Glossary

Glossary

This glossary provides definitions for key terms and concepts used throughout Simba Intelligence documentation, user interfaces, and technical architecture. Terms are organized alphabetically with cross-references to related concepts.

A

Admin Overlay A modal interface component that provides access to administrative functions within Simba Intelligence. Accessible only to users with supervisor permissions, it includes user management, license information, and system configuration options.

AI Provider External artificial intelligence services that power Simba Intelligence’s natural language capabilities. Supported providers include Google Vertex AI, OpenAI/Azure OpenAI, and AWS Bedrock.

AI Semantic Cache An intelligent caching system that stores and retrieves query responses based on semantic similarity rather than exact string matching. Uses vector embeddings to identify similar queries and reduce AI provider API calls.

API Key A unique authentication token that enables programmatic access to Simba Intelligence’s REST API. Users can generate and manage their own API keys through the web interface.

Authentication Token A temporary security token issued after successful user login that grants access to Simba Intelligence services. Tokens automatically refresh during active sessions and expire after a configurable timeout period.

B

Background Tasks Long-running operations executed asynchronously by Celery workers, such as AI processing, data source creation, and system maintenance tasks. These tasks run separately from the web interface to avoid blocking user interactions.

Bearer Token An authentication method where the API key or session token is included in the HTTP Authorization header as “Bearer [token]”. Used for both web session authentication and API access.

C

Capability A specific AI function that can be enabled for an LLM provider configuration, such as chat (natural language understanding), embeddings (semantic search), or vision (image analysis).

Celery The distributed task queue system used by Simba Intelligence for background processing. Includes Celery Workers (task processors) and Celery Beat (task scheduler).

Celery Beat The scheduler component that manages periodic and scheduled tasks in Simba Intelligence, such as cache cleanup, health monitoring, and maintenance operations.

Celery Worker Background service processes that execute queued tasks such as AI processing, data operations, and system maintenance. Can be scaled horizontally based on workload.

Composer Service The data discovery and query engine component that manages data source metadata, query execution, and integration with External Data Connectors. Provides the /discovery interface for data management.

Configuration Management The system for managing application settings, credentials, and deployment parameters across different environments using environment variables, configuration files, or Helm values.

Consul The service discovery system used by Composer services and External Data Connectors to register themselves and discover other services in the ecosystem.

D

Data Agent An AI-powered interface that automatically creates data sources based on natural language descriptions or dashboard image analysis. Users describe their data needs, and the AI agent handles technical configuration.

Data Connector A configured connection to an external database or data source. Data connectors enable Simba Intelligence to access and query external systems like SQL Server, PostgreSQL, or Snowflake.

Data Source A configured interface to specific tables, views, or datasets within a connected database. Created either manually through the Data Agent or automatically via AI analysis.

Docker Compose A tool for defining and running multi-container Docker applications. Simba Intelligence uses Docker Compose for development environments and small-scale deployments.

E

EDC (External Data Connector) A specialized microservice that translates between Simba Intelligence’s unified data interface and specific database systems. Each database type (PostgreSQL, SQL Server, Snowflake) has its own EDC.

Embedding A numerical vector representation of text content used for semantic similarity comparison. Generated by AI providers and used in semantic caching and query similarity detection.

F

Flask The Python web framework that powers Simba Intelligence’s backend API services. Provides REST endpoints, authentication, and business logic processing.

G

Gunicorn The Python WSGI HTTP server used to run the Flask application in production environments. Provides worker process management and performance optimization.

H

Health Check Automated endpoints that report the status of system components. Used by load balancers, monitoring systems, and operational scripts to verify service availability.

Helm The package manager for Kubernetes applications. Simba Intelligence uses Helm charts for deployment configuration and management in Kubernetes environments.

Horizontal Pod Autoscaler (HPA) Kubernetes feature that automatically scales the number of pod replicas based on CPU usage, memory consumption, or custom metrics.

I

Ingress Controller Kubernetes component that manages external access to services, typically providing load balancing, SSL termination, and name-based virtual hosting.

Inventory Item A data source object that appears in the Playground interface, representing an available dataset that users can query with natural language.

J

JSON Web Token (JWT) A secure token format used for transmitting user identity and permissions between services. Used internally for session management and service-to-service authentication.

K

Kubernetes (K8s) Container orchestration platform used for production deployments of Simba Intelligence. Provides automated deployment, scaling, and management of containerized applications.

L

Langchain The AI framework that provides unified interfaces to multiple LLM providers. Enables Simba Intelligence to work with different AI services through consistent APIs.

LLM (Large Language Model) Artificial intelligence models capable of understanding and generating human language. Examples include GPT-4, Gemini, and Claude models.

LLM Configuration Settings that define how Simba Intelligence connects to and uses AI providers, including credentials, model parameters, and enabled capabilities.

LLM Provider An external service that provides large language model capabilities, such as Google Vertex AI, OpenAI, or AWS Bedrock.

M

Message Item A structured data object that contains streaming query results, including different event types (start, data, complete) and associated content.

Microservices Architecture Design pattern where applications are built as a collection of loosely coupled services. Simba Intelligence uses microservices for different functions like web interface, AI processing, and data connectivity.

N

Natural Language Query Questions asked in everyday business language that Simba Intelligence translates into database queries. Example: “What are our top-selling products this quarter?”

Network Policy Kubernetes security feature that controls traffic flow between pods, implementing network-level access controls and micro-segmentation.

O

OAuth 2.0 Authentication protocol that allows secure access to user accounts without sharing passwords. Can be used for SSO integration with enterprise identity providers.

P

pgvector PostgreSQL extension that adds vector data types and similarity search capabilities. Used by Simba Intelligence for semantic caching and AI-powered query similarity detection.

Playground The interactive interface where users ask natural language questions about their data and receive real-time answers with explanations.

Pod The smallest deployable unit in Kubernetes, containing one or more containers that share storage and network resources.

Pod Security Context Kubernetes security settings that define privilege and access control settings for containers, including user IDs, filesystem permissions, and security capabilities.

Prometheus Open-source monitoring system that collects metrics from applications and infrastructure. Compatible with Simba Intelligence’s metrics endpoints.

Q

Query Engine The Composer service component (port 5580) that processes SQL queries, optimizes execution plans, and coordinates with External Data Connectors to retrieve data.

Query Suggestion AI-generated recommendations for questions users can ask about their data sources. Automatically created based on data source analysis and common analytical patterns.

R

RBAC (Role-Based Access Control) Security model that assigns permissions based on user roles. Simba Intelligence implements hierarchical roles from basic users to supervisors with different access levels.

Redis In-memory data store used by Simba Intelligence for caching, session storage, and message queuing between services.

REST API Representational State Transfer Application Programming Interface. Simba Intelligence provides REST APIs for programmatic access to all platform capabilities.

Role A permission level that determines what features and data sources a user can access. Standard roles include Basic User, ROLE_CREATE_SOURCES, ROLE_MANAGE_CONNECTIONS, ROLE_ADMINISTER_USERS, ROLE_ADMINISTER_GROUPS (tenant administrators need both), and supervisor.

S

Semantic Cache AI-powered caching system that identifies semantically similar queries and reuses previous responses, reducing AI provider costs and improving response times.

Server-Sent Events (SSE) A web standard for real-time communication from server to client. Simba Intelligence uses SSE to stream query results and progress updates to users.

Service Discovery The mechanism by which services automatically find and communicate with each other. Simba Intelligence uses Consul for service discovery between Composer services and EDCs.

Service Mesh Network infrastructure layer that handles service-to-service communication, security, and observability in microservices architectures.

Single Sign-On (SSO) Authentication system that allows users to access multiple applications with one set of credentials. Simba Intelligence can integrate with enterprise SSO providers.

SQLAlchemy Python SQL toolkit and Object-Relational Mapping (ORM) library used by Simba Intelligence for database interactions and schema management.

StatefulSet Kubernetes workload type used for applications that require persistent identity and stable network identifiers, such as databases and stateful services.

T

Token Response Data structure returned by authentication services containing access tokens and related metadata for API authentication.

TLS (Transport Layer Security) Cryptographic protocol for secure communication over networks. Simba Intelligence requires TLS for all external communications and recommends it for internal service communication.

U

User Session An authenticated connection between a user’s browser and Simba Intelligence, maintained through session tokens with automatic renewal capabilities.

V

Vector Database Database system optimized for storing and searching high-dimensional vectors. Simba Intelligence uses PostgreSQL with pgvector extension for semantic search capabilities.

Vertex AI Google Cloud’s machine learning platform that provides access to large language models including Gemini. Primary AI provider for Simba Intelligence with support for chat, embeddings, and vision capabilities.

Virtual Warehouse Snowflake’s compute resource that executes queries. Virtual warehouses can be automatically suspended and resumed to optimize costs while maintaining performance.

W

Webhook HTTP callback mechanism that allows Simba Intelligence to notify external applications when specific events occur, such as query completion or data source creation.

Worker Process Background service instances that process tasks from the queue system. Includes both Celery workers for background tasks and Gunicorn workers for web requests.

Z

Zero Trust Security model that requires verification of every user and device before granting access to applications and data, regardless of their location or network.

Zoomdata Legacy brand name for the Composer service components. In Simba Intelligence, services like zoomdata-web and zoomdata-query-engine provide data discovery and query execution capabilities.

---

Related Terms by Category

AI and Machine Learning

• AI Provider, LLM, LLM Configuration, Capability, Embedding, Semantic Cache, Vertex AI, Langchain

Data Management

• Data Source, Data Connector, EDC, Composer Service, Query Engine, Inventory Item, Schema Discovery

User Interface

• Data Agent, Playground, Admin Overlay, Query Suggestion, Message Item, Natural Language Query

Security and Authentication

• RBAC, Role, API Key, Bearer Token, Authentication Token, User Session, Single Sign-On, Zero Trust

Infrastructure and Deployment

• Kubernetes, Docker Compose, Helm, Pod, StatefulSet, Microservices Architecture, Service Discovery, Consul

Performance and Monitoring

• Health Check, Prometheus, Horizontal Pod Autoscaler, Server-Sent Events, Worker Process

---

Common Acronyms

Acronym	Full Term	Context
API	Application Programming Interface	REST API, API Key
CSS	Cascading Style Sheets	Frontend styling
EDC	External Data Connector	Database connectivity
HPA	Horizontal Pod Autoscaler	Kubernetes scaling
HTTP/HTTPS	HyperText Transfer Protocol (Secure)	Web communication
JWT	JSON Web Token	Authentication
K8s	Kubernetes	Container orchestration
LLM	Large Language Model	AI capabilities
ORM	Object-Relational Mapping	Database abstraction
RBAC	Role-Based Access Control	Security model
REST	Representational State Transfer	API architecture
SSE	Server-Sent Events	Real-time streaming
SSO	Single Sign-On	Authentication integration
TLS	Transport Layer Security	Encryption protocol
UI	User Interface	Frontend components
URL	Uniform Resource Locator	Web addresses

---

Usage Examples in Context

Example Sentences Using Glossary Terms

Data Analysis Workflow: “Users access the Playground to ask natural language queries about their data sources. The AI provider processes the question through Langchain, generates SQL queries via the Query Engine, and returns results through Server-Sent Events streaming.”

System Architecture: “Simba Intelligence uses a microservices architecture deployed on Kubernetes with Helm charts. External Data Connectors register with Consul for service discovery and communicate with the Composer Service to execute queries against external databases.”

Security Model: “The platform implements RBAC with roles ranging from basic users to supervisor level. Authentication tokens are validated for each request, and API keys enable programmatic access with appropriate permissions.”

AI Integration: “LLM providers like Vertex AI are configured with specific capabilities for chat, embeddings, and vision. The semantic cache uses vector databases to identify similar queries and optimize response times.”

---

This glossary serves as a comprehensive reference for understanding Simba Intelligence terminology. Terms are defined in the context of their usage within the platform and may have specific meanings that differ from general software industry usage. For additional context on any term, refer to the relevant user guides and technical documentation.