“[…] In the early days of BI, running queries was only possible for IT experts. The tremendous increase in available computational power and main memory has allowed us to think about a totally different approach: the design of systems that empower business users to define and run queries on their own. This is sometimes called self-service BI. An important behavioral aspect is the shift from push to pull: people should get information whenever they want and on any device [142]. For example, a sales person can retrieve real-time data about a customer, including BI data, instantly on his smart phone. Another example could be the access of dunning functionality from a mobile device. This enables a salesperson to run a dunning report while on the road and to visit a customer with outstanding payments if she or he is in the area. These examples emphasize the importance of sub-second response time applications driven by in-memory database technology. The amount of data transferred to mobile devices and the computational requirements of the applications for the mobile devices have to be optimized, given limited processing power an connection bandwidths. As explained previously, . An exception was the need to consolidate complex, heterogeneous system landscapes. As a result of the technological developments in recent years, many technical problems have been solved. We propose that BI using operational data could be once again performed on the operational system. In-memory databases using column-oriented and row-oriented storage, allow both operational and analytical workloads to be processed at the same time in the same system” (“In-Memory Data Management, Plattner and Zeier, Springer, p.183).
cosimoaccoto.com 