So why did Pervasive acquire Data Junction?
Show me the synergy
Pervasive last month acquired Data Junction. The question is: why (asks Phil Howard of Bloor Research)?.
Pervasive, like Hewlett-Packard, started life in a garage, though in the case of Pervasive it was in the 1980's. It eventually (albeit under another name) became well-known as the developer of Btrieve, before it was acquired by Novell. Then it was spun out as Btrieve Technologies and, most recently, changed its name to Pervasive and its product's name to Pervasive.SQL.
Despite the name change, most of us who have been around for any length of time will still tend to think of the product as Btrieve but, whatever you call it, the product is still widely used as an embedded database, particularly for applications going into smaller and medium sized organisations.
Data Junction on the other hand, initially came to fame (if that is the right word) as an ETL (extract, transform and load) vendor. However, the truth is that it never became one of the major players in this space. More recently, it turned its attention to the enterprise application integration space.
So, where is the synergy between the two companies? Well, both are profitable, they do not compete with one another, and they are approximately the same size. If nothing else, there should be economies of scale. But that hardly seems to be the whole answer.
It is quite a leap from databases to application integration though, no doubt, a database will be required for the latter for storing metadata. So perhaps it is more reasonable to think in terms of Data Junction's ETL capabilities.
Microsoft has built ETL capabilities (DTS) into SQL Server, Oracle has Oracle Warehouse Builder and even IBM has some ETL capabilities so there is a clear trend towards including such capabilities in database products. However, I don't know any other database vendors that are doing this.
The reason why Microsoft et al are providing ETL facilities is that they expect their products to be used for data warehouses purposes and it obviously makes like easier to load data is you have your facilities. But it seems unlikely that Pervasive expects to be used as a data warehousing platform.
The other good reason for wanting an integrated ETL tool is because you expect to pick up business from your competitors and you want to make it as easy as possible for customers to make that decision. This is a plausible explanation but it is difficult to see which users Pervasive will be targeting if, indeed, this is its aim.
Another conceivable possibility is that Pervasive is eyeing GoldenGate's market. GoldenGate markets what is effectively a combined ETL and replication tool, and it has proved quite successful at the mainframe level. Pervasive has an extension to Pervasive.SQL that provides replication and perhaps it sees this as a possibility.
The bottom line is that I don't why Pervasive is buying Data Junction. There doesn't seem to be any particularly obvious motive. I could, of course, ask the company. Indeed, I will do so. But I find that initial ruminations on subjects such as these are more interesting without reference to the vendor since, in any case, you may not get to here the whole truth. If I learn more from the company I'll let you know.
Phil Howard is an analyst at Bloor Research who specialises in database technology. His biog is here.