Datastage to informatica Migration

A forum for discussing DataStage<sup>®</sup> basics. If you're not sure where your question goes, start here.

Moderators: chulett, rschirm, roy

Post Reply
rajudx
Participant
Posts: 45
Joined: Tue Nov 14, 2006 1:58 pm
Location: NJ

Datastage to informatica Migration

Post by rajudx »

Hi,

We are planning to migrate datastage 8.7 version jobs to informatica.
Can someone share any ideas or steps to save the time during the migration process.
Saving time :i.e. Exporting/ importing the meta data.
Ran
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

I don't really think there is anything you can save from your DataStage world that can be directly leveraged in Informatica. For the metadata you'll want to import that using Informatica from the actual sources rather than a dump of the DDL, for example.

About the only thing I can suggest is make sure you have a full set of the requirements from your DataStage environment, especially if you bring in people to work on this that know Informatica but not DataStage. And don't expect to slavishly turn every single DataStage job into a functionally identical Informatica mapping, just take whatever it is that the job (or series of jobs) are doing and then decide on how best to do that in the new tool. Same goes for sequence jobs into workflows.

Oh and while Informatica does have the ability to "partition" your jobs, it isn't nearly as slick nor full featured as what DataStage has with its parallel framework. For example, there's no equivalent to a dataset. I've also not found an equivalent for "build ops" or custom C++ functions, the only tool "under the covers" of Informatica is... Java.

Good luck.
-craig

"You can never have too many knives" -- Logan Nine Fingers
Mike
Premium Member
Premium Member
Posts: 1021
Joined: Sun Mar 03, 2002 6:01 pm
Location: Tampa, FL

Post by Mike »

Good advice from Craig as usual and particularly valuable for your scenario since he has expertise in both tools.

I've always been a firm believer that once you've chosen one of the top enterprise ETL tools that it's a fool's game to switch.

Mr. Client, you're going to spend all of this big $$$ on an ETL tool conversion and when all is said and done you may be as well off as when we started if everything goes as planned.

You're much better off learning how to be successful with the chosen ETL.

What was that famous quote... "Don't blame the tool, blame the carpenter".

Mike
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

I've never really understood the desire to completely change horses in the middle of the stream. It hardly ever works with gaming companies and the engine they use and I've seen warehouse projects go down in flames for it. And as Mike noted you'll probably just end up back where you started (in a couple of years) if everything goes as planned... but it never does.

There are many key differences between the two tools and those will take a bit to get used to but one thing you'll find are some surprising similarities between them. For example, an "expression transformation" in Informatica is basically a Transformer and stage variables are called variable ports but are otherwise identical. Sorter, Joiner, Lookup, Filter (yada yada yada) all exist and do what you are already used to. As noted though, if your shop is any kind of "Parallel Power User" shop then I think you'll be a wee bit disappointed with your options in that regard in Informatica. Oh, and... no equivalent to transformer looping. Actually, there's no looping of any kind. :(
-craig

"You can never have too many knives" -- Logan Nine Fingers
eostic
Premium Member
Premium Member
Posts: 3838
Joined: Mon Oct 17, 2005 9:34 am

Post by eostic »

Yep. Yada, yada, yada. We've even seen this situation internally across the 20+ years of DataStage. Many ETL tools have come and gone, and been part of the family...helping customers build NEW functionality in DataStage, and forward transformation makes perfect sense, but converting all the "old" Jobs (in whatever tool, whether it was PWE/DS390/Data Manager/Rational Matters/DWE/and whatever RedBrick's ETL was called)? Much less benefit, and possibly more work than it's worth. That being said, if a particular old etl process is "running slow" or the "machine is going away" (and said tool doesn't run on the newly chosen platform) or the license cost is exceedingly high, there "might" be justification......but as Mike, Craig and others noted, "find that justification" before you begin such an endeavor.

Ernie
Ernie Ostic

blogit!
<a href="https://dsrealtime.wordpress.com/2015/0 ... ere/">Open IGC is Here!</a>
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

Guessing it's far too late for anything other than... the die is cast, gird your loins, full steam ahead!
-craig

"You can never have too many knives" -- Logan Nine Fingers
vmcburney
Participant
Posts: 3593
Joined: Thu Jan 23, 2003 5:25 pm
Location: Australia, Melbourne
Contact:

Post by vmcburney »

Take a look at Analytix DS Lightspeed. It can reverse engineer DataStage jobs into mapping specifications and then generate Informatica jobs. You need to verify they support your version of the products. It may save you up to 80% of the development effort. You still have a big testing exercise and data verification and performance improvement process to go through. These projects usually cost lots of money and deliver no business value. It is worth doing if you are re-engineering or rebooting your DW but not worth it if you are just swapping over the plumbing on a legacy DW.

http://analytixds.com/wp-content/upload ... ksheet.pdf
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

Very... interesting!
-craig

"You can never have too many knives" -- Logan Nine Fingers
rajudx
Participant
Posts: 45
Joined: Tue Nov 14, 2006 1:58 pm
Location: NJ

Post by rajudx »

Thanks for your suggestions and this will be helpful a lot for me.
Ran
Post Reply