Browse Prior Art Database

Efficient way to create high performance report utilizing visualized performance objects Disclosure Number: IPCOM000238923D
Publication Date: 2014-Sep-25
Document File: 6 page(s) / 164K

Publishing Venue

The Prior Art Database


Disclosed is a method to project performance information directly on report output, so that business intelligence application developers have the information necessary to make informed decisions when selecting the best design.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 38% of the total text.

Page 01 of 6

Efficient way to create high performance report utilizing visualized performance objects

In a system with multiple services in a distributed nature, it is always a challenge to identify performance issues due to the system's complexities and concurrencies. In business intelligence (BI) software, the report rendering optimization is also a challenge. The delivery of high performance reporting is business-critical; however, only a few subject matter experts know how to identify the optimization options or system bottlenecks.

Current methods and practices for addressing these challenges include performance logging, program code profiling, and a half-split (or binary chop) method. The first two methods are mostly used by software engineers who have in-depth product knowledge goals to improve software performance. Software engineers do not have the business knowledge to improve data

warehouse model design and develop a high performance report. The third method is often used by BI application developers who have business knowledge but do not have the insight of the software product. By manually simplifying a report, one at a time, BI application developers identify the worst performing component in the report before find a work -around. As a report becomes more complicated, identifying the bottleneck becomes more challenging. It is sometimes impossible to simplify the report because of the multiple interdependencies inside the report. The more complex a report is, the more tedious and challenging it becomes for a developer.

A method is needed to lower the technical threshold, thus allowing a report author to develop a high performance report without help from subject matter experts (SMEs).

The novel contribution is a method to use a framework, which is built into the report-rendering engine, to generate a visualized performance indicator directly on the report output. This allows BI application developers to know the exact cost of each component. With this knowledge, BI application developers can make informed decisions when selecting the best design and adopt the best practices in a self-sustainable manner. This method reduces software support costs and improves product performance perceptions.

A report usually consists of a set of layout components. It might contain dozens, if not hundreds, of these components. The following are a few samples of layout components:

Prompt List


Page 02 of 6

Chart Master detail report Header/Footer
Logic container (invisible)

Each layout component contains a physical layout container and a data container. In Figure 1, the physical layout container is shown as a crosstab. The data container represents data needed for the crosstab , which is usually a query result from a data store such as a relational database. The rendered output is the physical layout.

Figure 1: Layout components of a report

The proposed visualized performance component can be a simple textbox, a to...