Terminology

Version: 17.07

Container

A lightweight yet self-contained, fully isolated process with its own environment, filesystem, libraries, system tools etc. (except for the OS kernel) providing a platform- and environment-agnostic runtime for its child processes (ensuring that it runs the same, regardless of where, when or how it is run)

Container Image

A container (generally a Docker container) is based on a Docker image. A container image provides a basis for defining a container, with its own filesystem and environment configurations. IPS generally uses this term in reference to candidate container images for launching IPS-Worker instances.

Container Orchestration Framework

A platform such as Docker Swarm, Kubernetes, OpenShift or Amazon EC2 Container Service that facilitates spawning, controlling and maintaining one or more clusters (groups) of containers, often with advanced features such as networking and routing, monitoring and auto-healing, security and access control, metrics, versioning, etc.

IPS (Integration Platform Server)

A higher-level platform that leverages features of a container orchestration framework to facilitate easy, convenient and reliable deployment of integration solutions on top of containers, together with simplified management, comprehensive metrics and statistics and detailed monitoring with automatic recovery and troubleshooting

Master

Represents a "controlling" host machine (e.g. a Kubernetes master node) which holds the management and monitoring components of the underlying orchestration platform

Node

Represents a "worker" host machine (e.g. a Kubernetes node/minion) which provisions actual runtime components of the deployed application upon the underlying orchestration platform. In case of IPS, these runtime components include both the IPS management components (e.g. Configuration Server and IPS-Web) and the actual worker (IPS-Worker) instances

Instance/Pod

A collection of one or more containers which is created, scheduled and managed as a single deployable unit by the orchestration framework (such as a Kubernetes Pod). IPS-Worker, for example, is deployed as a single-container pod, whereas the kube-dns pod in the demo IPS installer is a multi-container pod.

Configuration Server

A centralised server that acts as an endpoint for serving projects, custom artifacts and other deployment parameters and configurations to IPS-Worker instances

IPS-Web

The management console of IPS, which provides a unified dashboard for viewing and managing all your clusters, projects, nodes, worker instances, configurations, logs and statistics

IPS-Worker

A tailored UltraESB-X instance hosted in IPS, which acts as a lightweight container for hosting and running integration projects developed based on the AdroitLogic Project-X framework

Replication

Existence of multiple runtime copies of an instance configuration (e.g. IPS-Worker), all with identical configurations, generally with load-balancing for distributing incoming workload across copies

Cluster

A group (replication) of IPS-Worker instances, based on a configurable container image and at a desired replication level

Project

An integration project that can be deployed inside an IPS-Worker instance

Integration Flow

A specific message processing flow in an integration project

Configuration Artifact

A custom library (e.g. third-party JAR file) or configuration file (e.g. logging customization, XML configurator for a third-party library) that can be included in a deployment in order to modify or enhance the default behaviour of a project or an IPS-Worker instance itself

Deployment

Based on the context, this may refer to

  • a set of configurations (projects, custom artifacts and other parameters) defining the runtime state of a cluster

  • an orchestration platform-level governance entity for pod replicas (such as a Kubernetes Deployment)

Service

An abstraction that encapsulates and exposes a set of pods (replicas) (e.g. by mapping their message-accepting network ports from the local (container) network on to the external network) along with a policy by which to access them (e.g. round-robin load balancing across the replicas)

In this topic
In this topic
Contact Us