Comment You may have seen that Oracle is buying TimesTen. The question is why?

Is it because it thinks that TimesTen has great database technology? Hardly: Oracle does not exactly have a reputation for recognising database technologies other than its own.

Then is it perhaps because Oracle thinks that it is worth harvesting the TimesTen user base? Given that this is peanuts compared to Oracle and that many of them are joint customers anyway then it is obvious that this is not the case.

The only logical conclusion is that TimesTen offers capabilities that extend Oracle's capabilities in a way that the company sees as a critical. TimesTen is able to do this in two ways: it can be used on an Application Server to improve performance; and its in-memory capabilities enable such things as real-time trading and, potentially, RFID.

However, TimesTen has been doing this for years and Oracle never saw the need to purchase the company before. You could argue that RFID now makes the difference, but TimesTen does not have any explicit RFID capabilities at present (at least that I am aware of) and if Oracle is going to have to develop that from scratch anyway, why does it need TimesTen to do so?

I don't think that any of this will wash. I think that we need to go back to the markets that TimesTen has come out of and have another look.

In the Application Server go-faster market there used to be a number of players apart from in-memory databases like TimesTen. In particular, there was Persistence, which was arguably the market leader with its EdgeXtend product and certainly had the best object-relational mapping capabilities. What used to be eXcelon was less important in the market with its Javlin product but could boast the best caching. Well, what has happened? Progress has bought both eXcelon and Persistence and is merging the two products to provide a go-faster product that should dominate the market.

And what about real-time environments? The problem with using an in-memory database to support real-time capabilities is that you actually have to commit incoming data to the database. In a real-time environment the resulting delay can be critical. This is why Apama developed capabilities to provide the same functionality but without the need for database commits. So, what happened next? Progress bought Apama.

Do you begin to see a pattern here? It certainly looks as if the acquisition of TimesTen is actually about preventing Progress from making inroads into Oracle accounts. Progress is miniscule compared to Oracle so it would be difficult to conclude that Oracle was running scared of Progress per se but this strengthens the argument that I have written about before: that Oracle is increasingly under attack and, moreover, is increasingly seen as vulnerable by other suppliers. If that is the case then one part of a defensive strategy would be to shore up the battlements – the purchase of TimesTen seem sto fit within that picture.

Copyright © 2005 IT-Analysis.com

Related stories

Database growth boosts Oracle
Oracle becomes EJB 3.0 'standard'
Ex-MS man joins Oracle
Oracle looks forward to SMB