|
« The importance of Infrastructure teams |
Main
| Infrastructure - A key success factor »
This week has seen a significant announcement from Aster Data Systems, an upcoming data warehouse appliance (implemented as a software appliance on commodity hardware, website www.asterdata.com). Aster announced the implementation of the famed Google MapReduce framework into is architecture.
There are fellow analysts and other pioneers in the database field, who question the validity and the scalability provided by MapReduce. My goal is not to debate on MapReduce, but what I'm fascinated by and believe will help users especially in a large scale data warehouse, is the ability to provide a language independent search capability.
This feature can be implemented by FAST or ENDECA, but what the framework from Aster will help bring is the scalability and all the bells and whistles of a data warehouse appliance, with the search capability of a Google like architecture.
Yes, we are a long ways away from achieving a true Google like search on the data warehouse, but the new crop of vendors are taking baby steps to get to that point.
Greenplum has also announced that in a future release of their software this year they are implementing the MapReduce framework.
We will see more of these trends coming from the data warehouse appliance and new business intelligence vendors. The traditional RDBMS platform may think this direction too, but they definitely have a long pace to catch-up. Though all this is bleeding edge, these are the features that will dictate the future of the data warehouse and business intelligence.
|