DSXchange

Hi ALL,

I recently started reading for Certification . Got doubts about few of the objectives . Could any body please throw some light ? ( I searched the forum but could not get any answers)

i) Section 5 - Databases (15%) :- Describe how to control restart of DB target (separate the "L" from ETL) -

What is separating L from ETL ( hope it is not a silly question) ?

ii) Section 7 - Combining and Sorting Data (10%) :- Describe the differences between unique option in sort, RemDup stage -

I feel unique option works same in all the cases

iii) Section 8 - Automation and Production Deployment (10%) Import/Export
Including compiled Transformer

what is so special about exporting/importing compiled Transformer?

vi)Section 8 - Automation and Production Deployment (10%)-Explain how a DataStage job may be run using a third party scheduler (e.g., autosys, control U, CRON, etc.)
Explain how performance run statistics from a job run can be captured

Can any body please let me know ehere I can get the documentation for these ( I checked dsjob command But could not able to find out) ?

Thanks a lot all for your help
Srni

1) Separating L from ETL basically means creating two jobs. One for Extraction and Transformation and writing the results to a DataSet. And a second job for Loading the data to the database from the DataSet. In ETL if the target database fails during the load then you will have to extract and transform the data again. By using two jobs i.e ET and L, you can just load the data from the DataSet.

2) One difference between Unique option in sort and Remove duplicates stage is that, in Remove duplicates stage you can specify which duplicate record to keep, the first or the last.

3) If you are using a parallel routine then, by using the export option on the routine will not export the object file. So that will have to be exported separately and make sure that the path in the routine is the same.

dsjob is the way to obtain information externally. It has families of options, one for getting information from the log (all of the form -logxxxx) and one for getting information about the job and its components (all of the forum -xxxxinfo).

Deployment is really about making sure that the engine is accessible to all processing nodes and all the executable components are also accessible; either by sharing or by being distributed to all the nodes.

Thank a lot Bugfree and Ray .I owe you a lunch . Let me know when you come to USA, Detroit.

For Unique option , I thought , It is about the UNIQUE option on Advanced Tab ( Input Page).

Rergards
Srini