Edit

Share via


Performance and latency of Analytics

Azure DevOps Services | Azure DevOps Server 2022 - Azure DevOps Server 2019

When you use Analytics for reporting, you should understand data latency and query performance. To get started using Analytics, see what is the Analytics.

Note

The Analytics service is automatically enabled and supported in production for all services within Azure DevOps Services. Power BI integration and access to the OData feed of the Analytics service are generally available. You're encouraged to use the Analytics OData feed and provide feedback.

Available data is version-dependent. The latest supported version of the OData API is v2.0, and the latest preview version is v4.0-preview. For more information, see OData API versioning.

Note

The Analytics service is automatically installed and supported in production for all new project collections for Azure DevOps Server 2020 and later versions. Power BI integration and access to the OData feed of the Analytics service are generally available. You're encouraged to use the Analytics OData feed and provide feedback. If you upgrade from Azure DevOps Server 2019, you can install the Analytics service during upgrade.

Available data is version-dependent. The latest supported version of the OData API is v2.0, and the latest preview version is v4.0-preview. For more information, see OData API versioning.

Data latency

When you use Analytics, you query a curated copy of the data stored in Azure DevOps. The data copy helps optimize read and aggregation performance, and greatly reduces the effect reporting scenarios have on Azure DevOps.

Because the data is copied, Analytics is not a real-time time store. Copying the data introduces up to a 30-second delay before the data associated with any one change shows up in Analytics.

Query performance

Using the recommended query patterns, Analytics will respond to any aggregation or non-aggregated query query within 3 to 5 seconds. The query response will be paged if it exceeds 10,000 results.

Some of the entity sets available in Analytics are designed for aggregations. The service will limit the results from these Entities to a single page for any non-aggregated query as outlined in the recommended query patterns.