14 Glossary
OLIVE Concepts
| Term | Definition |
|---|---|
| admin / editor / user | The three permission levels OLIVE enforces. admin covers administrative functions for the tenant, including importing an OLIVE Connectivity Bundle, but not ordinary editing; editor creates and changes connectivity definitions and can export a bundle, but cannot import one; user executes what already exists. From OLIVE 1.3 these are genuinely different permissions. A person can hold both editor and admin, but it should be avoided. See Roles In OLIVE. |
| Agent | A system that uses an LLM to decide steps toward a goal (e.g., call tools, fetch data, summarize) and return a final answer. Combines policy (rules), memory/context, and a set of callable tools/connectors. |
| Common Assets Sandbox | Historical. Up to OLIVE 1.2, a dedicated tenant holding data shared by all tenants — templates, presets and Microfrontend components. OLIVE 1.3 has no common-assets deployment; shared content is distributed between ordinary tenants with an OLIVE Connectivity Bundle. If your organization keeps one tenant as its library, ask your Tenant Admin which one it is. |
| Connectable Element | Anything OLIVE can invoke: a Datapoint, a Connector or a Workflow. All are reachable through the Data Orchestrator. |
| Datapoint | The smallest executable unit of connectivity — one well-defined operation, such as a single HTTP request, a transformation or a prompt. Passive until invoked. See Datapoints. |
| Workflow | A Connectable Element that orchestrates others with explicit control logic — branching, iteration, error handling — executed by the Temporal engine. See Connectivity Orchestration with Workflows. |
| Caller | The service that knows how to execute one kind of Datapoint. Registering a Caller is what makes a new datapoint type available. See How to Extend The Modelling Language. |
| Auth | A separate element holding the credentials a Datapoint needs, referenced from datapoint.auth. Only Datapoints can carry it. See Authentication and Secrets Management. |
| Placeholder | ${…} in a definition, resolved at execution time against the namespaces available in that scope. See Placeholder Resolutions. |
ref | The name given to a Connector element so that later elements can read its output as ${ref.<name>}. |
| Connectivity Pattern | A group of Connectable Elements that a Template instantiates together. |
uiConfig | The form definition on an element declaring which parameters a user supplies when a Template is instantiated. |
| Data Orchestrator | The core service that invokes Connectable Elements and tracks the resulting jobs. The public entry point for running connectivity. |
| Connector | A connection between two or more Connectable Elements. A Connector is an OLIVE Connectivity Metamodel element. |
| Deployment | One installed instance of the OLIVE platform. A development deployment is a single instance; a production installation is one shared runtime pool plus one deployment per tenant. Not to be confused with a Kubernetes Deployment (a workload object) or with a release (a version of OLIVE). |
core (tag) | The tag carried by connectivity entries delivered with the foundational package. They are hidden in the editor unless it is opened with debug=true, they define how OLIVE's own services behave, and from OLIVE 1.3 only a user with the admin role may change them. |
| Development deployment | An OLIVE installation with one tenant, no users and no authentication, used to build and test connectivity. One namespace holds every service and store, so there is no tenant to choose. It can create, edit and execute connectivity exactly as a production deployment can. Deployment mode single-no-auth. Solutions built here are delivered to a production deployment with an OLIVE Connectivity Bundle. |
| Extension | An OLIVE service providing functionality that is not part of the OLIVE core — for example AI callers, workflow workers or format converters. |
| OLIVE Connectivity Bundle (OCB) | A ZIP package containing connectivity elements (datapoints, connectors, authentication entries, templates and presets), the Microfrontend components, and a manifest recording when the export was made and with which export version. Used to deliver a solution to another deployment, and as a backup. See How to Package a Use Case Solution. |
| Production deployment | A multi-tenant OLIVE installation with authentication and role-based access control. Each tenant has its own storage and Keycloak realm. It creates, edits and executes connectivity exactly as a development deployment does; what differs is that tenants, users and roles exist and are enforced. Deployment modes multi-shared and multi-tenant. |
| Release | A version of the OLIVE Platform — 1.2, 1.3 — as described in the release notes. In this documentation release always means a version; an installed instance is a deployment. |
| Sandbox | The single tenant of a development deployment — the one working environment that holds all its connectivity content and that the connectivity editor operates on. Because a development deployment has exactly one, there is nothing to select. The equivalent in a production deployment is a tenant, and there are many. |
| Template | A parameterizable collection of Connectable Elements and Microfrontends that can be instantiated and customized for a single tenant. A Template is an OLIVE Connectivity Metamodel element. |
| Tenant | An isolated customer or organizational unit using the OLIVE platform. A production deployment has many; a development deployment has one, called the sandbox. |
| Tenant Admin | The role responsible for one tenant: onboarding scenarios, importing bundles, managing users and roles, and reviewing connectivity and UI components before they are enabled. The escalation point for anything inside a tenant. See Roles In OLIVE. |
| Operator | The role responsible for the platform itself: the cluster, the Helm values, the common infrastructure, and onboarding and offboarding tenants. The escalation point for anything that needs a change to the deployment rather than to a tenant. See Roles In OLIVE. |
AI & Language Concepts
| Term | Definition |
|---|---|
| LLM (Large Language Model) | A probabilistic, transformer-based model trained on large text corpora to predict tokens. Strengths: language understanding, generation, summarization, reasoning. Limitations: may produce incorrect or overconfident answers; works best with grounding (retrieval/tools) and clear prompts. |
| Prompt | A set of instructions and context given to a model to shape its output. Can include roles, goals, constraints, examples, and tool-use guidance. Good prompts are clear, scoped, and reference authoritative data; poor prompts increase hallucinations. |
| Vector Database | A database optimized for storing and searching vector embeddings, enabling similarity search and semantic retrieval. Used for LLM grounding, recommendations, and semantic search. |
| Semantic Search | Search technique that uses vector similarity (semantic meaning) rather than exact keyword matching to find relevant results. Often powered by embeddings from LLMs. |
Architecture & Deployment
| Term | Definition |
|---|---|
| Microservice | A small, independently deployable service that performs a specific business function and communicates via APIs. Enables scalability and maintainability. |
| Microfrontend | A modular frontend architecture where UI components or sections are built and deployed independently, often by separate teams. |
| Piral | A framework for developing modular frontends (Microfrontends) using "Pilets" that can be composed dynamically at runtime. |
| Pilet | A self-contained Microfrontend module built with Piral that can be independently developed, deployed, and loaded into a host application. |
| Cloud Agnostic | Designed to run across multiple cloud providers (e.g., AWS, Azure, GCP) without dependency on proprietary services. |
| High Availability (HA) | A system design principle ensuring minimal downtime and continuous service operation through redundancy and failover mechanisms. |
Infrastructure & Deployment Tools
| Term | Definition |
|---|---|
| Argo CD | A declarative, GitOps-based continuous delivery tool for Kubernetes that automates application deployment and synchronization with Git repositories. |
| Helm | A package manager for Kubernetes applications, allowing configuration and deployment using reusable charts. |
| Chart | A Helm packaging format containing all Kubernetes resource definitions needed to deploy an application. |
| Helm release | Helm's own term for one installed instance of a chart, as listed by helm list. An OLIVE deployment is made up of one or more Helm releases, so this term is used only where the Helm object itself is meant. |
| Deployment (Kubernetes) | A Kubernetes workload object that keeps a set of identical pods running. Written Deployment when the Kubernetes resource is meant, to distinguish it from an OLIVE deployment. |
| Kubernetes | An open-source container orchestration system for automating deployment, scaling, and management of containerized applications. |
| RKE2 | A lightweight, secure Kubernetes distribution by Rancher (also known as RKE Government), designed for production and edge environments. |
| Container | A lightweight, isolated runtime environment that packages an application with its dependencies, ensuring consistent behavior across environments. |
| NGINX | A high-performance web server, reverse proxy, and load balancer formerly used as an ingress controller in Kubernetes. End of Security support. |
| Traefik | A modern, cloud-native ingress controller for Kubernetes that provides dynamic routing, load balancing, and automatic TLS/SSL certificate management. |
| Ingress | A Kubernetes resource that manages external access to services, typically via HTTP/HTTPS routing rules. |
| CertManager | A Kubernetes add-on for automatically issuing and renewing TLS/SSL certificates from certificate authorities like Let's Encrypt. |
| HashiCorp Vault | A tool for securely managing secrets, credentials, and encryption keys used by applications and infrastructure. |
| GitLab | A DevOps platform for source code management, CI/CD pipelines, and infrastructure automation. |
| OCI (Open Container Initiative) | A set of open standards for container image formats and runtimes, ensuring interoperability across container tools. |
| AMD | Processor architecture type (Advanced Micro Devices) commonly used in general-purpose servers and desktops. |
| ARM | Energy-efficient processor architecture widely used in mobile, embedded, and increasingly cloud environments. |
Monitoring, Logging & Observability
| Term | Definition |
|---|---|
| Prometheus | An open-source monitoring system that collects metrics from targets, stores them in a time-series database, and supports alerting. |
| Grafana | A visualization and analytics platform for dashboards based on data from sources like Prometheus and Loki. |
| Grafana Alloy | A unified observability agent combining metrics, logs, and traces collection for Grafana Cloud or on-prem deployments. |
| Grafana Tempo | A distributed tracing backend that stores and visualizes traces for debugging and performance analysis. |
| Grafana Alerting | A centralized system in Grafana for managing alerts from various data sources. |
| Promtail | A log collection agent that ships logs to Loki, typically paired with Grafana for log visualization. |
| Loki | A log aggregation system by Grafana Labs, optimized for cost-efficient querying by labels instead of full-text indexing. |
| OpenTelemetry | A standard for collecting telemetry data (traces, metrics, logs) from applications for consistent observability across tools. |
Networking & Communication Protocols
| Term | Definition |
|---|---|
| HTTP (Hypertext Transfer Protocol) | The foundational protocol for web communication, defining how requests and responses are exchanged between clients and servers. |
| REST (Representational State Transfer) | An architectural style for designing networked APIs using HTTP verbs (GET, POST, etc.) and stateless interactions. |
| JSON (JavaScript Object Notation) | A lightweight data format used for data exchange between systems, particularly in APIs. |
| SMTP (Simple Mail Transfer Protocol) | A standard protocol for sending and relaying email messages. |
Authentication & Authorization
| Term | Definition |
|---|---|
| Basic Auth | A simple authentication method using a username and password encoded in the HTTP header. |
| Bearer Token | A security token used in HTTP authentication, usually representing a user or service identity in API requests. |
| OAuth2 | An authorization framework enabling applications to access user data on another service without sharing credentials. |
| OIDC (OpenID Connect) | An identity layer on top of OAuth2 providing authentication and user identity verification. |
| AWSv4 | The AWS Signature Version 4 signing process used to authenticate requests to AWS APIs. Deprecated — no longer supported from OLIVE 1.3. |
| Keycloak | An open-source identity and access management system providing SSO, OAuth2, and OIDC support for applications. |
Cloud & External Systems
| Term | Definition |
|---|---|
| AWS (Amazon Web Services) | A comprehensive cloud computing platform offering compute, storage, networking, and AI services. |
| CMDB (Configuration Management Database) | A database containing information about IT assets and their relationships, used for configuration and change management. |
| OpenAI | An AI research and deployment company providing advanced models like GPT for natural language and reasoning tasks. |
| MS Teams | Microsoft's collaboration and communication platform integrating chat, meetings, and app extensions. |
Related Glossary
For the related glossary about cloud technologies and deployment see the Laboratory Setup "From Blech to Cloud" documentation.