Page tree
Skip to end of metadata
Go to start of metadata

The Advanced Reporting node on the Administration tab allows you to achieve two primary objectives:

  • Provide a set of features for designing, scheduling, and publishing reports that have a custom structure and presentation. For more information, see Reports and report templates (Works folder).
  • Provide an interface to BMC TrueSight Capacity Optimization data and model results to third-party reporting tools.

Overview of databases in BMC TrueSight Capacity Optimization

Before discussing specific concepts, it is useful to recall some common definitions.

A relational database stores information in data structures called tables. More specifically, a data warehouse (such as the BMC TrueSight Capacity Optimization DWH) uses two kinds of tables:

  • Fact tables (FT): Contain all recorded data, also called observations.
  • Dimension tables (DT): Contain all the information necessary to decode keys used in fact tables to identify data samples; decode keys are also called analysis dimensions.

Dimension tables have a simple primary key, while fact tables have a compound primary key consisting of the aggregate of relevant dimension keys.

Additional information

The data warehouse organizes tables in a star schema, in which a few fact tables reference a number of dimension tables. Data stored in fact tables is classified in different dimensions; dimension tables describe all available dimensions and can be joined to fact tables as needed.

An example of such a star schema is shown in the following figure:

Example of star schema



The BMC TrueSight Capacity Optimization database model is a relational star schema organized with fact tables at different time resolutions and dimension tables.

All tables containing the suffix _DATA in their names are fact tables, wherease tables identified by the same prefix but not suffixed with _DATA are the corresponding dimension tables.

Another important concept is the difference between horizontal and vertical tables:

  • horizontal table stores all data related to a single fact in one row; this practice saves disk space but requires to modify the table structure, should the collected data types change.
  • vertical table enforces a fixed structure (typically: timestamp, fact ID, name, value) and uses one row for each data value. This is more space-consuming but allows for greater flexibility.

The following table is an example of a vertical table:

TS

ID

METRIC

VALUE

01/01/2007 00:00:00

12345

CPU_UTIL

60.75

01/01/2007 00:01:00

12345

CPU_UTIL

52.15

01/01/2007 00:00:00

12345

MEM_UTIL

80.20

01/01/2007 00:01:00

12345

MEM_UTIL

82.00

...

...

...

...

The following table is an example of a horizontal table:

TS

ID

CPU_UTIL

MEM_UTIL

...

01/01/2007 00:00:00

12345

60.75

80.20

...

01/01/2007 00:01:00

12345

2.15

82.00

...

Vertical tables can be transformed into horizontal ones.

The BMC TrueSight Capacity Optimization DWH schema is based on a vertical structure; should a horizontal structure be required for exporting and reporting, the AR module offers several ways to export data in a horizontal format.

There are two main kinds of fact tables, identified by the prefixes SYS and WKLD: one for system and the other for business driver observations. The main dimensions used by BMC TrueSight Capacity Optimization are: entity, object, subobject, and location; for more details, refer to ETL tasks.

The basic structure that BMC TrueSight Capacity Optimization uses to collect counters is shown in the following figure:

The BMC TrueSight Capacity Optimization data model

Working with Advanced Reporting

The features of different Advanced Reporting components are described in the following sections:

Where to go from here

The Advanced Reporting module produces a number of public views that can be accessed using the CPIT_REP account. For more information on these public views, see Accessing data using public views.

To view a list of queries that are based on public views in chargeback, see Chargeback reports based on Public Views.