Recommendation

  Data transformation can be done independently from arcpy or any other Esri software. Using the latest GDAL python or CLI is recommended for transformation.

  Some arcpy functions needd Portal signin thus installing ArcGIS Portal and federating it with ArcGIS Server is recommended.

  If  input data source is not on S3 through CUMULUS, consider using OPeNDAP to subset data. Use NcML to modify some attributes and overwrite variable to make data CF-compliant.

  If ArcGIS Pro on Windows will be used for creating mosaic dataset, working with a large input table is fine.

  If arcpy and MDCS on Linux Windows Server will be used, there's a limit in input table size for creating mosaic dataset.

  If AWS Lambda will be used for data transformation, make sure that data size is small and transfer & transformation doesn't take longer than 15 minutes.

  Creating input table on AWS RDS is recommended for parallel / asynchronous data transformation through Lambda.

  If MDCS fails, export the table into CSV and modify header to standardize. Then, try ArcGIS Pro using the CSV file.

  Convert any mosaic dataset into CRF to optimize service performance especially on cloud. It will cost extra for storage but use to improve user experience.