DSXchange

i know that!

My question is : is there a best pratice to know which link is to be considered as the left one?

I worked with Informatica also and performance was impacted depending of which link was the left one.

to be more precise,

i wonder why there are 2 options (left and right outer join) as you can change the order of the links easily.

Hi all,

What is the impact of the order of the input links in a join stage when you do an outer join (in term of performance)?

Should the "biggest" link be the left or right link, or is the performance equal , regardless the configuration?

Thanks for your help.

this is C-ISAM developped by Informix.

I would like to read and write to it.

thanks for your help.

Hi all,

is it possible to access a unix isam file using datastage? (PX or server)

thaks all.

[quote="Thibal"][quote="chulett"][quote="Thibal"]About my "cleanliness" issues, I'll find tips when developing (It's a pity it doesn't exist a kind of Sequence Container)[/quote] Actually, there is! It's called another Sequence job. :wink:[/quote] Sorry, I did...

in a job control : the DSPrepareJob command is equivalent to the Reset if requierd and run of the job sequence.

kumar,

Thibal would like to know some opinions about job sequences, that's what i did.

naveen, it is obvious that Job Controls are more flexible than Job Sequences! You can't dynamically set parameters to a job activity, maintenance is harder when you have to add a parameter to all the jobs, etc... But Job Sequences are very useful for many ways! I was saying that there is no miracle ...

hi all,

is it possible to set the path of the log files generated by the PX load, without changing the Scratch ressource disk of the APT config File?

thanks.

Hi, I want to add a new column to my flow, initialized with a job parameter. Th problem is: in some cases, i don't want to do so. Consequently i have to use a orchestrate schema file (when i don t want to set it, i use an empty file) How can i set the parameter in the schema file? I tried the [&...

maybe u've not set the impersonation mode!

Check it!
Otherwise, set the ulimit for dsadm.

you also have to kill all the processes owned by the user who runs the jobs.
Otherwise the ulimit tab won't be updated!

well, i did this for a very large dimension!

it was not so brutal, besause the join does not consume very much CPU.

The partition and the sort do, so u have to be very careful ! put your lookup in a well partitionnend and sorted dataset in another job before.

I've already seen this problem. My solution was ( i'm not sure it was the most efficient, but it worked and i implemented it on schedule) : a join between 2 flows with the key - your flow - the lookup flow ( contains, id, effective_dt and expiration_dt) then you filter the data and you keep the reco...