Cloud Ingestion

Cloud ingestion is the mechanism that moves the collected measurements from the edge device to the Cloud.

Ingestion is resilient, as Edge devices buffer when offline and automatically backfill when connectivity returns. Depending on your need for latency vs. cost, there are different transport options.

Transport options

Choose how each Tag sends measurements to the cloud:

• Hot Path Streams measurements to the cloud with minimal delay for real-time/streaming use cases (dashboards, alerts, live KPIs). This typically incurs higher cloud costs relative to batching, depending on volume.

• Cold Path Batches measurements into files and uploads them at a defined cadence (see Cold path upload rate below). This is the most cost-effective option but introduces a delay by design. Edge buffering ensures that, even if the site is offline, data is retained and forwarded when the connection is restored.

• Disabled Cloud upload is turned off for this Tag.

Selecting Cloud ingestion method for a Tag

Cold path upload rate

When Cold path (batched) ingestion is selected, the cold path upload rate control appears, where it is possible to change the default ingestion rate for each individual Tag.

• Defaults to a global platform setting of 10 minutes.

• Override on Tag level, by setting a custom ingestion interval.\

  

  Cold path ingestion upload rate override, on Tag level.
• Override on Device level, by changing the default upload rate value in Ingestion Endpoint configuration. This can be found in the Device Configuration for the device, in the Settings tab.\

  

  To override cold path ingestion rate on Device level, select the Settings tab in the Device Configuration page for the specific Device, and select "Change Ingestion".

Cold path ingestion upload rate override, on Device level.

In the Change Ingestion settings, you can configure the Upload Mode for the device. This determines whether uploads are batched into a single file or emitted as one file per Tag. For most deployments, choose Single File—it reduces object counts and improves throughput. Use Per-Tag only when a downstream system explicitly requires separate files.