Tuesday, 18 October 2011

IBM has made low budget data warehouse possible with Informix Warehouse

I don’t think anybody will deny that a business analytics is a value adding intelligence to small, mid and enterprise level business and not just an enterprise level requirement. This necessarily means that, an analytics can be performed on any size of data. But with existing warehouse solutions which needs special infrastructure, analytics is beyond the budget of small and mid market. As per Gartner’s report, the small and mid market needing analytics is 75% and 25% is enterprise market.

The IBM Informix Warehouse Accelerator is the solution to address warehouse need of all classes but is certainly being well accepted by small and mid market due to its low cost of ownership. This doesn’t need any specialized infrastructure and work excellent on Intel 64bit machine with Linux OS. The in-memory compressed warehouse with massive parallelism and capability of applying predicate on compressed data ensures extreme high performance and almost zero administration. If you closely study your warehouse schema, you will find that the index size is more than one third of data size. Also for faster summary reports, you will have multiple summary tables. These factors actually increase the size of database drastically. But with Informix Warehouse, you don’t need index or summary tables as the data is stored in memory. This actually can reduce your data warehouse size almost by 50%. With reduced size, definitely the cost of ownership reduces drastically. More over you don’t need any data warehouse tuning, which is biggest gain.

As the warehouse is in memory, you don’t need cube generation either and can perform summary, drill down, roll up operations run time and that to with predictable performance.

You can follow my blog to know more about a) How the total cost of ownership affected b) What is the significance of predictable performance c) Approximate cost to deploy small and mid market analytics d) How analytics improves the performance of your business. e) Typical data warehouse architecture

Disclaimer: The postings on this site are my own and don't necessarily represent IBM's positions, strategies or opinions

Monday, 17 October 2011

IBM's Way of managing time series data

Off-course you can store time series data in proprietery database or in traditional RDBMS way in any 2nd generation database. But do you know the IBM's innovative way of managing time series data efficiently?

Understanding the need of today's era when everybody is talking about real time data management and micro data analysis, the IBM had envisaged the challenges of handling such big volume with high performance and hence had introduced Informix TimeSeries back in 90's.

You can read more about TimeSeries here - http://www.dbta.com/Articles/Editorial/Trends-and-Applications/Managing-Time-Series-Data-with-Informix-76454.aspx