NAVER Cloud Platform Fundamentals and Architecture
Domain 1 - Fundamentals and Architecture. This domain tests placement reasoning: before choosing a product you must decide which region, zone, account, VPC and subnet a workload belongs to, and where the provider's responsibility ends and yours begins.
Placement hierarchy
- Region: a geographic deployment unit. Choose it for latency to users, data-residency rules and service availability.
- Zone: an isolated fault domain inside a region. NCP's Korea region exposes multiple zones (KR-1, KR-2), so highly available designs spread instances across zones.
- Account and Organization: accounts own resources and billing; Organization groups accounts for central governance.
- VPC and subnet: your private network; public subnets route to an Internet Gateway, private subnets route out through a NAT Gateway.
Scope rules
Zone-scoped: Server, Block Storage, Load Balancer nodes. Region-scoped: VPC, Object Storage buckets, managed databases. Global: console, identity and billing. A frequent question is which failure a design survives - one zone, one region, or one account.
Worked example
A Seoul-facing order service must survive a data-centre failure and keep data in Korea. Reasoning: Korea region for residency and latency; one VPC; web servers in two subnets in different zones behind a Load Balancer; managed database with HA; backups copied to a second zone. Putting both web servers in one zone because the subnet is private fails the requirement.
Traps
- Assuming every service is available in every region - verify first.
- Confusing a zone with a subnet: zones are fault domains, subnets are network segments.
- Designing high availability inside a single zone.
- Forgetting that console identity (Sub Account) and API identity (IAM access key) are separate surfaces.