How to improve Geo-processing Execution Speed
Geoprocessing services hosted in the ArcGIS server needs to be pretty fast and accurate. User will always want the service to be the fastest. Since ArcGIS Server will be able to service multiple users at one instant of time, inefficient services can overload your server and increase the processing time. The more efficient and intelligent your services, the more users can be serviced with the same computing resources at a smaller amount of time.
Use local paths to data and resources
The more faster the data requires to be processed is accessible by the server the more faster your service will be.It is good to copy your data to the server rather than placing it on the shared location and asking the server to access it.(Copying the server will be effective only if the Geo processing input data is very small in size and also when the input data is not dynamic)
Reading and Writing the Intermediate data IN_Memory
The service execution speed will get increased by multiple times when the intermediate data is allowed to stored in the machine memory(RAM).Input and output operation is the slowest of all the computing operations.In Geoprocessing storing the intermediate data in a file geodatabase or personnel Geodatabase will take maximum of the execution time.
Please be sure the data which your are storing in the memory as an intermediate data need not be necessary after the execution is complete.Because once the execution is complete the data in the memory will get cleared.Don't write large amount of data to the in_memory as it may affect the performance of the processing service.
To write to your computer's memory, use the pathname in_memory, as follows: