February 25, 2009

Partial overview of Ab Initio Software

Ab Initio is an absurdly secretive company, as per a couple of prior posts and the comment threads on same. But yesterday at TDWI I actually found civil people staffing an Ab Initio trade show booth. Based on that conversation and other tidbits, I think it’s fairly safe to say:

Comments

14 Responses to “Partial overview of Ab Initio Software”

  1. Ed on March 27th, 2009 1:53 am

    Having used AbInitio, Talend and almost every other ETL/ELT product on the market. I would very much doubt such a claim by Talend that they come close to AbInitio in performance. AbInitio has an amazing capacity to squeeze out every ounce of performance from the hardware it is running on. I do understand why they are so secret, their approach is elegant and simple but could be easily copied.

  2. Dan on September 28th, 2010 3:00 am

    ab initio’s secrecy has since 1995 always been part of their marketing strategy. I.e., generate buzz in inside circles. That being said, they had some sharp technologists at the outset focusing on building a very scaleable high performance platform bundled with professional services to ensure successful deployment.

    That being said, it will be interesting to see where mapreduce-based ETL solutions will be in perhaps 5 years. I think parity is at least that far off given the lack of pipelining and the significant investment in application-layer tools for data hygiene, precedence, conflict resolution, change management, etc. But I never bet long term against the power of an open-source movement with momentum.

  3. Savan on October 27th, 2010 9:16 am

    Clover ETL is as good as Ab Initio at a fraction of cost. It is also very easy to use (if not easier than Ab Initio)

  4. Savan on October 27th, 2010 9:17 am

    A book on Ab Initio for those want a head start – http://www.lulu.com/product/paperback/ab-initio-etl-made-easy/11920613

  5. ETL Maven on August 22nd, 2011 7:22 pm

    Evidently that book on Ab Initio is gone– the Secrecy Squad must have gotten to them…

  6. Curt Monash on August 23rd, 2011 12:33 am

    The Clover ETL guys reached out to me, then didn’t follow up.

    Sometimes I think the whole ETL industry is screwy. (Just kidding … partly.)

  7. Jura on September 30th, 2011 4:15 pm
  8. john on March 4th, 2014 6:42 am

    Is this a Spanish software made in Mexico? And what’s up with the ‘Ab’ word? It sounds like one of those late night infomercial advertising

  9. Curt Monash on March 4th, 2014 7:03 am

    Not Latin America. Just Latin.

  10. Antonio Romero on August 4th, 2014 7:26 pm

    Looks like the ab Initio book is gone again from Lulu. Do they have Silencer Squads out that take out anyone who leaks the Secret Lore?

    I’d love to know whether their usability really is what it’s cracked up to be– the few screenshots I have found on the web looked like nothing special, and the rest of the world is still moving… As far as squeezing all the performance out of your hardware, I can only speculate what they’re doing… maybe they’re really good at tuning integration/transformation to e.g. fit within the large caches on modern CPUs. And in the age of Hadoop, can they still justify their price point in terms of high performance on massive data volumes? so many unknowns.

  11. Experienced enthusiast on April 15th, 2015 3:25 pm

    Having worked multiple years with Ab Initio GDE, Co>OS, Technical Repository and their Metadata Hub up until v3.1.7.5, I recently used IBM DataStage v8.7:
    First impression was: OK, this looks quite similar, components have become stages. Record formats are now specified on the links instead of in ports of components, makes more sense!
    Second impression: NO WAY, DataStage, your user interface is this bad? I have to jump through hoops to write an expression containing an input field in an Transformer stage (equivalent of Reformat component)? I can only perform a lookup in a lookup stage?

    There is no integrated source control in DataStage v8.7, no way to know if another developer worked on a parallel job (equivalent of graph) yesterday and is not finished editing it. Setting up a source control workflow using Git would suck because you’d have to explicitly export every job after every change going through a cumbersome slow export process…

  12. Experienced enthusiast on April 15th, 2015 3:38 pm

    What I’m saying, I’ll choose Ab Initio over DataStage anytime but I suppose tools like Pentaho or SAS DI studio should be better than DataStage as well.

    3 more enormous advantages for Ab Initio:

    – Their documentation contain clear documentation with clear useful examples, written by developers themselves I’d say. DataStage documentation is one of the worst I’ve seen.

    – Ab Initio support is heaven on earth as well, you send them the error of your graph that can’t read data in parallel from a Teradata DB over 32 instances and they’ll tell you Teradata only accepts 4 instances by default. Expert specialist advice about a product they just interface with.

    – Debugging at every point between every 2 components (by enabling a watcher). Sounds normal until you use DataStage: put a Copy and a Watch stage in between your flow, copy over the record formats and you can look at a few samples of data… Unbelievable.

  13. herbert on October 12th, 2015 4:54 am

    the AI book can not be found there in lulu. Can someone post the new url?

  14. Jobot on May 2nd, 2016 3:58 pm

Leave a Reply




Feed: DBMS (database management system), DW (data warehousing), BI (business intelligence), and analytics technology Subscribe to the Monash Research feed via RSS or email:

Login

Search our blogs and white papers

Monash Research blogs

User consulting

Building a short list? Refining your strategic plan? We can help.

Vendor advisory

We tell vendors what's happening -- and, more important, what they should do about it.

Monash Research highlights

Learn about white papers, webcasts, and blog highlights, by RSS or email.