Platform

Operational business intelligence (OBI) means many things to many people, but key is that it delivers information to decision makers in near real time, usually within seconds, minutes hours. The purpose is to empower front-line workers and mangers with timely information so they can work efficient and proactively to improve performance. This sounds easy but it’s hard to do.

Important decisions requires low latency or OBI systems which can handle moving parts and can recover easily from errors. This is especially difficult in highvolume environments. The key decisions you need to make when architecting a low-latency system is wether to use the data warehouse to support your OBI environment. The ramifactions are significant.

On one hand , the datawarehouse will ensure the quality of low- latency data. However, doing so may disrupt existing processes, add undue complexity, and adversely impact performance. On the other hand, creating a standalone operational business inteliigence system may be simpler and provide tailored functionality and higher performance.

But there are solutions on the way due to customer demands. The market has developed approach which fullfill the demand to be able to create rapid analysis which can be feeded and used on data from several large databases without a data warehouse or ETL layer.

Solutions

These architecture solution makes use of selective key or relevant data sources and real-time event processing. This eliminates the need for the collection large data volumes and makes it possible to create analysis of current business within 24 hours.  No more waiting time until the data in batches is updated in the datawarehouse. It uses advanced techniques for accelerating complex cross-database calculations, such as outer joins, "which in many cases required to create useful business analysis. The data dimensions are automatically fixed and the building and pre-defining "data cubes optimized for query performance is no longer needed. Moreover, the cost of change and the enormus flexibility thereby greatly reduced when the analytical requirements need to be adjusted.

Compared with the traditional architecture of data warehouses and BI technology a similar architecture will improve significantanly. The performance of operational systems will improve significantly because of the intelligent manner of the query load on the systems. The performance of operational databases are optimized and expensive hardware investments are minimized.

With such an operational business intelligence approach a data warehouse or ETL is not perse nessesary. This will save implementation time which is typically less than two weeks in comprahension with traditional implementations from nine to eighteen months.