DSXchange

Thanks guys.
The transformer was parallel one.. i am brining 4 columns from the source and adding one new column for the output.. no derivation just a hardcoded single alphabet value to it.. I am not sure how come the performance on the output link has degraded so much by just doing this.

ummm well i have before that transformer i have around 40k rows/ sec being processed - I am just using the transformer to add a new column to the input metadata and assigning a default value to it ( no constraints or anything). and the out put drops to 3k rows / sec..... could'nt think of anything e...

Guys,

Just a quick question. Can be pass hardcoded values in a modify stage? i do not want to use transformer for this as it kills in the performance.

Thanks.

The dsdlockd.log file keeps on growing continously although the start in the dsdlock.config file is set to 0.Any ideas on how to stop it growing ?

qt_ky wrote:Are you asking about comparing, for example, an RCP disabled job with 20 columns defined vs. an RCP enabled job with the same 20 columns passed through?

Yes for a start.

Hi, Can anyone logically explain the difference on the load a Unix Machine has (in terms of CPU , memory and I/O) between a job that has RCP enabled and versus a job that has columns defined(Which one of them will consume more resources).If someone can give a logical analysis with some concrete anal...

ray.wurlod wrote:Never seen it. How are your users authenticated - internal IIS, operating system or LDAP?

Through LDAP - kerbrose authentication.

Hi,

Is there a limit of users and projects on Datstage.Some users have their roles automatically revoked on projects without the administrators officially revoking them.Have you experienced anything like this ? and is there a way to check out how to find out why the access is being revoked ?

Thanks

Thanks.The system is healthy at the moment.I will monitor when there is extreme load.Based on your experience can you tell me why the DS Server will hang or require a reboot when the CPU maxes out on Unix ? Do you have any guidelines on how to avoid a reboot (or DS hang) ?

Hi, Can anyone guide me how to find the list of the 10 users and 10 active projects (jobs running on various projects) that are consuming the most CPU , memory and Disk I/O on the Datastage Server which is on a Unix AIX platform.I have tried to look at NMON graphs bu they only give me the datastage ...

ummm any other tips pertaining to this ?

The jobs are taking along time to compile now.. as far as i remember i could change the logging method from tables back to the old one (RT_XXX) which made the compilation time less.. does anyone remeber where the flag RT_LOGGING something like this used to be set to 1 for this?

Hey all,

Can someone please help me understand how can we reindex in the new vesion of datastage. Before release 8 it could be done using the command prompt using REINDEX ALL command as far as i rememeber.

Many thanks.

Yes, but that option is only available when I specify read method as "Specific file(s)", whereas, I am using the "File Pattern" option to open hundreds (maybe thousands) of files at the same time.

Thanks Ray. That's a good suggestion.

I know there's a filter command in the Seq. File stage in server jobs, but is there such a thing in PX? I can't see any such option at the moment. Is it hidden/obscure?