I have a large set of featureclasses that need to be run into a workspace, one at a time because loading the dataset all at once will make my FME workspace terminate because the memory is full. I have used workspacerunner on a folder with multiple csv files, which solves the memory problem. But I can't figure out how to do a simular thing with multiple featureclasses. I wouldnt mind if the featureclasses need to merged into one and the input the first xx number of features into the workspacerunner. If that is even possible.
Page 1 / 1
First of all, did you try and run the workspace with featurecaching off? Because as great as it is for developing, it eats ram and is not needed in production runs.
In the child workspace, create user parameter for the feature class name. Use this user parameter read in the FeatureReader reading the featureclass.
In the parent workspace, use a FeatureReader to read only the schema features. Then feed the features attribute fme_feature_type_name into the workspacerunners user parameter for featureclass. Set the workspacerunners parameter wait for job to complete as yes, and you are good to go.
First of all, did you try and run the workspace with featurecaching off? Because as great as it is for developing, it eats ram and is not needed in production runs.
In the child workspace, create user parameter for the feature class name. Use this user parameter read in the FeatureReader reading the featureclass.
In the parent workspace, use a FeatureReader to read only the schema features. Then feed the features attribute fme_feature_type_name into the workspacerunners user parameter for featureclass. Set the workspacerunners parameter wait for job to complete as yes, and you are good to go.
Attached workspace demonstrating this.
@nielsgerrits that works perfectly!