DSXchange

Hello folks,
This is a open thread to list out some of the DataStage features, we feel like lacking in the currently available versions. Please share your knowledge if you feel some of the features listed are already available [or there is an alternative approach] :-

your contributions/suggestions are very welcome :-

Features "looking forward"
================

1)Jobs in Designer should have a last run parameter view , where it should automatically pick up the last run parameter. this is helpful when debugging [especially complex jobs]. example: view data should automatically fill in the last run parameters.

2)Custom aggregates [as supported in Oracle ODCIAggregate, lack of this in an ETL tool is very surprising].
oracle custom aggregates

3)External metadata for the repository which can be organized in 3rd party databases like oracle. [Informative is great in this extent]

4) Performance statistics of how much time the job is spending on each stage, how much memory is used, IO time etc

5)Unconnected lookups. Meaning if the lookup of a source row gives more than 1 row. it should pass on both the rows instead of 1. [i can understand "one row at a time" is the basis of datastage. At least it makes sense to pass 2 rows in case of a lookup]

Cool Features
========
1)Same "Look and Feel" in all the components Designer, Manager and Director.

2)Copy paste of stages , links , what not

3) Find and replace in transformer[ even though, no shortcuts like CTR+F , i know of]
4)regular expression
..
..
..

top of my head

regards,
Prabu

i believe the Subject is non-editable after publishing..

please read the subject as
Open -> "Good to have" Features in DataStage next edition

i tried editing it, but its not allowing me to change

Moderators please move the topic to "Wish list"

prabu wrote:Hello folks,
Features "looking forward"
================

1)Jobs in Designer should have a last run parameter view , where it should automatically pick up the last run parameter. this is helpful when debugging [especially complex jobs]. example: view data should automatically fill in the last run parameters.

Agree, I had to write one of these in BASIC code so when a job started it saved parameter to a file and when it restarted it loaded from that file.

prabu wrote:2)Custom aggregates [as supported in Oracle ODCIAggregate, lack of this in an ETL tool is very surprising].
oracle custom aggregates

And while you are at it how about a parallel aggregation stage that can sum and count at the same time! (Cannot currently do this, the count function excludes all other functions).

prabu wrote:3)External metadata for the repository which can be organized in 3rd party databases like oracle. [Informative is great in this extent]

Done, this is in the Hawk release. A metadata repository on Oracle/SQL server/DB2 or Sybase and SOA metadata backbone.

prabu wrote: 4) Performance statistics of how much time the job is spending on each stage, how much memory is used, IO time etc

Available for parallel jobs. Improved in Hawk with extra monitoring tools.

prabu wrote: 5)Unconnected lookups. Meaning if the lookup of a source row gives more than 1 row. it should pass on both the rows instead of 1. [i can understand "one row at a time" is the basis of datastage. At least it makes sense to pass 2 rows in case of a lookup]

Available in parallel jobs.

vmcburney wrote:

prabu wrote: 5)Unconnected lookups. Meaning if the lookup of a source row gives more than 1 row. it should pass on both the rows instead of 1. [i can understand "one row at a time" is the basis of datastage. At least it makes sense to pass 2 rows in case of a lookup]

Available in parallel jobs.

And Server jobs using the ODBC or UV stages.

thanks vmcburney & Craig, loking forward to Hawk, especially for the metadata part of it :D