NAVER Cloud Platform Certified Professional key terms

A physically separate geographic area where NAVER Cloud Platform resources are deployed, containing multiple availability zones and independent infrastructure. Choosing a region affects latency, data residency, service availability, and pricing for workloads.

An isolated data center grouping inside a region with independent power and network paths. Distributing resources across zones improves fault tolerance, and some services only offer certain features in specific zones.

A logically isolated network environment that you define with your own IP address range. Resources such as servers and load balancers launch inside a VPC, and connectivity between VPCs requires explicit configuration.

A segmented IP range within a VPC that groups resources and controls routing. Subnets can be public-facing or private, and their associated route tables determine whether traffic reaches the internet.

A set of rules attached to a subnet that decides where outbound traffic is forwarded. Entries point to local networks, NAT gateways, or internet gateways, and misconfigured routes are a common cause of connectivity failures.

A VPC component that enables communication between resources in public subnets and the internet. It performs address translation for assigned public IPs and is referenced by route tables as a default route target.

A managed service that lets instances in private subnets initiate outbound internet connections without exposing them to inbound traffic. It translates private source addresses to its own public address and scales automatically.

A stateful virtual firewall applied to server instances that allows or denies inbound and outbound traffic by protocol, port, and source or destination. Rules are allow-only, and multiple groups can apply to one instance.

A stateless subnet-level filter that evaluates ordered allow and deny rules for traffic entering or leaving a subnet. Because it is stateless, return traffic must be explicitly permitted, unlike instance-level security groups.

A service that distributes incoming traffic across multiple backend targets to improve availability and throughput. NAVER Cloud Platform offers application and network load balancers with health checks, listeners, and target groups.

A logical collection of backend resources, such as server instances or IP addresses, that a load balancer forwards traffic to. Health checks run against targets, and unhealthy targets are removed from rotation automatically.

A periodic probe that tests whether a backend target is responding correctly, using protocol, port, path, and interval settings. Consecutive failures mark a target unhealthy so traffic stops being routed to it.

A managed domain name service that resolves hostnames to IP addresses and supports record types such as A, CNAME, and TXT. It integrates with platform resources and can be combined with health-checked routing for failover.

A content delivery network that caches static and dynamic content at edge locations close to users, reducing origin load and latency. Caching rules, TTL settings, and purge operations control how content is served.

An encrypted tunnel between an on-premises network and a VPC over the public internet, configured with pre-shared keys and security associations. It provides hybrid connectivity at lower cost than dedicated lines but with variable latency.

A virtual machine instance with selectable vCPU, memory, and disk specifications, launched from an image into a subnet. Billing depends on specification and usage, and instances can be resized or captured as images.

A template containing an operating system and optional software used to create server instances. Custom images capture a configured server so identical instances can be launched repeatedly and consistently.

A shell script or automation payload that runs when a server instance first boots, used to install packages, configure services, or register with monitoring. It supports repeatable provisioning without manual login.

A policy-driven service that adds or removes server instances based on metrics such as CPU utilization or schedules. It works with launch configurations and load balancer target groups to maintain capacity and control cost.

A managed Kubernetes offering that provisions and operates the control plane while you manage worker node pools. It integrates with platform networking, load balancing, and container registry services.

A group of worker nodes within a Kubernetes cluster that share the same instance specification and scaling settings. Multiple pools let you mix instance types and isolate workloads inside a single cluster.

A private repository for storing and versioning container images, with access control and vulnerability scanning. Kubernetes clusters and other runtimes pull images from it using repository credentials.

A highly durable service for storing unstructured data such as files, backups, and media as objects inside buckets. It offers tiered storage classes, lifecycle rules, versioning, and access policies for cost and security control.

Network-attached disk volumes that are mounted to a single server instance and behave like local drives. Volumes persist independently of the instance lifecycle and can be snapshotted for backup or cloning.

A point-in-time copy of a block storage volume that can be used to restore data or create new volumes. Snapshots are incremental and stored durably, making them a core backup mechanism.

A shared file storage service that multiple server instances can mount simultaneously over a network protocol. It suits shared content, web serving, and workloads needing a common filesystem across nodes.

A managed relational database service where the provider handles installation, patching, backup, and failover. You choose instance specifications, configure parameters, and connect from applications inside a VPC.

A read-only database copy that receives changes from a primary instance and serves read-heavy queries. It offloads reporting and scaling pressure, and can be promoted during failover in some configurations.

A managed in-memory data store used for caching, session storage, and fast lookups. It supports replication and persistence options, and reduces database load by absorbing repeated reads.

An identity created under the main account with its own credentials and assigned permissions. Sub accounts enable least-privilege access for people and automation instead of sharing root credentials.

A JSON document defining which actions are allowed or denied on which resources, attached to sub accounts or groups. Policies can be managed by the platform or written as custom statements.

An identity assumed by trusted principals or services to obtain temporary credentials scoped to specific permissions. Roles avoid embedding long-lived keys and are common for cross-service and cross-account access.

A credential pair of identifier and secret used for programmatic API calls and CLI authentication. Keys should be rotated, stored securely, and scoped to the minimum permissions required.

A managed service for creating, rotating, and controlling encryption keys used to protect data at rest. It separates key custody from data storage and records key usage for audit purposes.

A service that collects activity and threat signals, detects suspicious behavior, and raises alerts for investigation. It supports compliance reporting and rapid response to unauthorized access attempts.

An audit service that records control-plane actions such as resource creation, deletion, and permission changes. Logs identify who did what and when, supporting forensics and compliance reviews.

A service that collects metrics and events from platform resources, visualizes them on dashboards, and triggers alerts against thresholds. It helps detect performance degradation and capacity shortages before users notice.

A serverless compute service that runs code in response to events or HTTP requests without managing servers. It scales automatically and bills per invocation and execution time, suiting glue and automation tasks.

A managed front door for APIs that handles routing, authentication, throttling, and usage plans. It decouples clients from backend implementations and can invoke serverless functions or private endpoints.

A managed environment for training, deploying, and serving machine learning models with GPU-backed compute. It supports notebooks and model endpoints so teams avoid building infrastructure for experiments and inference.

A cost strategy where committing to a fixed term or capacity yields lower unit prices than on-demand billing. Comparing total cost of ownership across options guides procurement and workload placement decisions.