Oracle® Real User Experience Insight User's Guide Release 11.1 for Linux x86-64 Part Number E22309-01 |
|
|
PDF · Mobi · ePub |
This appendix explains the use of the Enriched data exchange facility. The structure of the database tables used by it is also explained.
The Enriched data exchange facility enables you to combine the data gathered by RUEI with other data sources. These could include, for instance, Customer Relationship Management (CRM) or Business Intelligence (BI) systems. Using this functionality, you can produce customized analysis of your Web environment using your own BI tooling, as well as integrate RUEI's rich set of collected data with offline data to obtain greater insight into what drives your sales and revenue.
The facility works by exporting the data collected every 1-minute period to a database. By default, the data is exported to the same database instance as used by the Reporter. However, it is strongly recommended that you configure an alternative database instance for enriched data export. Access to data in the export database is available via SQL. The procedure to do this is fully described in the Oracle Real User Experience Insight Installation Guide.
As described later in this section, you can customize the content of the exported data to include information not normally collected by RUEI. For example, the contents or value of visitors' shopping baskets. Because the exported data is page-based, the available data is restricted to applications and suites, and does not include service-related data.
In addition to the user experience data gather by RUEI, current and historical KPI data can also be exported for customized analysis. This facility enables deep-dive analysis of the performance of your network environment and business-critical applications.
Controlling the Availability of Exported Data
The amount of data available in the export database is controlled via the Enriched data exchange retention setting in the defined Reporter data retention policies. These are fully explained in Section 12.9.1, "Defining Reporter Retention Policies". The structure of the database tables used within the export database are described in Appendix R, "Enriched Data Export Facility".
Example BI Implementation Using Enriched Data Exchange
This section presents an outline of a BI solution utilizing data from the Enriched data exchange facility. In this case, it makes use of Oracle Business Intelligence foundation (part of the Oracle Fusion Middleware product family). Its schematic structure is shown in Figure R-1.
Figure R-1 Schematic Overview of Data Warehouse Staging Area
The framework is based on Oracle Warehouse Builder (OWB). The RUEI-captured data is exported to a database. From the export database, it is uploaded via SQL scripts to a staging database. This then populates the production database. Once in the production DWH, the RUEI data is available through a wide variety of reports and dashboards. An example of these reports is shown in Figure R-2.
Enabling and Disabling Enriched Data Exchange
To enable the Enriched data exchange facility, do the following:
Select Configuration, then Applications, and then Enriched data exchange. The screen shown in Figure R-3 appears.
Use the Enriched data exchange enabled/disabled check box to specify whether the Enriched data exchange facility should be enabled. Use the KPI data exchange enabled/disabled check box to specify whether the export of current and historical KPI data should be enabled. By default, both are disabled. Note that the availability of export data is determined by your Reporter data retention polices (see Section 12.9.1, "Defining Reporter Retention Policies").
If either are enabled, it is recommended that you configure an alternative database for export data. The procedure to configure an alternative database is fully described in Chapter 9 of the Oracle Real User Experience Insight Installation Guide.
Optionally, you can define additional data items to be included in the exported enriched data. Typically, these are elements in the client request or server response headers that are not normally collected by RUEI, but which you want included in the exported data. To do so, click Add new item. The dialog shown in Figure R-4 appears.
Figure R-4 Add Enriched Data Export Item Dialog
Use the Source type menu to define how the required item should be identified within the data collected by RUEI, and the scope of the search. You can specify to search within the client request header or server response header, using either a literal search or an XPath expression, or to search within a custom page-tagging implementation for a specific tag. Further information about support for custom page-tagging schemes is available in Appendix A, "Tagging Conventions".
Use the Source value field to specify the specific argument or element from which the data item's value should be taken.
Use the Export name field to specify the name to be assigned to the data item. This becomes the item's element name. In the case of a custom tag, this will appear in the CONTENT column in the BI__BIDATA_MASTER (see Table R-2) in the format &
name
=
value
=
. Similarly, XPath expression items would be reported in the REQUEST_HEADERS or REPLY_HEADERS columns. Note that the Export name field is not available if you select a header-related option in the Source type menu. In this case, the name of the header is used in the appropriate column. When ready, click Save. The new item, if found in the monitored traffic, will start to appear in the reported data within 5-10 minutes.
Existing data items can be modified by right-clicking them within Figure R-3, and selecting Edit. You can also select Remove to delete it, or select Remove all to delete all currently defined items.
Be aware that the SQL queries used to access exported data can place a significant performance overhead on the export database. For this reason, it is recommended that you pay particular attention to the following points:
Try to limit the number of SQL queries run during a 1-minute period to a minimum. In particular, try to avoid querying the same data more than once.
Use simple SQL queries to access the required data. If particular table columns are not required, they should be dropped from the returned query.
If large volumes of data are required to be handled, you should consider the use of a separate export database. The procedure to configure an alternative database is fully described in Appendix B of the Oracle Real User Experience Insight Installation Guide.
This section explains the structure of the database tables generated by RUEI for Enriched data exchange. The table used for KPI data export is explained in Section R.3, "KPI Data Exchange Database Table Structures". These tables are located in the database (local or remote) used by your RUEI installation. They can be accessed through SQL queries.
At the highest level, the WG__BIDATA_PERIOD table, shown in Table R-1, provides an outline of the available exported data. While the export data tables do not enforce referential integrity, the PERIOD_ID (minutes since 1970 UTC) column provides a link to other related tables.
The STAMP column indicates the 1-minute interval during which the data export was triggered. As data is purged from the database, according to the specified Enriched data exchange retention policy (see Section 12.9.1, "Defining Reporter Retention Policies"), rows are removed from this table. Similarly, rows are added to the table every one minute. Note that a new row will only appear in the table if exporting of the associated 1-minute period has been completed. The availability of export data is determined by the Enriched data exchange retention setting described in Section 12.9.1, "Defining Reporter Retention Policies".
The actual export data is held in the WG__BIDATA_MASTER table, show in Table R-2. Within each PERIOD_ID, each page view receives a unique PAGEVIEW_ID, which is incremented for each page view. When a new PERIOD_ID is encountered, the PAGEVIEW_ID numbering re-starts from 1. The STAMP column specifies the page view's true timestamp, rather than a 1-minute interval. Other columns specify the page view's properties. Note that the SESSION_ID column provides a link to the viewed pages within specific sessions.
Table R-2 WG__BIDATA_MASTER Table
Column | Type |
---|---|
PERIOD_ID* |
NUMBER |
PAGEVIEW_ID* |
NUMBER |
APPLICATION |
VARCHAR2 (4000 BYTE) |
BROWSER_DETAIL |
VARCHAR2 (4000 BYTE) |
BROWSER_TYPE |
VARCHAR2 (4000 BYTE) |
CLIENT_CITY |
VARCHAR2 (4000 BYTE) |
CLIENT_COUNTRY |
VARCHAR2 (4000 BYTE) |
CLIENT_IP |
VARCHAR2 (4000 BYTE) |
CLIENT_NETWORK |
VARCHAR2 (4000 BYTE) |
CLIENT_OS_CLASS |
VARCHAR2 (4000 BYTE) |
CLIENT_OS_VERSION |
VARCHAR2 (4000 BYTE) |
CLIENT_PROVIDER |
VARCHAR2 (4000 BYTE) |
CLIENT_REGION |
VARCHAR2 (4000 BYTE) |
CONTENT |
VARCHAR2 (4000 BYTE) |
COOKIE |
VARCHAR2 (4000 BYTE) |
DYNAMIC_NETWORK_TIME |
NUMBER |
DYNAMIC_SERVER_TIME |
NUMBER |
FULL_URL |
VARCHAR2 (4000 BYTE) |
HITS |
NUMBER |
HTTP_RESULT |
VARCHAR2 (4000 BYTE) |
NAMED_CLIENT_GROUP |
VARCHAR2 (4000 BYTE) |
NAMED_CLIENT_NAME |
VARCHAR2 (4000 BYTE) |
NAMED_SERVER_GROUP |
VARCHAR2 (4000 BYTE) |
NAMED_SERVER_NAME |
VARCHAR2 (4000 BYTE) |
PAGE_DELIVERY_TYPE |
VARCHAR2 (4000 BYTE) |
PAGE_DELIVERY_VALUE |
VARCHAR2 (4000 BYTE) |
PAGE_GROUP |
VARCHAR2 (4000 BYTE) |
PAGE_LOAD_TIME |
NUMBER |
PAGE_NAME |
VARCHAR2 (4000 BYTE) |
PAGE_READ_TIME |
NUMBER |
REFERRER_URL |
VARCHAR2 (4000 BYTE) |
REPLY_HEADERS |
VARCHAR2 (4000 BYTE) |
REQUEST_HEADERS |
VARCHAR2 (4000 BYTE) |
SERVER_IP |
VARCHAR2 (4000 BYTE) |
SESSION_ID |
VARCHAR2 (4000 BYTE) |
SET_COOKIE |
VARCHAR2 (4000 BYTE) |
STAMP |
TIMESTAMP |
STATIC_NETWORK_TIME |
NUMBER |
STATIC_SERVER_TIME |
NUMBER |
SUITE_TYPE |
VARCHAR2 (4000 BYTE) |
URL_ARGUMENTS |
VARCHAR2 (4000 BYTE) |
URL_FILE |
VARCHAR2 (4000 BYTE) |
URL_PARAMS |
VARCHAR2 (4000 BYTE) |
URL_POST_ARGUMENTS |
VARCHAR2 (4000 BYTE) |
USER_ID |
VARCHAR2 (4000 BYTE) |
VHOST |
VARCHAR2 (4000 BYTE) |
Information about the reported data items in Table R-2 is available from Section D, "Summary of Data Items".
The WG__BIDATA_PROPERTIES Table
The WG__BIDATA_PROPERTIES table, shown in Table R-3, contains additional page view properties. Note that while each row in the WG__BIDATA_MASTER table refers to one page view, multiple rows in the WG__BIDATA_PROPERTIES table can refer to the same page view.
The TYPE column shown in Table R-3 indicates whether the item refers to a defined Enriched data exchange item, or to a custom dimension (as described in Section 3.11, "Working With Custom Dimensions"). The NAME column specifies the name of the page property. This is either a custom export item, or a custom dimension.
Table R-3 WG__BIDATA_PROPERTIES Table
Column | Type |
---|---|
PERIOD_ID* |
NUMBER |
PAGEVIEW_ID* |
NUMBER |
TYPE* |
VARCHAR2 (64 BYTE) |
NAME* |
VARCHAR2 (255 BYTE) |
VALUE* |
VARCHAR2 (4000 BYTE) |
Information about the custom items within a page view reported in the WG__BIDATA_MASTER table can be retrieved from the WG__BIDATA_PROPERTIES table using a SQL query based on the appropriate PAGEVIEW_ID.
The WG__BIDATA_SUITES table, shown in Table R-4, specifies the suite types for which export information is available. A suite type only appears in this table if a suite instance has been defined for the suite type.
The individual suite tables are essentially extensions of the WG_BIDATA_MASTER table, and provide the suite-specific information associated with each page view. An example, the WG__BIDATA_SUITE_EBS table, is shown in Table R-5.
Table R-5 WG__BIDATA_SUITE_EBS Table
Column | Type |
---|---|
PERIOD_ID* |
NUMBER |
PAGEVIEW_ID* |
NUMBER |
EBS_FORMNAME_ID |
VARCHAR2 (4000 BYTE) |
EBS_FORMNAME_NAME |
VARCHAR2 (4000 BYTE) |
EBS_FWK_NAME |
VARCHAR2 (4000 BYTE) |
EBS_JSP_FILENAME |
VARCHAR2 (4000 BYTE) |
EBS_MODULE_ID |
VARCHAR2 (4000 BYTE) |
EBS_MODULE_NAME |
VARCHAR2 (4000 BYTE) |
EBS_REGION_ID |
VARCHAR2 (4000 BYTE) |
EBS_REGION_NAME |
VARCHAR2 (4000 BYTE) |
EBS_RESP_KEY |
VARCHAR2 (4000 BYTE) |
EBS_RESP_NAME |
VARCHAR2 (4000 BYTE) |
The CLIENT_COUNTRY reported within the exported data is based on the ISO 3166-1 standard. This uses a 2-character abbreviation (for example, "AU" for Australia) to indicate the end-user's country location. However, in cases were it is not possible to determine the end-user's location, a number of special codes are reported. These are shown in Table R-6.
Table R-6 Exceptions to ISO 3166-1 Country Code Reporting
Code | Description |
---|---|
-- |
A local (rather than top level) domain name is used for a home network. |
A1 |
An anonymous proxy is being used as an intermediary for requests from the client. |
A2 |
Client access to the Internet is via an ISP satellite. |
EU |
A corporate proxy located in Europe is being used. |
AP |
A corporate proxy located in Asia or the Pacific region is being used. |
For the USA and Canada, the reported CLIENT_REGION is based on the ISO 3166-2 standard. This uses a combination of country code and region. For example, the Texas region of the USA would be reported as "US-TX". For locations in the rest of the world, the relevant FIPS 10-4 region codes are reported. For the special country codes shown in Table R-6, the region code is reported as "00". For example, "A1-00".
This section explains the structure of the database tables generated by RUEI for the export of KPI data. These tables are located in the database (local or remote) used by your RUEI installation. They can be accessed through SQL queries.
At the highest level, the WG__BIDATAKPI_PERIOD table, shown in Table R-7, provides an outline of the available exported KPI data. While the export data tables do not enforce referential integrity, the PERIOD_ID (minutes since 1970 UTC) column provides a link to the other KPI-related table, WG__BIDATAKPI_MASTER.
The STAMP column indicates the 1-minute period interval during which the data export was triggered. As data is purged from the database, according to the specified Enriched data exchange retention for KPIs policy (see Section 12.9.1, "Defining Reporter Retention Policies"), rows are removed from this table. Similarly, rows are added to the table every minute. Note that a new row will only appear in the table if exporting of the associated 1-minute period has been completed. The availability of KPI data is determined by the KPI data exchange retention setting described in Section 12.9.1, "Defining Reporter Retention Policies".
The actual KPI data is held in the WG__BIDATAKPI_MASTER table, shown in Table R-8. Within each PERIOD_ID, each KPI receives a unique KPI_ID.
Table R-8 WG__BIDATAKPI_MASTER Table
Column | Type | Description |
---|---|---|
PERIOD ID* |
NUMBER |
Timestamp. |
KPI_ID |
NUMBER |
Internal unique KPI ID. |
NAME |
VARCHAR (255 CHAR) |
User-defined KPI name. |
CATEGORY |
VARCHAR (255 CHAR) |
User-defined KPI category name. |
DESCRIPTION |
VARCHAR (255 CHAR) |
User-defined KPI description. |
TARGET_TYPE |
NUMBER |
(0=none, 1=fixed, 2=automatic). |
TARGET_MIN |
NUMBER |
KPI's minimum target value. |
TARGET_MAX |
NUMBER |
KPI's maximum target value. |
NUMERATOR |
BINARY_DOUBLE |
Numerator value for period. |
DENOMINATOR |
BINARY_DOUBLE |
Denominator value for period. |
VALUE |
NUMBER |
Current KPI value calculated over the SPAN period. |
STATUS |
NUMBER |
Status of KPI (-1=unknown, 0=fail, 1=okay). |
SPAN |
NUMBER |
Periods (in minutes) over which the KPI value is calculated. |
DATA_TYPE |
VARCHAR (255 CHAR) |
Data access definition for KPI ("null"=generic, "app"=application-generic, "suite"=suite-specific, "service"=service-specific)Foot 1 . |
SUITE_TYPE |
VARCHAR (255 CHAR) |
Data access suite type definition (for example, "EBS" and "Siebel"). |
FILTERS |
CLOB |
Dimension-level filters definitions. |
REQUIREMENTS |
CLOB |
Metric-level requirement definitions. |
Footnote 1 The precise values are based on those used in the GUI, and can change.