This article describes key technologies and methodologies used by Report Portal.
What is OLAP?
OLAP (On-Line Analytical Processing) is a technology that provides solution to three fundamental reporting problems:
Slowly running reports. For example, calculating total sales for all products might
take very long time. Moreover, each report might keep calculating the same value all
over again. To resolve this problem, OLAP can store all possible totals
in summary (aggregation) tables.
The aggregations are usually re-calculated daily or hourly.
Report Explosion. The number of possible static reports that can be
created from a database may be too large. For example, we have a sales
table with three columns: product, date and dollars. Suppose that we sell
just 10 products and we have just 10 days worth of data.
We can have 29 possible reports:
- Total sales
- Total sales by product
- Total sales by product for each day - 10 reports
- Total sales by date
- Total sales by date for each product - 10 reports
- Total sales by date (rows) and product (columns)
- Total sales by date (columns) and product (rows)
- Total sales by date (rows) and product (rows)
- Total sales by product (rows) and date (rows)
- Total sales by date (columns) and product (columns)
- Total sales by product (columns) and date (columns)
If we are to add an additional column: "customer name" to our table,
the number of possible reports would increase exponentially.
Because an average organization can have much more than three columns
associated with its sales, the number of possible reports can
approach or exceed trillions!
OLAP technology allows users to design and modify reports themselves
via a simple drag and drop interface, eliminating the need of
having an "army" of report writers on staff.
- Statistical calculations are usually not supported by a relational database.
For example, SQL (Structured Query Language) requires users to run multiple
queries to get the mean, median or the standard deviation.
Because OLAP has pre-calculates totals, it supports a wide range of statistical
functionality that can be calculated in a single step.
What is Data Warehousing?
Data Warehousing is a common approach to creating a reporting database that
provides the following advantages:
All organizational data in stored in one place.
The data is sourced from different operational systems such as accounting,
human resources and payroll. When operational records are imported into
the data warehouse, the system makes sure that all information can be
linked to the rest of the data.
Sometimes it is important to have a historical snapshot of the data in history.
The operational database is updated constantly to reflect the latest changes and
may not store the old data. The data warehouse will store all the old and new data.
Running reports off an operational database can cause problems for
operational system users.
An OLAP database can retrieve the data from an operational database or a data warehouse.
What is Data Mining?
Data Mining is the automated extraction of hidden predictive information from database.
Data Mining does two things:
- Identifies hidden patterns in your data.
- Attempts to predict the feature based on the historical patterns.
Report Portal lets users browse a Microsoft decision-tree
data-mining models to identify hidden patterns.
What is XMLA?
XML for Analysis (XMLA) is a SOAP-based API.
XMLA provides the most efficient way to access an OLAP database over the Internet.
Hyperion, Microsoft and SAS first proposed the standard April of 2001.
The standardís organization is XMLA.org council.
XMLA Consulting is a permanent XMLA.org council member.
Currently, Microsoft Analysis Services has the
leading market share in the OLAP market.
Report Portal is based on
AJAX, the same technology used by Google
Maps and Gmail. The AJAX approach provides zero-footprint platform with
mimimum of page reloads. Essentially, Report Portal is an IE SOAP client
that consumes the XMLA web service provided by Microsoft. This gives Report Portal
a technological advantage over other simalar products ProClarity
(classic asp based) and Panorama (applet based).
Want to learn more?
- OLAP and Microsoft Analysis Services:
- Data Warehousing:
- Data Mining:
- XML for Analysis: