using shared containers

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
zulfi123786
Premium Member
Premium Member
Posts: 730
Joined: Tue Nov 04, 2008 10:14 am
Location: Bangalore

using shared containers

Post by zulfi123786 »

we are trying to implement range look up using a combination of join stage and filter stage and for that we are thinking of putting the framework in a shared container which would be accesable in any job that requires a range look-up
However the metadata varies from job to job (the number of columns going in and coming out of the container)

We have created a container and placed it in the job and modified the metadata within the container to suit at that particualar point of the job however this change is being reflected in the shared container.

Is there any possibility to have just the stages put in the container and add columns to it when actually being used in the job without changing the content of the container so that other jobs using it are not affected?

I guess placing the shared container in the job and then converting it into local container would help this ? Any other thoughts
?

Thanks in advance
samyamkrishna
Premium Member
Premium Member
Posts: 258
Joined: Tue Jul 04, 2006 10:35 pm
Location: Toronto

Re: using shared containers

Post by samyamkrishna »

You can do it by enabling RCP on the columns. The mandatory colums you can enter the metadata and teh other columns you can pass through by enabling RCP.
Althaf6553
Participant
Posts: 64
Joined: Wed Sep 26, 2007 6:52 am
Location: Syracuse ,NY

Post by Althaf6553 »

Just wanted to understand as what's the point of having a shared container with only 2 stages in it and editing it each time you want to use it in a different job..Here you are not reusing any thing but editing it each time you use it in job

Please clarify if i have misunderstood some thing .
Althaf
samyamkrishna
Premium Member
Premium Member
Posts: 258
Joined: Tue Jul 04, 2006 10:35 pm
Location: Toronto

Post by samyamkrishna »

You do not have to give any metadata. Enable RCP and pass teh columns as parameters to the container on which you are doing calculation.
zulfi123786
Premium Member
Premium Member
Posts: 730
Joined: Tue Nov 04, 2008 10:14 am
Location: Bangalore

Re: using shared containers

Post by zulfi123786 »

samyamkrishna wrote:You can do it by enabling RCP on the columns. The mandatory colums you can enter the metadata and teh other columns you can pass through by enabling RCP.
As we never know what and how many columns and with what metadata would be used for the range look up we cant judge what are mandatory columns.
zulfi123786
Premium Member
Premium Member
Posts: 730
Joined: Tue Nov 04, 2008 10:14 am
Location: Bangalore

Post by zulfi123786 »

Althaf6553 wrote:Just wanted to understand as what's the point of having a shared container with only 2 stages in it and editing it each time you want to use it in a different job..Here you are not reusing any thing but editing it each time you use it in job

Please clarify if i have misunderstood some thing .
There are a total of 5 stages and we want to have the same framework for implementing a range look up across all jobs untill we move to 8.x . There fore we have chosen to use a shared container which could be pullled in and modify only the metadata.
zulfi123786
Premium Member
Premium Member
Posts: 730
Joined: Tue Nov 04, 2008 10:14 am
Location: Bangalore

Post by zulfi123786 »

samyamkrishna wrote:You do not have to give any metadata. Enable RCP and pass teh columns as parameters to the container on which you are doing calculation.
Did not understand what is meant by passing columns as parameters. Could you please elaborate?
jwiles
Premium Member
Premium Member
Posts: 1274
Joined: Sun Nov 14, 2004 8:50 pm
Contact:

Re: using shared containers

Post by jwiles »

zulfi123786 wrote:we are trying to implement range look up using a combination of join stage and filter stage and for that we are thinking of putting the framework in a shared container which would be accesable in any job that requires a range look-up
However the metadata varies from job to job (the number of columns going in and coming out of the container)

We have created a container and placed it in the job and modified the metadata within the container to suit at that particualar point of the job however this change is being reflected in the shared container.

Is there any possibility to have just the stages put in the container and add columns to it when actually being used in the job without changing the content of the container so that other jobs using it are not affected?
When you add a shared container to a job within Designer you are not creating a copy of it, but rather a reference. The stages to the shared container are not physically included until the job is compiled.

When you've attempted to edit the shared container within the job, Designer opens the shared container itself and changes will be saved back to the shared container. Existing jobs will not be affected until they are recompiled.
zulfi123786 wrote: I guess placing the shared container in the job and then converting it into local container would help this ? Any other thoughts?
This would have the effect you are looking for and would not change the original shared container when you edit the local container. Based on your posts in this thread, this is probably your best option at this time.
Post Reply