Apatar – a few extracts short of a load

I’ve been meaning to try out the Apatar ETL/Mashup tool for sometime and today being yet another rainy day in this the worst Irish summer that I can remember (and Irish summers are not renowned for the lack of rainfall) I decided to give it a try out. Not impressed I’m afraid; comes up short when compared to either Kettle (Pentaho PDI), Proto, RSSBus or Talend. Very few database connectors (e.g. no SQLite,DB2 or Firebird support) this wouldn’t be a problem if the product offered a generic JDBC or ODBC connector. It does have one nice feature the others (other than RSSBus) haven’t, an Amazon S3 connector. But the thing that I find amazing is that a product that’s positioning itself as an Enterprise 2.0. mashup tool doesn’t have the ability to read and write Excel files! And no, CSV files don’t count.

Advertisements

4 responses to “Apatar – a few extracts short of a load

  1. Thanks for sharing your experiences with Kettle and Talend. Have you looked at or tried Jitterbit? I have narrowed my list to these three and was looking for someone that might have tried it.

  2. Brian,

    I’ve had it on my list for some time now to try out Jitterbit (that and Splunk http://www.splunk.com). My first impressions were of a high-end tool-set more suitable for very large organisations rather than the micro-ETL that I’m more interested in, so I’m unlikely to make time in the short-term (especially as it doesn’t support SQLite!!!) but if I do I’ll make sure to blog about it.

    Tom

  3. Apatar introduces a concept that is quite interesting though it is not designed so as to deal with large amounts of data. Indeed, the “no code” solution has its advantages but there are strings attached: it does hide the complexity of integration, but it also limits the flexibility of the software.

    In any case, Apatar is, indeed, more geared towards data mashups. In order to perform ETL, a tool that can manage great quantities of data is required, and in the open source world I do have been running into Talend Open Studio a couple times.

  4. @I.Kato

    Yes, Talend (or Pentaho PDI aka Kettle) is what you need if you intend to do serious heavy lifting of data, but to be fair to Apatar, their product has improved a lot since I wrote this (including the addition of an Excel component!). Even so, even for ‘simple’ mashups, I find Talend a more rewarding tool to use, particularly now that I’ve “rediscovered” Groovy as a scripting language for Talend.

    Tom