A geoprocessing service can contain one or more tools that use input data from a client application, process it, and return output in the form of features, maps, reports, files, or services. You can author a notebook using ArcGIS Notebooks and share it as a web tool. A web tool published from ArcGIS Notebooks is shared as an asynchronous geoprocessing service. To publish a webtool from ArcGIS Notebooks, see Publish a notebook as a webtool.
Service structure
A geoprocessing service is a resource that can be accessed through a URL. The URL format for a geoprocessing service is https:////rest/services//GPServer, for example:
https://myOrg.esri.com/arcgis/rest/services/ae34f9eacf09447dbfe562a7a6945b17/gpServerA geoprocessing service has properties including Service description, Tasks, and Execution type. The geoprocessing tasks that are published in a geoprocessing service are its child resources.
A geoprocessing service does not include operations, but in an asynchronous geoprocessing service, tasks support submitJob and cancelJob operations.
The tasks in a geoprocessing service inherit the Execution type property and include additional properties, such as the task Description and Parameters. The properties of the service and the task are determined by the publisher of the service and cannot be modified by the clients. However, a notebook author can edit the notebook and republish the webtool.
Input and output
Most geoprocessing tasks have both inputs and outputs. There are slight differences between the handling of the input and output of geoprocessing services.
Input
For asynchronous geoprocessing services, the input of a task follows the syntax of the data type of a parameter.
Output
For asynchronous geoprocessing services, the result must be accessed through the GP Result resource after the submitJob operation. The syntax of the result parameters will be based on their data types.
During the execution of the submitJob operation, the status of the job can be accessed through the GP Job resource.