Azure Time Series Insights数据源咨询：数据库及CSV时序数据能否接入

Hey there! Let's break down your questions about Azure Time Series Insights (TSI) data sources clearly:

1. Can database-stored data be used as a data source for Azure Time Series Insights? #

Azure Time Series Insights doesn’t support direct ingestion from databases (like SQL Server, Cosmos DB, or other relational/NoSQL databases) out of the box. That said, you absolutely can get that data into TSI with a bit of middleware:

• Use Azure Functions or Logic Apps to periodically pull data from your database, transform it into TSI-compatible JSON format, and send it to an Azure Event Hub or IoT Hub (both are native TSI data sources).
• For larger-scale or scheduled syncs, Azure Data Factory can handle extracting data from your database, transforming it, and routing it to a TSI-ingestible endpoint.

2. Can CSV-formatted or file-based time series data be used with Azure Time Series Insights? #

Yes, you can use CSV or file-based time series data, but you’ll need to route it through a supported ingestion channel. Here are the most common approaches:

• Upload your CSV files to Azure Blob Storage, then use Azure Data Factory or Logic Apps to parse the CSV, convert rows into TSI-compliant JSON events, and send them to an Event Hub/IoT Hub for TSI to ingest.
• For one-off or small datasets, write a simple script (using Python, PowerShell, etc.) to read the CSV, format the data correctly, and push it directly to an Event Hub. TSI will automatically pull the events from there.
• If you’re using Azure Data Explorer (ADX) alongside TSI, you can also ingest the CSV into ADX first, then link your ADX cluster to TSI for querying and visualization.

内容的提问来源于stack exchange，提问作者Shyam V