how many process will create

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
4friends
Participant
Posts: 71
Joined: Tue May 01, 2007 12:51 pm
Location: Mumbai

how many process will create

Post by 4friends »

if am runing a job on three node and having 3 stages then how many process will create
thanks & regards
kk
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Thirteen.

Unless there is operator combination or some other optimization happening.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Raamc
Premium Member
Premium Member
Posts: 87
Joined: Mon Aug 20, 2007 9:08 am

Post by Raamc »

Rey,

May i know the process how you calculated the number of processes?
ArndW
Participant
Posts: 16318
Joined: Tue Nov 16, 2004 9:08 am
Location: Germany
Contact:

Post by ArndW »

This answer might also change if you are using a partitioned DB2 database as source/target/lookup as well.
4friends
Participant
Posts: 71
Joined: Tue May 01, 2007 12:51 pm
Location: Mumbai

how we can say as 13 process,can u explain me

Post by 4friends »

ray.wurlod wrote:Thirteen.

Unless there is operator combination or some other optimization happening. ...

if i am having stages like

dataset---->transformer---->targetdatabase.
and running on three node.

can u tell me where we can get this information.plz
thanks & regards
kk
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

The simplest model is where there is no operator combination. Each operator (for simplicity) represents one stage. You have three stages, therefore three operators, on each of three processing nodes. That makes for a total of nine "player" processes executing the operators. Each processing node has a "section leader" process looking after co-ordination on that particular node, and there is one "conductor" process (the startup process) managing overall co-ordination and logging.
Nine player processes, three section leader processes and one conductor process equals a total of thirteen processes.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply