CDL digital transformation covers approximately 10-15% of the exam. Understanding why businesses move to cloud, core concepts, and service models is essential.
Why Businesses Move to Cloud
Just like companies buy electricity from a utility rather than running their own generators, cloud transforms IT infrastructure into a service. The key motivations are:
CapEx to OpEx shift: Traditional IT requires large upfront server purchases (Capital Expenditure). Cloud shifts this to pay-as-you-go (Operational Expenditure) — you only pay for what you use. A retailer can scale up 10x for Black Friday and scale back down without wasted capacity.
TCO reduction: Running your own servers involves hidden costs beyond hardware — electricity, cooling, data center space, networking equipment, and maintenance staff. Total Cost of Ownership (TCO) drops significantly when Google absorbs these overhead costs.
Global scale and innovation speed: Deploying globally used to require building data centers in each region. With Google Cloud, you can launch in a new region in minutes. New ideas can be experimented with in minutes instead of months.
Core Cloud Characteristics
| Characteristic | Core Meaning | Example | |---------------|-------------|---------| | Scalability | Ability to adjust resources | Scale from 2 to 20 servers | | Elasticity | Auto scale up/down | Auto-add servers during traffic spikes | | Reliability | Continue operating despite failures | Failover to another region | | Agility | Fast provisioning | 20 servers ready in 5 minutes |
Scalability is the ability to scale. Elasticity is scalability with automation — the system scales itself based on demand, like a rubber band that stretches and bounces back.
Cloud Deployment Models
| Model | Ownership | Key Trait | Best For | |-------|-----------|-----------|----------| | Public Cloud | Cloud provider | Low upfront cost, shared infra | Startups, fast scaling | | Private Cloud | Organization | Full control, high security | Finance, healthcare, regulated industries | | Hybrid Cloud | Mixed | Flexibility + security | Orgs with existing infra | | Multi-cloud | Multiple providers | Avoid vendor lock-in | Optimizing specific workloads |
Hybrid Cloud connects public and private clouds — keep sensitive data on-premises while running web servers on Google Cloud. Multi-cloud uses multiple public clouds simultaneously to avoid vendor dependency.
Google Cloud Infrastructure: Regions and Zones
Regions are independent geographic areas (e.g., = Seoul). Data in one region is isolated from disasters in other regions. Important for data sovereignty regulations.
Zones are independent data centers within a region (e.g., , , ). High-availability apps deploy across multiple zones. Think of regions as countries and zones as cities within them.
IaaS, PaaS, SaaS
Think of it like pizza: making from scratch at home (on-premises), buying ingredients delivered (IaaS), ordering delivery (PaaS), eating at a restaurant (SaaS).
| Model | Customer Manages | Google Manages | Example | |-------|-----------------|----------------|---------| | IaaS | OS, middleware, apps, data | Virtualization, servers, storage, network | Compute Engine | | PaaS | Application code, data | OS, runtime, middleware, all infra | App Engine, Cloud Run | | SaaS | Data, access settings | Everything else | Google Workspace |
IaaS gives maximum control but maximum responsibility. PaaS lets developers focus on code, not infrastructure. SaaS requires no management at all — just log in and use.
Shared Responsibility Model
Security in the cloud is a shared duty between Google and the customer. Think of it like an apartment building: the landlord (Google) secures the building exterior and elevators, while the tenant (customer) secures their front door and belongings.
| Responsibility Area | IaaS | PaaS | SaaS | |--------------------|------|------|------| | Data | Customer | Customer | Customer | | Application | Customer | Customer | Google | | OS | Customer | Google | Google | | Virtualization | Google | Google | Google | | Physical security | Google | Google | Google |
As you move from IaaS → PaaS → SaaS, customer responsibility decreases and Google's responsibility increases.
Exam Key Points
"Pay only for what you use instead of buying servers" -- CapEx → OpEx
"Total cost including hardware, power, staff, space" -- TCO
"Auto scale up/down with demand" -- Elasticity
"Fast resource provisioning, quick experiments" -- Agility
"Service continues despite failures" -- Reliability
"Independent geographic area, data sovereignty" -- Region
"Independent data center within a region" -- Zone
"Manage OS yourself, maximum control" -- IaaS (Compute Engine)
"Deploy code only, Google manages the rest" -- PaaS (App Engine, Cloud Run)
"Just log in and use, no management needed" -- SaaS (Google Workspace)
"Physical security always Google's responsibility" -- Shared Responsibility Model
"Data and access control always customer's responsibility" -- Shared Responsibility Model
"Public + private cloud combined" -- Hybrid Cloud
"Multiple cloud providers simultaneously" -- Multi-cloud