Analytics, AI, and Integration Services Overview
Analytics and data services
Azure Synapse Analytics combines SQL pools, Spark, and pipelines for enterprise analytics. Azure Databricks provides a collaborative Apache Spark platform. Azure Data Factory orchestrates data movement and pipelines. Azure Stream Analytics processes streaming data with SQL-like queries. Azure Data Explorer is optimized for log and telemetry analytics.
AI services
Azure AI services provide prebuilt models for vision, speech, language, and search. Azure OpenAI provides access to large language models with enterprise controls. Azure Machine Learning is for building, training, and deploying custom models. Azure AI Search adds indexing and retrieval. Bot Service builds conversational agents.
Integration services
Azure Logic Apps provides low-code workflow orchestration. Azure Functions provides code-first event handlers. Service Bus provides enterprise messaging with queues and topics. Event Grid routes reactive events. Event Hubs ingests high-volume telemetry streams. API Management publishes and governs APIs.
Traps
Questions often confuse Event Grid, Event Hubs, and Service Bus. Event Grid is for discrete event notifications; Event Hubs is for streaming telemetry; Service Bus is for reliable enterprise messaging with sessions, dead-lettering, and ordering. Another trap is treating Azure AI services as custom model training, which belongs to Azure Machine Learning. Logic Apps versus Functions is also commonly tested: workflow connectors versus code.
Worked example
An IoT solution receives millions of device messages per second, then triggers a workflow when a device reports an anomaly. Use Event Hubs for ingestion, Stream Analytics for anomaly detection, and Event Grid or Service Bus to trigger downstream processing.
Rule of thumb
Choose Event Grid for events, Event Hubs for streams, Service Bus for commands and enterprise messaging, and Logic Apps for connector-heavy workflows.