Bloor: Open Source Data Integration Tools Not as Widely Reused


I'm all about the free stuff, so I was pretty excited to see that Bloor Research is giving away copies of its recent report, "Comparative Costs and Uses of Data Integration Platforms."

Bloor surveyed 200 companies about their use of data integration platforms, with a particular look at the costs over the long haul and "how effective those costs were when spread across the number of projects that the platform supported," according to Philip Howard, a data management research director and author of the report.

For those of you who don't have the time or interest to read the whole 16-page whitepaper, Howard also wrote a column for IT-Director, which summarizes the most surprising findings.

One odd-ball revelation was the correlation between the type of tool used, the length of projects and the number of projects. So, for instance, open source solutions and Microsoft tended to have peak numbers of projects, and those projects tended to be of medium length. On the other hand, people tended to use IBM and Informatica for longer, more complex projects, but not for shorter projects.

My interest was piqued by these findings, though:

  1. Open source data integration platforms don't seem to fare that well in the long run. Of course, they're cheap, so you'd expect a certain amount of use. Problem is, they're not being reused in other projects. Howard noted this raises the total cost of ownership per project. Howard cautioned this could change as the products mature -- but it's certainly something you might want to consider when you're looking at your options. (I bet I get a few choice e-mails from vendors about that one -- but, you know what they say: Don't spam the messenger.)
  2. SOA is still a big so-what for data integration platform use. Howard attributed this to the application focus of SOA, a situation he believes will change as more companies start to look at how SOA can change data integration. He seemed surprised by this. Mostly, he wrote, data integration platforms are used for data migration and conversion projects, which I thought made sense.

Of course, for those of you looking to buy, the big news will be how well which solutions fared. Howard doesn't keep you in suspense about who came out on top in the long run -- it was Progressive -- but for more details, you'll need to download the report.

The full report includes findings, broken down by vendor, on:

  • How many man-weeks required for needs analysis, broken out by solution
  • How many weeks it took to learn each product and build the first solution
  • A comparison of initial and yearly administrative costs
  • A look at TCO and costs per project
  • A comparison of how custom code fared in each category

Hey -- what else have you got to read this weekend?