Skip to content

Commit ec1376a

Browse files
Merge pull request #215303 from santiagxf/santiagxf/aml-batch-patch
Update how-to-use-event-grid-batch.md
2 parents e4e7d93 + 1ecff7c commit ec1376a

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

articles/machine-learning/batch-inference/how-to-use-event-grid-batch.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -27,7 +27,7 @@ The workflow will work in the following way:
2727
4. It will trigger the batch endpoint (default deployment) using the newly created file as input.
2828

2929
> [!IMPORTANT]
30-
> The proposed Logic App will create a batch deployment job for each file that triggers the event of *blog created*. However, keep in mind that batch deployments distribute the work at the file level. Since this execution is specifying only one file, then, there will not be any parallelization happening in the deployment. Instead, you will be taking advantage of the capability of batch deployments of executing multiple scoring jobs under the same compute cluster. If you need to run jobs on folders, we recommend you to switch to [Invoking batch endpoints from Azure Data Factory](how-to-use-batch-azure-data-factory.md).
30+
> When using Logic App connected with event grid to invoke batch deployment, a job for each file that triggers the event of *blog created* will be generated. However, keep in mind that batch deployments distribute the work at the file level. Since this execution is specifying only one file, then, there will not be any parallelization happening in the deployment. Instead, you will be taking advantage of the capability of batch deployments of executing multiple scoring jobs under the same compute cluster. If you need to run jobs on entire folders in an automatic fashion, we recommend you to switch to [Invoking batch endpoints from Azure Data Factory](how-to-use-batch-azure-data-factory.md).
3131
3232
## Prerequisites
3333

0 commit comments

Comments
 (0)