List the four core set of capabilities typically characterizing mashups and give your explanation of each one.
Tight budgets bring out the resourcefulness in us all. Enterprise IT groups have responded to the current economic downturn by consolidating systems, cutting back staff, delaying projects, and rethinking processes in order to minimize impact on service levels.
Unifying access to enterprise intelligence is difficult.
Many organizations seek a single unified view of disparate enterprise data. However, this is easier said than done. Enterprise intelligence — the trusted, authoritative reference data for answering critical business questions — often originates in many scattered silos, comes in myriad open and proprietary formats, and conforms to a wide range of schemas and models. Unifying heterogeneous information for BI is a very complex, time-consuming, labor-intensive, costly process. Given how difficult it is to find and aggregate all this data, it is no surprise that most BI users report that they don't have access to all the information they need.
Modifying rigid data structures requires tricky data modeling.
Typically, organizations strive to consolidate most of their structured transactional data in an enterprise data warehouse (EDW) with online analytical processing (OLAP) data marts, the primary repositories accessed by BI applications. For all the benefits of this approach, EDW and OLAP data marts often have a less rosy flipside: rigid data structures, such as “star schemas,” that require attention from skilled data modelers within the IT organization. For example, a requirement from the business to connect, load, transform, join, aggregate, and display data from new sources in existing reports would require a complex and time-consuming change to the logical and physical data models supporting the BI environment.
Designing BI applications is best left to the pros.
BI reports, dashboards, and other applications are often so complex and convoluted that only experienced developers should design or maintain them. Even the most user-friendly, point-and-click BI applications require users to slog through a daunting range of user interfaces, features, reports, metrics, dimensions, and hierarchies. As a result, most BI end users must run back to IT for help in creating new reports, queries, and dashboards.
Handling BI service requests can lead to major back-end consequences.
BI apps often hook into a deep pipeline of OLAP cubes, EDW production tables, persistent staging nodes, extract transform load (ETL) scripts, data cleansing rules, source data models, and other data structures, metadata, and programming artifacts. A user's request for a small change — such as a new field — in a BI report can require extensive modifications to many back-end systems (see Figure 1). By the same token, one minor upstream change to a single source-data element may require a handful of changes to ETL and data cleansing jobs, which may in turn lead to even more data model changes in the EDW, OLAP data marts, and other affected databases.
2. Visit www.kapowtech.com and write a short report on what you discovered.
We have discovered that Kapow Software pioneered a new and innovative approach for radically improving ROI and delivering ...