Prepare and test your skills

Prepare and test your skills

Cloud transformation is the move from owning and maintaining on-premises servers to running applications and operations on a public cloud such as Google Cloud (opens in a new tab). Organizations make this move for business reasons that justify the investment and for technical reasons that unlock new capabilities. Together, these drivers push companies away from rigid legacy hardware toward agile, scalable environments that are paid for as they are used.
Business drivers for cloud adoption are the business outcomes that justify moving to the cloud. The core business drivers are improvements in cost, speed, resilience, and innovation, and the final decision depends on balancing priorities rather than on any single driver alone. Temporary needs can also drive cloud use when a company needs capacity quickly without buying hardware.
Short-term examples include large-scale projects that need infrastructure for three to six months, multi-year programs that need a hybrid setup during transition, and company mergers where two different IT systems must be integrated quickly.
Technical drivers for cloud adoption are the technology gains or limits that shape architecture choices. The most common temporary driver is simply to facilitate migration, because enterprises must move applications and data in stages over a hybrid connection. Other technical drivers focus on building capabilities that are difficult or slow to create in existing data centers.
When these technical drivers reduce delay, simplify operations, or unlock new services, they directly support the business drivers of agility, lower cost, and innovation.
The shift from legacy infrastructure to agile, scalable environments is the move from fixed on-premises systems to flexible cloud architectures. A typical legacy system is a monolithic application built on the .NET Framework, hosted on an on-premises Windows server, and connected to an on-premises SQL Server. Legacy systems force the team to manage physical servers, pay Windows licensing costs, struggle to upgrade a large monolith, and work with fixed capacity that is hard to plan, budget, and scale.
| Dimension | Legacy application | Modern cloud-based application |
|---|---|---|
| Management | Team manages on-premises servers, Windows, and SQL Server | Minimal overhead by integrating with managed services |
| Mobility | Limited deployment tied to Windows with licensing costs | Full mobility with containers and no Windows dependency |
| Updates | Difficult to upgrade a monolithic codebase | High-velocity upgrades through independently deployable microservices |
| Scale and budget | Little agility to plan, budget, and scale fixed hardware | Full agility to scale and budget with elastic and serverless architecture |
Modern cloud architectures remove legacy limits by using managed services, containers, independently deployable microservices, and serverless scaling. Capacity can grow or shrink with demand, and new features can be released without rebuilding the whole system, which makes the cloud approach more cost-effective, efficient, and resilient.
The shift from capital expenditure (CapEx) to operational expenditure (OpEx) is the change from buying hardware upfront to paying only for cloud resources as they are used. In an on-premises environment, the organization owns everything, including cooling, physical security, hardware maintenance, virtualization, and applications. In a private hosting environment, a provider manages physical security, power, and cabling while the customer still manages virtualization and applications. In a public cloud environment, the provider also manages hardware and the virtualization platform, so the customer focuses on applications and can hand off even more work with fully managed services.
This ownership shift makes temporary and research environments practical because teams can provision large distributed infrastructure for a limited time and shut it down when the project ends. Teams can still use on-premises data over a secure connection while using on-demand pricing, high compute capacity, and quick setup in the cloud.
Exam tip: CapEx means large upfront spending to buy servers you own, while OpEx means on-demand cloud billing for what you use; the cloud driver is lowest possible CapEx with fast provisioning.
Migration is the act of moving a workload from on-premises, private hosting, or another public cloud to Google Cloud (opens in a new tab). Modernization is the act of changing an application so it uses cloud-optimized capabilities such as managed services, microservices, containers, and serverless computing. The two are distinct but linked, because using the public cloud can make modernization easier, and modernizing helps an organization get more value from the cloud.
Migration approaches describe how much a workload changes as it moves to the cloud. The six common paths range from moving with almost no change to replacing the application with a cloud service. Choosing a path depends on dependencies, cost, skills, and whether the current technology can run in the cloud.
Lift-and-shift and re-architecting differ in how much they change an application and what they aim to achieve. Lift-and-shift focuses on moving quickly with minimal change, while re-architecting focuses on redesigning for cloud-native behavior. The following comparison shows the tradeoff between speed and cloud benefit.
| Dimension | Lift-and-shift (rehost) | Re-architecting |
|---|---|---|
| What changes | Workload moves as-is with minimal changes | Application code changes how it functions to use cloud-optimized properties |
| Primary goal | Move quickly and optimize from an infrastructure perspective | Gain scalability, security, agility, and cloud-native design |
| Typical effort | Shorter and simpler than refactoring | More complex, takes more time and effort, and requires testing |
| Example | Move Windows Server and SQL Server instances to Compute Engine virtual machine (VM) instances | Turn a monolithic VM-based app into independent microservices on GKE or Cloud Run |
Exam tip: Lift-and-shift moves a workload to the cloud with minimal changes, while re-architecting changes the application's code and design to use cloud-native features.
How migration and modernization work together is a phased cycle where an early move makes later improvements easier. The primary goals of modernizing are greater agility, lower infrastructure and operations costs, and higher reliability and resiliency. Because it may not be feasible to modernize every application at once, teams often move first and improve after.
Lifting and shifting first helps optimize from an infrastructure perspective. After applications run in the cloud, it is easier to integrate cloud services and adopt cloud-first architectures. Modernization can also require procedural and cultural changes, and the first step to implementing SRE is to get leadership buy-in.
Each application following a different path means an organization does not have to choose one migration approach for everything. A hybrid and multicloud strategy lets each workload or application stack follow the path that fits its business and technical requirements. Even components of the same stack can follow different approaches.
On-premises frontend VMs, MySQL and hardware LB/WAF each take a different path to Google Cloud: refactor to GKE Autopilot, replatform to Cloud SQL, and replace with Cloud Load Balancing and Cloud Armor.
For example, a backend on-premises database can be replatformed from self-hosted MySQL to Cloud SQL. Frontend virtual machines can be refactored to run on containers using GKE Autopilot, where Google manages cluster configuration including nodes, scaling, security, and other preconfigured settings. On-premises hardware load balancing and web application firewall (WAF) capabilities can be replaced with Cloud Load Balancing and Google Cloud Armor.
Cloud-native features that modernization unlocks are the scalable, managed capabilities that replace self-managed infrastructure. A monolithic VM-based application can be refactored into several independent microservices using managed container services like Google Kubernetes Engine (GKE) or Cloud Run. Serverless architecture then provides full agility to scale and budget without managing servers.
Managed services shift platform management to Google Cloud and reduce overhead. Examples include using Cloud Storage instead of network attached storage (NAS), Pub/Sub instead of self-hosting MSMQ, RabbitMQ, or Kafka, Identity-Aware Proxy (IAP) instead of coding authentication into the application, and Cloud Key Management Service (Cloud KMS) and Secret Manager instead of storing secrets locally. A hybrid architecture can also modernize legacy services with API interfaces, and an API management platform like Apigee can add security, analytics, and scalability with minimal application changes.
Application migration pathways are the strategies an organization uses to move each application to the cloud. A migration is a journey, and different applications can follow different paths based on business goals, technical constraints, and how much cloud integration is desired. Common modernization goals are greater agility, lower infrastructure and operations costs, and higher reliability and resiliency.
Migration pathways in Google Cloud (opens in a new tab) describe how much an application changes as it moves. The main pathways range from moving with almost no change to completely replacing the workload. The right pathway balances speed, cost, risk, and desired cloud integration.
Exam tip: Rehost moves an application with little or no change, replatform makes limited changes to use managed services, and refactor changes the workload to improve it for cloud capabilities.
Decision criteria for common pathways match the workload's condition and goals to the right level of change. Rehost and replatform favor speed and simplicity, while refactor and rebuild favor deeper optimization at higher cost and effort. The choice should weigh measurable business effect, migration risk, and technical difficulty.
| Pathway | Best fit workloads | Main tradeoff |
|---|---|---|
| Rehost | Few dependencies; not worth refactoring; third-party software | Fast and simple, but limited cloud optimization |
| Replatform | Limited changes needed to use managed services | Some cloud benefit with fewer changes than refactor |
| Refactor | Dependencies to untangle; unsupported OS, hardware, or database; inefficient compute or storage; manual deployment | More cloud benefit, but longer and requires new skills |
| Rebuild | No longer meets requirements; replaceable by similar apps; third-party technology at end of life; uneconomical licenses | Full optimization and technical debt removal, but longest and requires rewriting |
Phased and mixed migration means moving in stages and mixing approaches rather than changing everything at once. A common sequence is to start with lift-and-shift or replatforming, then refactor or rearchitect as the next step. Lifting and shifting first optimizes from an infrastructure perspective, and after applications run in the cloud it is easier to integrate cloud services with cloud-first architectures. Applications in the cloud can still communicate with other environments over a hybrid network connection.
Different parts of the same stack can follow different pathways:
A large monolithic VM-based application can also be refactored into independent microservices on GKE or Cloud Run.
The migration journey and planning process is the ordered set of phases an organization follows to migrate applications to Google Cloud. Discovery and assessment should be ongoing, and risk assessment should run in parallel. Starting with a workload that is not business-critical, has minimal dependencies, and can serve as a blueprint helps establish patterns for later waves.
A process flow showing the four phases of the migration journey to Google Cloud — Assess, Plan, Deploy, and Optimize — connected in order by arrows, with a loop back from Optimize to Assess and a note about starting with a low-risk, minimally dependent workload.
A migration wave is a group of applications that share characteristics or interdependencies. Standalone applications and databases are good candidates for a first wave because they have low external dependencies. Within a wave, applications are divided into move groups, which are infrastructure resources and workloads that must migrate together. Research and development environments are typically migrated first, followed by pre-production and production.
Choosing a migration pathway means matching each workload to business goals, technical constraints, and desired cloud integration. After assessing workloads, an organization can prioritize them and define migration waves and approaches, then re-evaluate after the first cloud deployments because assessments can change over time. The decision considers measurable effect, risk, and difficulty.
Modernization and organizational readiness describe the technical and cultural changes needed to gain lasting cloud value. Modernization aims for agility, lower infrastructure and operations costs, and higher reliability and resiliency. It can require adopting Site Reliability Engineering (SRE) or DevOps principles, and it may need to extend to private environments in a hybrid setup.
The Google Cloud Adoption Framework helps assess readiness for Google Cloud. The framework evaluates four themes: Learn, Lead, Scale, and Secure. For each theme, an organization can be in one of three phases: Tactical, Strategic, or Transformational. This structure helps identify current capabilities and the gaps to fill before and during migration, since SRE is more of a transformation process than a technical challenge.
Migration is the act of moving a workload from on-premises, private hosting, or another public cloud to Google Cloud, while modernization is the act of changing an application so it uses cloud-optimized capabilities such as managed services, microservices, containers, and serverless computing. The two are distinct but linked because using the public cloud can make modernization easier, and modernizing helps an organization get more value from the cloud.
Rehost (lift and shift) moves a workload as-is with minimal changes for speed and should be chosen when the workload has few dependencies, is not worth refactoring, cannot be refactored before migration, or is based on third-party software. Rearchitecting changes how application code functions to optimize for scalability, security, and agility but is more complex, takes more time and effort, and requires testing.
The migration journey runs in order through Assess, Plan, Deploy, and Optimize. Assess inventories applications and environments, identifies dependencies and requirements, calculates total cost of ownership, and establishes performance benchmarks, while Plan creates cloud infrastructure, identity, organization and networking and defines migration waves, move groups, and success criteria. Deploy executes migration iteratively in waves and tests requirements before moving to production, and Optimize moves the application to production and uses Cloud Monitoring to gain insights into performance, availability, and health.
An enterprise is evaluating a migration from its traditional on-premises data center to Google Cloud. The executive team wants to eliminate multi-year hardware procurement cycles, align infrastructure costs with actual customer demand, and empower development teams to rapidly provision compute resources as needs change.
Which pair of business and technical drivers best justifies the organization's cloud transformation?
Last updated