I'm not a database guy but I do know something about what it costs to put together data centers and how the database itself can represent a significant chunk of IT budget change. So while Aster's acquisition by Teradata for a reported $263 million is noteworthy in itself, is there a bigger picture to consider?
Right now, colleagues are reporting this as a big data play and to a certain extent that is right. But think back a couple of years to the time when Teradata acquired Kickfire. At the time, that was was spun as a small data warehouse analytics play:
Trying to gain a foothold in a crowded market for analytic database systems, Kickfire aimed at customers that had smaller-sized data stores and a desire to analyze that information, but no appetite for the price tags of products from larger vendors.
But Kickfire lacked MPP (massively parallel processing) capabilities, which was "a recipe for disaster ... in the 'Big Data Era,'" Abadi said.
"It is well known that more than 95 percent of data warehouses are smaller than 5TB, and that MPP is not strictly necessary for less than 5TB," he wrote. "So it is easy to get into the trap of Kickfire's thinking that the mass market is addressable without building an MPP product. However, businesses are looking forward, and seeing much more data in their future ... and can often be reluctant to select a product with known scalability limits."
Over at Briefings Direct, guest writer Tony Baer of Ovum argues:
Acquisition of Aster Data, probably the best pick of the remaining lot of columnar database challengers, provides Teradata yet another facet of an increasingly well-rounded product portfolio. Going forward, we expect that Teradata will continue its offerings of vertical industry data templates to extend to the columnar world.
Viewed from a market perspective, Teradata’s acquisition marks the home stretch for consolidation of the current crop of analytic database challengers, who are mostly spread in the columnar field. Dellis the last major platform player standing that has yet to make its move.
But there is another possible play.
My understanding is that both Kickfire and Aster use open source databases - MySQL and Postgres but they can work with pretty much anything. One can expect these will be swapped out for Teradata's own DB system to allow the offering to expand into services with which Teradata is familiar. Now think about the cost to the customer. A fraction of Power 7 and Exadata boxes sold by IBM and Oracle respectively.
Now think one step further. What does this mean in a world where data volumes are exploding yet the need to parse data is also increasing...and where the back end OLTP systems might well be an SAP in large scale situations?
All of a sudden Teradata starts to look attractive because it can put solutions into the market at a fraction of the price that IBM and Oracle can offer. One school of thinking believes there is no reason to dismiss Teradata (with Kickfire and Aster) as potential OLTP candidates even though their design is aimed at analytics style applications.
Even when you take into account all the moving parts of a deal and the monetary tags that go with it, Teradata could chew off a big chunk of IBM and Oracle's business and still profit handsomely. For companies looking at their total IT infrastructure cost landscape and already attempting to divest, Teradata could provide an intriguing possibility.
At least that's the way I am being encouraged to look at it from 'those close to events.' But what do you think? Does such a scenario make sense?
As I said at the top of the post - I'm not a DB guy...just askin'.