Hikube Glossary
Find here the definitions of terms and concepts used throughout the Hikube documentation.
| Term | Definition | Documentation |
|---|---|---|
| Add-on / Plugin | An extension that can be enabled on a Kubernetes cluster (cert-manager, ingress-nginx, monitoring, etc.) to add functionality without manual configuration. | Kubernetes API Reference |
| AMQP | Advanced Message Queuing Protocol. A standard messaging protocol used notably by RabbitMQ for communication between applications. | RabbitMQ Overview |
| ClickHouse Keeper | A distributed consensus service built into ClickHouse, used for cluster node coordination (alternative to ZooKeeper). | ClickHouse Overview |
| Cloud-init | An automatic initialization tool for virtual machines at first boot. It allows configuring users, packages, scripts, and networking via a YAML file. | Compute API Reference |
| CNI (Container Network Interface) | A standard that defines network management for containers in a Kubernetes cluster. Hikube uses Cilium as the default CNI. | Kubernetes Overview |
| Control Plane | The set of components that manage the state of the Kubernetes cluster (API server, scheduler, controller manager). The number of replicas determines high availability. | Kubernetes API Reference |
| Golden Image | A preconfigured base image for virtual machines, optimized for a given operating system (Ubuntu, Rocky Linux, etc.). | Compute Overview |
| Ingress / IngressClass | A Kubernetes resource that manages external HTTP/HTTPS access to cluster services. IngressClass defines the controller used (nginx, traefik, etc.). | Kubernetes API Reference |
| JetStream | A streaming and persistence system built into NATS, enabling durable message storage, replay, and guaranteed delivery. | NATS Overview |
| Kubeconfig | A configuration file containing access information for a Kubernetes cluster (server URL, certificates, tokens). Required to use kubectl. | Kubernetes Quick Start |
| Namespace | A logical space within a Kubernetes cluster used to isolate and organize resources. Each tenant has dedicated namespaces. | Key Concepts |
| NodeGroup | A group of worker nodes in a Kubernetes cluster with shared characteristics (instance type, min/max scaling, roles). Allows adapting resources to different workloads. | Kubernetes API Reference |
| Operator | A Kubernetes pattern that automates the management of complex applications. Hikube uses specialized operators: Spotahome (Redis), CloudNativePG (PostgreSQL), etc. | Redis Overview |
| PVC (PersistentVolumeClaim) | A persistent storage request in Kubernetes. Allows pods to retain data beyond their lifecycle. The size is defined by the size parameter. | Kubernetes Quick Start |
| Quorum Queues | A RabbitMQ queue type based on Raft consensus, providing strong replication and fault tolerance for critical messages. | RabbitMQ Overview |
| ResourcesPreset | A predefined resource profile (nano, micro, small, medium, large, xlarge, 2xlarge) that simplifies CPU/memory allocation for managed services. | Redis API Reference |
| Sentinel | A Redis component that continuously monitors the cluster state, detects master failures, and automatically orchestrates failover to a replica. | Redis Overview |
| Shard / Replica | A shard is a horizontal data partition (used by ClickHouse). A replica is a copy of the data for high availability and fault tolerance. | ClickHouse Overview |
| StorageClass | Defines the type of storage used for persistent volumes. replicated provides data replication across multiple datacenters for high availability. | Kubernetes API Reference |
| Tenant / Sub-tenant | An isolated and secure environment within Hikube. A tenant can contain sub-tenants to separate environments (production, staging, development). | Key Concepts |