Edenbase: a multi-purpose solution

I recently wrote about compliance solutions for the telecommunications and allied markets and, as a result, I was contacted by yet another company (in addition to those mentioned by readers) with an appropriate solution. However, what has particularly attracted my interest is that this company, Edenbase, also addresses the data warehousing market.

The two products the company markets are Edenbase CompactDB (for storing and retrieving call data records and the like) and Edenbase CompactWarehouse but they are both based on the same technology. In either case, what you do is to create a new schema on the same instance of your existing database (Oracle, DB2 or SQL Server) using Edenbase’s software, which compresses the existing data into a much more compact database.

The compression (which is based on the use of multiple algorithms) used by Edenbase is impressive: the resulting database is between 2% and 8% of the size of the original. Indeed, the company cites one user of this technology that has reduced a 4TB system down to 60GB, which is a ratio of 1.5%. Note that there is no change to the structure of the database or any of the table definitions so the impact on any existing applications will be minimal.

Along with compression, Edenbase also builds indexes. These will likely be minimal in a compliance environment but in CompactWarehouse will normally be applied to every column in the warehouse. Note that the indexes, which are themselves compressed, are created after the data is compressed rather than before. This means that the creation of indexes is very much faster than normal (the company claims 32 times for indexing the entire database) and, of course, there is minimal index maintenance as everything is indexed (so you don’t have to make any decisions about what to index). In addition, note that Edenbase includes support for text-based indexes, as the origins of the product (see below) were in the pharmaceutical sector, where it was used to compress and index results of drug trials and similar, where it was necessary to be able to search against particular medicines, for example.

Data retrieval itself is also very much faster, not merely because everything is indexed but also because you can retrieve 20 times as much data (approximately) in a single disk read. The company claims, and this seems perfectly plausible, that the resulting efficiencies mean that you do not need to pre-aggregate data, in much the same way that you do not need to do so using appliance technology.

As far as Edenbase itself is concerned, it is an Israeli company founded in 2005. However, the software has a much longer heritage than that as it was originally developed by a company called Intercon in the ‘90s, specifically for the pharmaceutical sector. This company’s technology and intellectual property was acquired by Edenbase, which has productised what Intercon was previously only supplying in conjunction with its consulting services.

Edenbase presents an interesting alternative solution in its chosen markets. In both of these environments there seems to be another new supplier announcing its presence every other day—which makes it difficult to determine likely winners—however there is no doubt that Edenbase is a contender.