Datastage vs Seebeyond

vyyuri · Post by **vyyuri** » Sat Apr 12, 2008 5:10 am

Could some body please help me on this?

For a given scenario how do we chose which product exactly fits in ( I mean to say the comparison between Seebeyond and Datastage)

I listed down some of the points for both of these

1) Data Volume me is high go for Datastage
2) For complex scheduling go for Seebeyond
3) I know one is integration product and other one is ETL tool
4) Ofcourse license cost matters

My scenario is simple:-

1) Source data is stored in Oracle Database (Volume is huge)
2) Initially do the full load to target system ( I am preferring Datastage for this)
3) Generate files every day for the changed data and send it to target (Daily changes will be hardly 1/2 million for each file) . The file structure is bit complex (Some how I feel Seebeyond is suitable here , write down stored procedure and call it from seebeyond , But Can somebody advise why I can not choose Datastage here)

Can any body please suggest me advantages and disadvantages by using these products for the above scenarios .

Thanks a lot your help

Note :- I am working on both these products as a developer , I am not able to find out or analyze more than the first 4 points I mentioned .

Regards
Srini

chulett · Post by **chulett** » Sat Apr 12, 2008 6:49 am

SeeBeyond? Is that the product (ICAN, I believe) that Sun acquired a couple of years ago and calls Java CAPS now? I really don't know much of anything about it, but thought it was more healthcare related for some reason.

vyyuri · Post by **vyyuri** » Sat Apr 12, 2008 2:28 pm

Yes you are right Sun bought it . It is now JCAPS.
It is an integration tool

ray.wurlod · Post by **ray.wurlod** » Sat Apr 12, 2008 5:09 pm

It's a scheduling tool more than anything.

However, if all you want to do is move stuff around within Oracle, then a combination of PL/SQL scripts scheduled from SeeBeyond/JCAPS will do the trick for you.

When, in future, because of amalgamations, mergers, the need to capture data from or write data to other sources, or whatever other reasons, then you will have difficulty doing that in PL/SQL (for example, how do you get data from Excel?).

Plan ahead; businesses do change over time, so an ETL tool for managing the deltas is probably your better long-term solution. The jobs created by that tool can still be externally scheduled, if required.

Ultimately, the decision is one of maintenance; in a multiple data source/target environment a more flexible ETL tool will require fewer skills to maintain than a disparate set of vendor-specific scripts.

vyyuri · Post by **vyyuri** » Sat Apr 12, 2008 7:12 pm

Thanks a lot Ray and chulett for your responses.

They are going to stick to Oracle (That is what the message I got from business , But of course no body know the future)

They are asking me to prepare a Presentation on comparisons and differences of these two tools. ( In the end of presentation I have to choose one tool for the development and have to support that)

Somehow I feel uncomfortable comparing ETL tool with EAI tool. If they ask me to write a comparison on ETL products Or EAI tools that will be easier and I can suggest some thing, but they are mixing these two

Problem here is I have to suggest one solution sticking to Database Oracle

If I say Seebeyond they are asking why can not I use Datastage , If I say Datastage they are saying why Can't I use Seebeyond . I am totally got confused.

When I say I will write PL/SQL block and use seebeyond for scheduling, Even the same can be done in Datstage . We can call Pl/SQL anonymous block and can schedule daily to generate the file.

I am sorry, I am not trying to push my work to your head , Just looking for suggestions from experts

Regards
Srini

chulett · Post by **chulett** » Sat Apr 12, 2008 8:19 pm

I'm curious how you can be expected to prepare any kind of comparison between tools that you aren't intimately familiar with? Or did you... 'volunteer'?

I'd be surprised if you can find any site where both have been compared head-to-head already, but you never know. Put your google-fu to the test. Contact both companies and see what they can provide for you. Make something up! </kidding>

vyyuri · Post by **vyyuri** » Sat Apr 12, 2008 9:10 pm

'volunteer' :D . For the past 8 years I have been working on J2EE,Oracle,Seebeyond and Datstage . My managers introduced me to client and in the discussion one funny guy got this silly doubt and did put me in trouble .

I did search on internet ( asked Google uncle ) , Oops no answer

ray.wurlod · Post by **ray.wurlod** » Sun Apr 13, 2008 4:18 am

OK start your presentation with a slide that outlines the difference between a hammer and a screwdriver. If you only have one you can do the job of the other, but it's not pretty.

vyyuri · Post by **vyyuri** » Sun Apr 13, 2008 6:03 am

Good said Ray , Let me take Hammer and bang their heads .

eostic · Post by **eostic** » Sun Apr 13, 2008 9:02 am

The EAI vs ETL comparison is a classic one....and even though the lines have greyed somewhat (ETL tools can do more real-time these days, and EAI tools introduced more large volume capabilities), the characteristics can still be fairly used. ETL tools have historically been stronger at high volume, batch oriented, metadata driven, decision support activities, and EAI have historically been stronger at transaction oriented, hierachical data structures that are often message driven. Part of that transaction support is often out-of-the-box awareness of transaction standards such as EDI, SWIFT, EDIFACT, ACORD, etc. Also, largely in part because of their original marketplace years ago in decision support and data warehousing, ETL tools "tend" to have tooling that is more appealing to data oriented folks. Ease of use is of course "in the eyes of the beholder," but long ago, EAI was paying attention to 24 by 7 ensured delivery while ETL was pioneering drag and drop. As noted earlier, the lines are far more grey now, but you should still be able to find a significant amount of support for both tool types in most organizations, depending on the problems to be solved and the skill sets of the supporting teams.

Ernie

vyyuri · Post by **vyyuri** » Sun Apr 13, 2008 1:09 pm

Thanks a lot eostic . Your message helped me a lot. You are great. I am pointing out every thing you said ( Got few points from Rays message)

I am even writing on Exceptional Handling , Scheduling , restarting the job if it fails, restarting jobs and server if there is any maintenance and even required skills ( Seebeyond guy should know Java and the tool)

Thinking more and leaning toward using seebeyond for this case because already they have well defined special framework for handling exceptions ( Sending pager , Mail and sms to cell) . In addition to that already there are some projects in production on seebeyond

Once again thanks a lot
Srini

chulett · Post by **chulett** » Sun Apr 13, 2008 2:28 pm

vyyuri wrote:they have well defined special framework for handling exceptions ( Sending pager , Mail and sms to cell)

Just as an FYI, this same 'framework' is pretty straight-forward to do in DataStage as well and - if you don't feel like rolling one yourself - you can download one for free from our own Ken Bland's website.

vyyuri · Post by **vyyuri** » Sun Apr 13, 2008 3:12 pm

Thanks chulett for providing information on framework.
Looks like left with no other option

But not sure whether client know about this framework :D

At least got some information to fill the Presentation . But thinking about concluding the presentation . I am thinking about memory,Licensing and cost .In the end I am trying to push the ball to their court

Regards
Srini

ray.wurlod · Post by **ray.wurlod** » Mon Apr 14, 2008 5:06 am

This may also prove useful in your presentation.

Or maybe not.

DS_FocusGroup · Post by **DS_FocusGroup** » Mon Apr 14, 2008 5:17 am

http://eai.ittoolbox.com/documents/popu ... d-etl-3462

this might help you a bit