Page 1 of 1
Number of Nodes used by datastage
Posted: Wed Mar 28, 2012 5:09 am
by karthi_gana
All,
Our configuration file has 3 nodes. This file is being used in all our jobs. When I run a job, will it use all three nodes for its processing? or will it use 2 nodes and the remaining one for operating system processes?
Inputs are welcome!
Posted: Wed Mar 28, 2012 7:56 am
by mobashshar
Will use all the nodes mentioned in config file
Posted: Wed Mar 28, 2012 8:11 am
by jwiles
It will use all nodes within the configuration file, depending upon the file and the job design.
Stages running in sequential mode will run on only one node regardless of the number of nodes defined in the config file. Stages constrained to certain nodes or node pools will run only on those nodes or node pool member nodes. Stages that are not constrained and are running in parallel mode will run on nodes which are members of the default node pool.
Pools are documented in the Parallel Job Developer's Guide section on the Configuration File.
DataStage does not typically control where "operating system processes" run unless you are calling external programs/commands/scripts specifically within the job or sequence.
Regards,
Posted: Wed Mar 28, 2012 6:37 pm
by qt_ky
To see more details, dump your job score to the job log and examine it. To find out how, search this forum on job score.