongoing payment for consumed virtual machines, containers, and storage that scale up and down with demand. Total cost of ownership compares both models over one and five years, including maintenance, support, migration, and network transfer costs.
The four phases are Assess, Plan, Deploy, and Optimize in that order. Assess inventories apps, dependencies, costs, and performance benchmarks, Plan builds the foundation with identity management, organization and project structure, networking, plus a prioritized migration strategy, Deploy moves workloads and data to Google Cloud, and Optimize improves performance, scalability, disaster recovery, and cost and opens the door to artificial intelligence (AI).
Use Spot VMs for fault-tolerant batch jobs with no commitment that can be preempted, and use committed use discounts (CUDs) for predictable steady use. Committed use discounts require a quantity or spend commitment for 1 or 3 years for a lower billing rate, while Spot VMs require no commitment.
Open source software is code available for community use and improvement, with Kubernetes as an example, while open standards are shared public rules followed by multiple vendors for interoperability, such as a common specification for data exchange. Open source supports community innovation and easier hybrid use, while open standards let products from different vendors work together, and both help avoid vendor lock-in.
Last updated