def create(
config_path,
cluster_mode,
destination_system,
destination_database,
destination_environment,
algorithm_instance,
ext_params_str
):
data_system = DataSystem(
config_path,
cluster_mode,
destination_system,
destination_database,
destination_environment
)
if data_system.database_type == DataSystem.DatabaseType.EMR:
config = AlgorithmConfigurationHadoop.create_with_ext_params(
config_path,
cluster_mode,
destination_database,
destination_environment,
algorithm_instance,
ext_params_str
)
execution_system = EMRSystem.from_data_system(data_system, config.get_emr_cluster_id())
return AlgorithmExecutorHadoop(execution_system, config)
else:
raise M3DUnsupportedDatabaseTypeException(data_system.database_type)
def create(config_path, cluster_mode, destination_system,
destination_database, destination_environment,
destination_table, load_type, emr_cluster_id, spark_params_str):
data_system = DataSystem(config_path, cluster_mode, destination_system,
destination_database, destination_environment)
if data_system.database_type == DataSystem.DatabaseType.EMR:
execution_system = EMRSystem.from_data_system(
data_system, emr_cluster_id)
spark_params_dict = json.loads(spark_params_str)
return LoadExecutorHadoop(execution_system, load_type,
destination_table, spark_params_dict)
else:
raise M3DUnsupportedDatabaseTypeException(
data_system.database_type)