Is it possible to build a multi-terabyte data warehouse in less than an hour? That's the pitch made by two-year-old Dataupia, which today revealed partnerships with AMD, Informatica, and a dozen other tech companies. But if it sounds too good to be true, well, you know.

Two months ago, Dataupia (sounds like "day-topia" and rhymes, not coincidentally, with Utopia) announced availability of its first product, the Dataupia Satori Server 12000, which the company positions as a data warehousing appliance. The system is sold in an expandable, blade configuration; each blade includes a dual-core AMD Opteron microprocessor, 2 terabytes of Seagate storage, and a massively parallel "database engine." Dataupia's server doesn't come with its own relational database management system, but functions as an architectural layer that brings massively parallel processing (MPP) scalability to symmetric multiprocessing (SMP) computing environments.

Dataupia's president and CEO Foster Hinshaw was a co-founder and CTO at Netezza, which also sells a data warehousing appliance and raised $108 million last week in an IPO. Hinshaw left Netezza two years ago to start Dataupia. He says the companies don't compete head-to-head, characterizing Dataupia's appliance as "mainstream" compared to Netezza's "high end" appliance. A more direct competitor, according to Hinshaw, is DATAllegro.

Hinshaw's pitch is simplicity: Dataupia's appliance connects to existing Oracle, DB2, or Microsoft SQL Server databases, creating a fast-and-easy data warehouse. The system scales to tens or hundreds of terabyes by adding more blades. At a cost of $19,500 per blade, a 20 terabyte data warehouse would cost under $200,000 – a bargain compared to a comparably sized data warehouse from Hewlett-Packard, IBM, Oracle, or Teradata. One of Dataupia's first customers is Sendio, a startup that sells anti-spam appliances.

Hinshaw calls the effort required to create one of these data-warehouses-in-a-box "trivial." So easy, he says, that even an IT professional who's not a data warehousing specialist could create a data warehouse using Dataupia's system in less than an hour.

Sounds great, but I have to wonder how well that promise will live up to the test of real-world data centers. Data warehouses are notoriously difficult to build and maintain, requiring systems integration, data extraction and cleansing, query optimization, data mining algorithms, desktop analysis tools, and more. It remains to be seen just how much of that can be done in the time it takes to bake a cake.