Eye on Oracle

Feb 5 2007   5:00PM GMT

Oracle Data Integrator execs size up the competition

Derek Kuhr Derek Kuhr Profile: Derek Kuhr

As a technology reporter, I’ve come to expect two different types of responses whenever I ask software executives about their competition: Either they go on a near-tirade talking smack about their chief rival, or they go the other way and don’t say anything about them at all. (The latter response oftentimes reminds me of a Tonight Show guest referring to a television show which appears on “another network,” as if the mere mention of another network’s actual name would immediately send droves of viewers into a channel-changing frenzy and ultimately destroy NBC’s business. But that’s a bit off topic and bad for search engine optimization.)

That’s why you could color me surprised the other day when I was interviewing two Oracle executives about the newly updated Oracle Data Integrator, a data integration offering Oracle acquired along with Sunopsis Inc. last October. When I asked who Oracle’s chief data integration competitors were and how Oracle’s approach differed from theirs — admittedly a bit of a softball question — I’m pretty sure I got a straight answer. Oh, it definitely sounded a little bit like a sales pitch, as responses from marketing execs tend to, but it had substance.

“From my perspective the main competitors would be the data integration players, which are really comprised by Informatica, IBM and to a lesser extent companies like Tibco,” said Jeff Pollock, senior director of Oracle Fusion Middleware.

What differentiates Oracle from those vendors, Pollock continued, is Oracle’s strong support for heterogeneous (a word I still can’t spell on my own after seven years in the business) architectures; a useful set of pre-fabricated templates called Knowledge Modules that simplify the data integration mapping process; support for both batch and real-time updates; and a strong ETL (extract, transform, load) engine — which Oracle refers to somewhat confusingly as an ELT (extract, load, transform) engine —  that gives the system greater performance and broader support for third parties.  Finally, Pollock said, the fact that Data Integrator runs on Oracle Fusion Middleware and is 100% Java-based makes it the right choice for folks looking to build a service-oriented architecture (SOA).

At that point I suggested in passing that existing Oracle customers might prefer being able to go to an existing vendor — Oracle — for their data integration needs, and that seemed to throw the executives off a bit. It was then clear that they didn’t want to focus on the old “one-stop-shop” marketing campaign and rather focus on the features of the product.

“It’s also hot-pluggable,” said Ashish Mohindroo, senior product director for Oracle Data Integrator continued, “which means that it can work with the existing infrastructure. So, if you have invested in Teradata or IBM WebSphere [or others], that’s all fine because the majority of the components of the Oracle Fusion Middleware family can plug right on top of that to extend and evolve, rather than rip and replace, that infrastructure.”

I’d like to ask some questions to folks out there with data integration experience: Am I right? Did Oracle give me a straight answer or did I just get shamboozled by a couple of corporate slick talkers who were presumably wearing nice suits. And if you believe they are on the level, do you see a need for these “differentiating” features in your organization? Let me know by sounding off on this blog thread. If there’s a good amount of interest, I’ll write a follow-up story showcasing your opinions and getting to the bottom of what end users really want from data integration providers.

Talk to you soon.

Mark Brunelli, News Editor

3  Comments on this Post

There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when other members comment.
  • Simon Ellis
    Straight Answer - Maybe. Oracle is heavily developing their proposition into what they term the "fabric" layer - data and system integration. For years they have had parts of the solution, but with their acquisition strategy over the past few years they are really starting to develop a complete solution while adopting the Open Standards, SOA stance. As for "hot-pluggable", well I have not come across anyone yet who wants to really do this - most organisations I know take the one-stop-shop view to these types of things - Oracle V IBM V Microsoft. The combination of Oracle saying "our tech will work with (almost) anything" with many organisations looking for single vendor solutions may be a good play for Oracle.... time will tell...
    0 pointsBadges:
  • Bob Zurek
    Not a straight answer. First of all, let's be really honest hear, the core players in the ETL marketspace all support heterogenous data connectivity, so I would say this isn't a strong differentiator. You wouldn't be a key player in the ETL market unless you had heterogenous data connectivity. All major players also have real-time and batch and even on-demand (ETL process invoked by supporting SOA), so that is not a differentiator. Saying that "Data Integrator runs on Oracle Fusion Middleware and is 100% Java-based makes it the right choice for folks looking to build a service-oriented architecture (SOA)" is also not quite right. Making something 100% java doesn't mean it is SOA ready. For example, what if a developer wants to invoke a data integration process via EJB or HTTPS? As long as the solution supports neutral bindings like bindings for EJB, HTTPS, XML/SOAP it doesn't matter if the tool is based on Java or not. Key differentiators in the ETL space include. 1. Support for open standards and open interface along with capabilities to easily extend the product by novices. 2. Scalability, Security and Ability To Recover From Failures. On the scalability front, the ability to run processes in parallel across distributed and grid infrastructure to support the the largest data volumes in the enterprise. Terabytes, not megabytes. 3. Supports the end-to-end process of the complete data integration lifecycle where the data integration platform MUST include Data Profiling, Data Quality and robust Metadata management capabilities. If you are going to extract, transform and move data, then you should be able to first profile the data to make sure it is the proper data and you should be able to then apply data quality processes and rules to the data as it moves to the warehouse. 4. Change Management thru Metadata. What happens when data sources change? Typically, an ETL processes without core metadata management capaiblities will break. However, with a strong metadata management framework, the ETL professional can run impact analysis and lineage for getting information like "where used". This helps with managing the change process just to name a few of the key attributes of a metadata management system as part of the ETL solution. 5. Business Glossary. A key diffentiator of an ETL solution is one that employs a business glossary so that users and IT professionals agree upon the "meaning of things" The business glossary is part of the Metadata management framework. Think of it as wikipedia for your data and metadata. Hope this helps others understand the true differentiators one must look for in an enterprise data integration platform.
    0 pointsBadges:
  • Bob Zurek
    Sorry, the second paragraph should say "here" and not "hear".
    0 pointsBadges:

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

Share this item with your network: