Warm Data
Warm : having a comfortable/agreeable degree of heat.

Warm data is primarily of interest to a specialist audience of data ‘explorers’ who design more complex, iterative, ad-hoc, ‘train of thought’ queries against recent and historical data.

Response time for warm data queries is measured in minutes or hours (even days) and schema complexity may be high.

Warm Data Solution
Most organisations store data of all temperatures in an ever expanding, centralised Enterprise Data Warehouse (EDW).

The RDBMS-based EDW is optimised to satisfy the warm data consumers.

However, the EDW struggles to support fast query access to a wide audience of hot data consumers. Furthermore, it is not an economically viable solution for the storage of very large volumes of cold data.

Against this backdrop, there is no real need to introduce a ‘warm data solution’. The challenge is to allow the EDW to satisfy the warm data consumers, which it is well placed to do, by adopting separate treatment strategies for the consumers of hot and cold data.

The following EDW platforms are very well placed to support warm data consumers: Netezza, SQL Server & Teradata.