| Oracle® Health Sciences Clinical Development Analytics Installation and Configuration Guide for Informatica Release 3.2 E74822-01 |
|
|
PDF · Mobi · ePub |
| Oracle® Health Sciences Clinical Development Analytics Installation and Configuration Guide for Informatica Release 3.2 E74822-01 |
|
|
PDF · Mobi · ePub |
This chapter describes the OHSCDA Installation tasks that you must complete and includes the following sections:
Section 2.1, "Installing Oracle Health Sciences Clinical Development Analytics"
Section 2.2, "Adding InForm Studies to the OHSCDA Warehouse"
This section describes the OHSCDA Installation tasks that you must complete.
Note:
All references to ${DAC_INSTALL_DIR} in this document refer to DAC installation folder.
OHSCDA application installation consists of the following components:
The basic OHSCDA components are installed using the Oracle Universal Installer. The installer gathers all the information about the database connectivity, data mart, Informatica repository by presenting a sequence of prompt screens and then installs the components accordingly.
Important:
Make sure that both the Oracle Database client and the Informatica client are available on the same system where you plan to execute the installer.Perform the following steps to install the OHSCDA application on Windows:
Extract the contents of the media pack into a temporary directory (For example, C:\ocda_temp).
Navigate to the \install directory under the extracted temporary folder.
Double-click the setup.exe file to launch the Oracle Universal Installer with the Welcome screen.
The installer will take you through a series of screens having fields. Attend to the Installer's prompts. The following sections describe each Installer screen, and the required action.
The Welcome screen provides information about the Oracle Universal Installer. The following function buttons appear on the installation screens:
Deinstall Products: Deinstall individual components or the entire product. This button appears only on the Welcome screen.
About Oracle Universal Installer: View the version number of the installer in use.
Help: Access detailed information about the functionality of each screen.
Installed Products: View currently installed products or deinstall the entire product or components.
Next: Proceed to the next screen.
Cancel: Cancel the installation process and exit the installer.
Click Next.
The Specify Home Details screen appears.
Enter the name of the installation and path where all the staged files from the Installer get copied to the local system. This is also the location from where Installer executes the database script.
Example Name: OCDAHome1
Path: C:\OCDA
Click Next.
The Install or Upgrade OHSCDA screen appears.
Select one of the following options:
Install: If you want to make a fresh installation of OHSCDA.
Upgrade: If you want to upgrade an existing OHSCDA 3.0 instance.
Click Next.
The OHSCDA Data Warehouse Details screen appears.
This screen collects all the information regarding the OHSCDA data warehouse.
Enter values in the following fields:
Data Warehouse Connect String: Connection of the database where OHSCDA Data Warehouse will be deployed.
Note:
Make sure the connect string matches the value in the tnsnames.ora file.Database Host Name
Database Port No
Database Service Name
Default Tablespace Name
Temporary Tablespace Name
Click Next.
The OHSCDA Data Warehouse screen appears.
Enter and confirm the system password of the database where OHSCDA Data Warehouse will be deployed.
Click Next.
The OHSCDA Data Warehouse Details screen appears.
Enter and confirm the OHSCDA Warehouse schema (RXI) password.
Click Next.
Enter and confirm the OHSCDA RPD (RXI_RPD) password.
The OHSCDA Informatica PowerCenter Details screen appears.
This screen collects all information to connect to the Informatica server.
Enter values in the following fields:
Informatica Repository Name: Informatica Repository Name where you plan to import OHSCDA Informatica mappings.
Informatica Domain Name: Domain of the Informatica repository.
Informatica Hostname: Host name of the Informatica server.
Informatica Repository Port Number: Port number of the Informatica server.
Informatica Username with Admin Privileges: User with Admin privilege to import XMLs.
Click Next.
The OHSCDA Informatica Details screen appears.
Enter and confirm the Informatica password for admin user.
Click Next.
The OHSCDA Informatica PowerCenter Client Home Directory screen appears.
Enter the Informatica PowerCenter client home path.
For example,
Informatica PowerCenter client home: D:\Informatica\9.5.0\clients\PowerCenter client\client
Click Next.
The National Language Support screen appears.
Select one the following options:
Yes: If the source data contains data in English and any other language.
No: If the source data contains data only in English.
Click Next.
The Summary screen appears.
Verify the details provided in the Summary screen.
Click Install.
Note:
You must record the log file path that appears on the bottom of the screen for future reference.For each database that will be a source for extracting data into the OHSCDA warehouse, perform the following steps to create a Relational Connection in the Informatica Workflow Manager:
Launch the Informatica PowerCenter Workflow Manager.
Connect to the repository where OHSCDA Informatica mappings are imported.
Select Connections, and then select Relational to display the Relational Connection Browser.
Click New to display the Select Subtype dialog.
Select Oracle as database type, and then click OK. The Connection Object Definition dialog box is displayed with options for the selected database platform.
Enter values in the following fields according to the Source database connection:
Connection Name: Enter the logical connection name.
User Name: Enter the user name that can access source data (For example, rxa_des for Oracle Clinical.)
Password: Enter the database password.
Connection String: Enter the Connect string for connecting to the database.
Code Page: Enter UTF-8 encoding of Unicode.
Note:
These values will also be required in setting up DAC. Make note of them to ensure that you enter the same during DAC setup.The TNS entry for all your source and target databases must be added on Informatica Server.
Note:
You must repeat steps 1 through 4 for each required source connection.You will be using CDA_Warehouse.zip file in this section.
Create a new DAC repository, as an Administrator.
Unzip OCDA_HOME\oracle.pharma.ocda.standard\DAC_Code\CDA_Warehouse.zip onto the computer where you will run DAC client.
Import the OHSCDA Warehouse Application metadata.
Start the Data Warehouse Administration Console (DAC) client.
From the Tools menu select DAC Repository Management, and then select Import.
Click Change import/export folder to navigate to the folder where you unzipped CDA_Warehouse.zip in step 2 of Section 2.1.3, "Preparing a DAC Repository for OHSCDA".
Click OK to display the Import dialog box.
Select the following categories of metadata you want to import: Logical, Overwrite log file, and User Data. Deselect the System check box.
Select CDA_Warehouse application in the ApplicationList.
Click OK.
Enter the verification code and Yes in the secondary window that is displayed after the import.
You can inspect the import log in ${DAC_ INSTALL_DIR}\log\import.log to verify if import is successful.
Configure Informatica Repository Service in DAC.
Navigate to the Setup view, and then select the Informatica Servers tab.
Click New to display the Edit tab below or select an existing Informatica server from the list.
If you are configuring a new installation, the Informatica Servers tab will be empty. If you are upgrading an existing installation, the Informatica Servers tab might contain existing Informatica servers.
Enter values in the following fields:
Name: Enter the Logical name for the Informatica server (for example, INFA_REP_SERVER).
Type: Select Repository.
Hostname: Enter the host system name where Informatica Server is installed.
Server Port: Enter the port number Informatica Server or Informatica Repository Server use to listen to requests.
Login: Enter the Informatica user login for the Admin user.
Password: Enter the Informatica Repository password.
Repository Name: Enter the Informatica Repository Name.
Test the connection to verify the settings.
Click Save to save the details.
Configure Informatica Integration Service in DAC.
Note:
Make sure that you use the same Login and Password that you have used in setting up Informatica.Click New to display the Edit tab below or select an existing Informatica server from the list.
If you are configuring a new installation, the Informatica Servers tab will be empty. If you are upgrading an existing installation, the Informatica Servers tab might contain existing Informatica servers.
Enter values in the following fields:
Name: Enter the Logical name for the Informatica server (for example, INFA_SERVER).
Type: Select Informatica.
Service: Informatica Integration Service Name associated with the Informatica repository added in Step 5.
Domain: Enter the Informatica domain name.
Login: Enter the Informatica Repository user login (Admin User).
Password: Enter the Informatica Repository password.
Repository Name: Enter the Informatica Repository Name.
Test the connection to verify the settings.
Click Save to save the details.
Configure source databases (Oracle Clinical, Siebel Clinical, and InForm) and the target database (the OHSCDA warehouse). For each database with which DAC will interact for OHSCDA, perform the following steps:
Navigate to the Setup view, and then select the Physical Data Sources tab.
Click New to display the Edit tab below or select an existing database connection from the list.
Enter values in the following fields:
- Name: Enter the Logical name for the database connection.
- Type: Select Source when you create the database connection for a transactional (OLTP) database. Select Warehouse when you create the database connection for a data warehouse (OLAP) database.
- Connection Type: Select a connection type for the database connection.
- Instance or TNS Name: Enter the Data Mart database instance name.
- Table Owner: Enter the Data Mart schema name.
- Table Owner Password: Enter the Data Mart schema password.
- DB Host: Enter the Data Mart host name.
- Port: Enter the Data Mart host port.
- Dependency Priority: Enter the user-defined priority of the data source.
- Data Source Number: Enter the user-defined number of the data source.
- Num Parallel Indexes Per Table: Enter a number to specify how many indexes are to be created in parallel.
Test the connection to verify the settings.
Click Save to save the details.
Note:
By Default, Oracle supports Data Source Number 1, 2, and 3 for Oracle Clinical, Siebel Clinical, and InForm respectively.The logical Names of the connections in DAC must be same as the connection names created in Informatica Workflow Manager.
OHSCDA Warehouse connection must be the same as provided during OHSCDA installation.
Table 2-1 Sample of Data Source Table W_RXI_DATASOURCE_S
| ROW_WID | DATASOURCE_NUM_ID | DATASOURCE_NAME | DELETE_FLG | INTEGRATION_ID | ENTERPRISE_ID | INFORM_STUDY_GUID |
|---|---|---|---|---|---|---|
|
1 |
1 |
ORACLE_CLINICAL |
N |
ORACLE_CLINICAL |
0 |
- |
|
2 |
2 |
SIEBEL_CLINICAL |
N |
SIEBEL_CLINICAL |
0 |
- |
|
3 |
3 |
INFORM |
N |
INFORM |
0 |
{FA5BCD87-B858-4F08-8BF8-1914B61C9DDF} |
|
4 |
101 |
INFORM_1 |
N |
INFORM_1 |
0 |
{CC0DBD73-FA40-4943-BDB3-60752C1A2732} |
Note:
If InForm is one of your sources and auto merge feature is not used, then the column INFORM_STUDY_GUID must be populated with studyguid for the corresponding InForm study.You can find the distinct studyguid from table RXI_STUDYVERSIONS in the consolidation database.
From DAC, the value passed for the data source parameter is always 3. Since we have one consolidation DB which holds all study data, the guid, when auto merge is set to N, helps resolving individual data source numbers.
If auto merge is set to Y, then a single entry for data source 3 for INFORM is sufficient and INFORM_STUDY_GUID need not be populated. In this case, the consolidation DB is considered as single source and duplicates are handled by auto merge.
Perform the following steps to modify the value for data sources:
Navigate to the Execute view, and then select the Execution Plans tab.
If Oracle Clinical, Siebel Clinical, and InForm are your source systems, select CDA - Complete Warehouse from the list.
If Oracle Clinical is your only source system, select CDA - Oracle Clinical Warehouse from the list.
If Siebel Clinical is your only source system, select CDA - Siebel Clinical Warehouse from the list.
If InForm is your only source system, select CDA - InForm Warehouse from the list.
Click Parameters subtab in the bottom pane.
For each row with TYPE equal to DATASOURCE, in the Value field, select the appropriate Physical Data Source Name from the drop-down list for the field.
Select a relevant value from the list for each of the data sources.
Click Save.
Note:
If Oracle Clinical is your only source system, use CDA - Oracle Clinical Warehouse.
If Siebel Clinical is your only source system, use CDA - Siebel Clinical Warehouse.
If InForm is your only source system, use CDA - InForm Warehouse.
Navigate to each of the containers and make sure that values are set for each of the available parameter.
Table 2-2 lists the DAC configurable parameters.
Table 2-2 DAC Configurable Parameters
| Parameter | Description |
|---|---|
|
START_TS |
This is the last refresh time of the source tables minus prune days. (@DAC_SOURCE_PRUNED_REFRESH_TIMESTAMP) |
|
END_TS |
Current Execution Plan's actual start time adjusted to source database time zone minus prune days. (@DAC_ETL_START_TIME_FOR_SOURCE) |
|
DATASOURCE_NUM_ID |
The ID associated with every source system. The default ID is 1 for Oracle Clinical, 2 for Siebel Clinical, and 3 for InForm. |
|
ENTERPRISE_ID |
The ID associated for every tenant. The default value is 0. |
|
DELETE_FLOW |
The default value is N and set it to Y if Deletes have to be captured in the data warehouse. |
|
EMAIL_SUFFIX |
You can provide domain name as a suffix to username. For example: oracle.com |
|
Prune Days |
This is used for setting the END_TS for incremental load. |
|
MPI_AUTHFILE |
Currently, not in use. |
|
MPI_USER |
Currently, not in use. |
|
AUTO_MERGE_FLG |
This is used to enable or disable auto-merge.
|
|
$OutputFile_OCDA |
Currently, not in use. |
|
$DBConnection_SP_OLAP |
Currently, not in use. |
|
$DBConnection_OLAP |
Currently, not in use. |
You will be using help.zip and images.zip files in this section.
You need to manually deploy OHSCDA's help and images files on Oracle WebLogic Managed Server. Perform the following steps:
Navigate to <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\ in Oracle WebLogic Server.
Create the following folders at <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\
s_ocda
Unzip help.zip from OCDA_Home\oracle.pharma.ocda.standard\Reporting\Help to <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\s_ocda\
Move customMessages and sk_ocda folders from <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\s_ocda\help\ to <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\.
Add the following tag in instanceconfig.xml, present in <MIDDLEWARE_HOME>\instances\instance1\config\OracleBIPresentationServicesComponent\coreapplication_obips1:
<UI><DefaultSkin>ocda</DefaultSkin>
<DefaultStyle>ocda</DefaultStyle></UI>
Unzip the Images.zip files from OCDA_Home\oracle.pharma.ocda.standard\Reporting\Images to <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\s_ocda
Launch Oracle WebLogic Administration Server Console.
Example: https://hostname.domain:port/console
This opens the Oracle WebLogic Server Administration Console.
Log on to Oracle WebLogic Server Administration Console as an Administrator.
In the left pane of the Administration Console, select Deployments.
On the left pane, click Lock & Edit.
In the right pane, click Install.
This opens the Install Application Assistant.
In the Path field browse to <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\.
Select analyticsRes and click Next.
Select Install this deployment as an application and click Next.
In Available targets for analytics, select the servers in the cluster on which you want to deploy OHSCDA.
Click Next.
In the Deployment targets, select bi_server1.
Click Next.
Select I will make the deployment accessible from the following location option for <MIDDLEWARE_HOME>\instances\<instancename>\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\analyticsRes\ in the Source accessibility section.
Click Finish.
analyticsRes appears in the Deployment section.
Click Activate Changes.
Select analyticsRes in the Deployment section.
Click Start to view the list and select Servicing all requests. The Start Application Assistant page is displayed.
In the content pane of the new page, click Yes to start the selected deployment.
State of analyticsRes should be Active after this deployment. You may confirm the same on Deployments page.
Log out from Oracle WebLogic Server Administration Console.
Log on to Oracle Enterprise Manager Fusion Middleware Control.
Restart the BI components.
Log on to OBIEE and verify the branding and help links on the dashboards.
See Also:
Oracle WebLogic Server Documentation Library
You will be using OCDA.zip and OCDA.rpd files in this section.
To deploy InForm web catalog, copy OCDA.zip from OCDA_Home\oracle.pharma.ocda.standard\Reporting\Webcat\Inform_Webcat to OBIEE server.
Else, copy OCDA.zip from OCDA_Home\oracle.pharma.ocda.standard\Reporting\Webcat to OBIEE server.
Copy OCDA.rpd files from OCDA_Home\oracle.pharma.ocda.standard\Reporting\RPD to OBIEE server.
Unzip OCDA.zip in the following folder:
Windows32 - <DRIVE>:\<MIDDLEWARE_HOME>\instances\instance1\bifoundation\OracleBIPresentationServicesComponent\coreapplication_obips1\catalog
UNIX - /<MIDDLEWARE_HOME>/instances/instance1/bifoundation/OracleBIPresentationServicesComponent/coreapplication_obips1/catalog
For fresh installation, create a TNS entry of OHSCDA database in %ORACLE_BI%\network\admin.
Create an ODBC entry (System DSN) to connect to (RXI) using Oracle Database 11g client driver.
In the Oracle BI Administration Tool, select File, then Open, and then Offline.
Navigate to the OCDA.rpd, and then click Open.
Password: Admin123
Click OK.
In the Oracle BI Administration Tool, select File, then Change Password.
Enter the current (old) password.
Enter the new password and confirm it.
Confirm the new password.
Click OK.
Modify the connection pools in the RPD as following:
In the physical layer, expand the OHSCDA Data Warehouse node and double-click Connection Pool object to open Connection Pool dialog box.
Set Data source name field to the ODBC entry for the warehouse you created in step 3.
Change the username to the username of the OHSCDA read-only schema. (RXI_RPD)
Change the password to the password of the OHSCDA read-only schema. (RXI_RPD)
Click OK.
Reconfirm the password.
Click OK.
From the File menu, select Save to save the rpd.
Click Yes for Do you wish to check global consistency?
Click Close in the Consistency Check Manager.
Click Save.
Click File and then click Close.
Click File and then click Exit.
Open NQSConfig.xml found at <MIDDLEWARE_HOME>instances\instance1\config\OracleBIServerComponent\coreapplication_obis1.
Set EVALUATE_SUPPORT_LEVEL =2.
Start the Oracle WebLogic Server and BI components.
Open the Fusion Middleware Control URL from the system where you saved the OCDA.rpd in step 13. The URL includes the name of the host and the port number assigned during the installation. The following shows the format of the URL:
https://hostname.domain:port/em
The login page is displayed.
Note:
Oracle recommends that you enable HTTPS on middle-tier computers that are hosting the Web services, since otherwise the trusted user name and password that are passed can be intercepted.Enter the Oracle Fusion Middleware administrator user name and password and click Login.
Expand the Business Intelligence folder and select the coreapplication node.
The Overview page displays the current status of the system, by providing information about current availability, performance, and issues identified within the BI domain. The Overview page also enables you to start and stop Oracle Business Intelligence.
Navigate the Repository tab of the Deployment page.
Click Lock and Edit Configuration.
Click Close.
In the Upload BI Server Repository section, click Browse and navigate to select the RPD.
Enter the RPD password in Repository Password and Confirm Password fields.
In the BI Presentation Catalog section, enter <ORACLE_INSTANCE>/bifoundation/OracleBIPresentationServicesComponent/<COMPONENT_NAME>/catalog/OCDA in the Catalog Location field.
Click Apply, and then click Activate Changes.
Return to the Business Intelligence Overview page and click Restart.
To configure maps:
Go to http://www.oracle.com/technetwork/middleware/bi-foundation/obieesamplesarchive-2026956.html.
Navigate to OBIEE 11.1.1.3 - Sample Application.
Click Documentation Downloads.
Click Save.
Unzip the folder.
Open SampleApp_Deploy_Instructions_825.pdf.
For instructions on how to configure maps, see section 6 Deploying SampleApp Mapviewer Content.
Perform the following steps to create groups:
Launch Oracle WebLogic Server Administration Console.
In Oracle WebLogic Server Administration Console, select Security Realms from the left pane and click the realm you are configuring. For example, myrealm.
Select Users and Groups tab, then Groups.
Click New.
In the Create a New Group page provide the following information:
Name: Enter the name of the group. Group names are case insensitive but must be unique. See online help for a list of invalid characters.
(Optional) Description: Enter a description.
Provider: Select the authentication provider from the list that corresponds to the identity store where the group information is contained. DefaultAuthenticator is the name for the default authentication provider.
Click OK.
The group name is added to the Group table.
Repeat step 4 through 6 with the following values:
Table 2-3 Security Group Parameters
| Name | Description | Provider |
|---|---|---|
|
OCDA-CRA |
(Optional) |
DefaultAuthenticator |
|
OCDA-DataManager |
(Optional) |
DefaultAuthenticator |
|
OCDA-ExecutiveManager |
(Optional) |
DefaultAuthenticator |
|
OCDA-RegionManager |
(Optional) |
DefaultAuthenticator |
|
OCDA-StudyManager |
(Optional) |
DefaultAuthenticator |
|
OCDA-WebcatAdim |
(Optional) |
DefaultAuthenticator |
Perform the following steps to create groups:
Log on to Oracle Enterprise Manager Fusion Middleware Control.
From the target navigation pane, open Business Intelligence and select coreapplication.
Right-click coreapplication, then select Security to display a submenu with Application Policies and Application Roles as options.
Select Application Roles. The Application Roles page is displayed.
Click Create to display the Create Application Role page. Complete the fields as follows:
In the General section:
Role Name - Enter the name of the Application Role.
(Optional) Display Name - Enter the display name for the Application Role.
(Optional) Description - Enter a description for the Application Role.
In the Members section, select Add Group. To search in the dialog box that displays:
Click the blue button to search. It will display a list of all the groups.
Select from the results returned in the Available box.
Use the shuttle controls to move the desired name to the Selected box.
Click OK to return to the Create Application Role page.
Repeat steps 5 and 6 for all the Roles listed in the following table:
Table 2-4 Application Roles and Groups
| Role Name | Description | Groups |
|---|---|---|
|
OCDA-CRA |
(Optional) |
OCDA-CRA |
|
OCDA-DataManager |
(Optional) |
OCDA-DataManager |
|
OCDA-ExecutiveManager |
(Optional) |
OCDA-ExecutiveManager |
|
OCDA-RegionManager |
(Optional) |
OCDA-RegionManager |
|
OCDA-StudyManager |
(Optional) |
OCDA-StudyManager |
|
OCDA-WebcatAdim |
(Optional) |
OCDA-WebcatAdim |
This section describes how OHSCDA extracts InForm's operational data and loads it into the OHSCDA warehouse. All of the data for each InForm study is maintained in a transactional database (actually a database schema). InForm maintains one or more InForm databases per Oracle database instance. Each study also has a Reporting and Analysis database (again, a set of objects in a schema, not necessarily an exclusive database instance) which contains data in views designed for query, as opposed to transactional activity.
Inform maintains the Reporting and Analysis database for each study in a study-specific schema. For efficiency and maintainability, OHSCDA has created a program that extracts data from a number of InForm Reporting and Analysis databases and places that data into a single consolidated database. Then, OHSCDA's source-dependent extract code reads from the consolidated database, and writes to the OHSCDA warehouse.
For the remainder of this section, refer to any given InForm Reporting and Analysis database as a source schema and the schema in which OHSCDA consolidates InForm data as the consolidation schema.
OHSCDA consolidates data from selected InForm source schema views, the selected ones being the ones that have data corresponding to columns in the OHSCDA data model. The following are the InForm source views that are consolidated for OHSCDA:
IRV_AF_SUBJECT_FORMS
IRV_CUR_QUERY
IRV_CUR_RULE
IRV_CUR_SITE
IRV_CUR_SPONSOR
IRV_CUR_SUBJECT
IRV_CUR_USER
IRV_FORM_REVS
IRV_ITEMSET_REVS
IRV_ITEM_REVS
IRV_QUERY_STATE_CHANGES
IRV_SECTION_REVS
IRV_STUDYVERSIONS
IRV_STUDYVERSION_FORMS
IRV_STUDYVERSION_VISITS
IRV_SUBJECT_STATE_CHANGES
IRV_SV_SUBJECTVISITS
IRV_USERS_SITES
For each of these source views, the consolidation schema contains a consolidation table. Each consolidation table has the same columns as its source view, except that the consolidation table has three additional columns:
STUDYGUID
CONSOLIDATION_DATETIME
MODIFIEDDATETIME
STUDYGUID is needed to retain the association of each record with its InForm study. CONSOLIDATION_DATETIME and MODIFIEDDATETIME allows OHSCDA ETL to detect which consolidated records have been added or updated since the prior ETL execution.
The names of tables in the consolidation schema match the names of their source views, with the following change:
IRV in the source view name is replaced with RXI in the consolidation table name.
Consolidation is the process of getting new or modified records from a source schema and updating the consolidation schema with those records. Consolidation is carried out by periodic execution of a scheduled job in each InForm schema that contributes to the consolidation schema. Execution is done through the DBMS_JOB facility. Each source schema is consolidated on its own schedule.
There is no master job that oversees or coordinates the consolidation of all the source InForm schemas. You can choose to consolidate different InForm schemas at different times, and with different intervals between consolidations. However, the more spread out consolidations are, the wider the time period represented by the records in the consolidation schema. So it is recommended that you schedule consolidation jobs to fall within a small time window. For details on scheduling, see Section 2.2.7, "Adjusting Scheduling of Consolidation Jobs".
When the Consolidation job executes for a particular source schema, it does the following:
For each source table from which OHSCDA extracts data,
Identify what records have been created or modified in this table since the previous consolidation
For each modified source record, drop the corresponding record in the consolidation table
Insert all new and modified records in the Consolidation schema table, setting their StudyGUID, and modification timestamps
The procedures that does this work reside in each InForm Reporting and Analysis Schema, they are placed there as part of the process of preparing that schema for consolidation.
Preparation for consolidation also creates an access account in the database holding the source schema. This account is the one that reads the Reporting views, and writes to the consolidation tables. This access account is named CDAxxx, where xxx is the Study name.
All movement of data is done by a pull from the Reporting and Analysis Database to the Consolidation database. This ensures that there is no path by which someone who can connect to the consolidation schema is able to modify the InForm database.
The instructions that follow tell how to do the initial setup of the consolidation schema (this also enrolls the first Study to be consolidated as part of the process) and then how to set up each additional InForm study so that it becomes a contributor to the Consolidation schema.
Before you begin:
Navigate to OCDA_HOME\oracle.pharma.ocda.standard\OCDA_Common_Scripts\Inform_Consolidate.zip.
Copy and unzip the InForm_consolidate.zip file to a server from which it is possible to access all InForm source database that contain studies that are to be consolidated and also to access the database in which the consolidation schema is located.
This section describes the steps to create the Consolidation Schema. As part of the process, it enrolls an initial InForm study as a source to be consolidated.
If one does not already exist, create a TNSNAMES.ORA entry for the Consolidation database instance on each InForm database server that contains a study that is to be consolidated.
Note:
This is not essential. If you prefer not to create the TNSNAMES entries, you can provide the complete connection string instead of the TNS name when specifying remote databases below.On the consolidation schema database server, add a TNSNAMES.ORA entry for each source schema database.
On consolidation schema database server, create a root directory for managing the files needed to configure and log consolidation of the source schemas.
We will refer to this directory as <$ConsolidationMgmtRoot>.
If trial source and target consolidation DB are in the same network, then a single <$ConsolidationMgmtRoot> is sufficient. Else, replicate the directory in both the servers.
During the setup of the source trial DB, use TrialCDA and logs folders. And, during the setup of the destination consolidation DB, use DestCDA and logs folder.
This section describes the steps you must follow to prepare a source schema for consolidation and perform the initial execution of consolidation of that schema. Subsequent consolidations occurs automatically at an interval you define as part of the preparation described here.
The setup and configuration is performed in the following two phases for each study trial:
Follow the instructions in this section for each InForm study that is to be consolidated for consumption by OHSCDA.
On either source or consolidation schema database server, based on the network architecture as described in Section 2.2.3.2, create a directory within <$ConsolidationMgmtRoot> /<Study-name> to hold the files for consolidation of this study.
Copy the contents of InForm_consolidate.zip into the <$ConsolidationMgmtRoot>/<Study-name>/TrialCDA directory.
Set your current directory to the <$ConsolidationMgmtRoot>/<Study-name> directory.
Edit CONFIGTRIALCDA.sql to set specific values of configuration parameters for this study. Make the following changes:
Table 2-5 Source Trial Configuration Parameters
| Parameter | Set value to |
|---|---|
|
Slash |
Use \ for Windows or / for Linux |
|
trial_name |
The name of the study. |
|
inform_release |
For InForm 4.6, enter R46. For later releases, enter R55. |
|
trial_tnsname |
Name in the local TNSNAMES file of the source database for the study. Can also supply entire TNS entry for the source database as a string. |
|
trial_schema_user |
Username of the InForm source schema. |
|
trial_schema_password |
Password of the InForm source schema. |
|
trial_ro_user |
This is OHSCDA reporting object schema user name and is defaulted to trail_name prefixed with 'cda'. |
|
trial_ro_password |
This is OHSCDA reporting object schema password and is defaulted to trail_name prefixed with 'cda'. You can change this password. |
|
trial_dba_user |
Database username who has the dba privileges. |
|
trial_dba_password |
Password for database username. |
|
trial_ts |
Name of a tablespace in the source schema database to which the access account for this study will be given access. |
|
trial_ts_temp |
Name of a temporary tablespace in the source schema database to which the access account for this study will be given access. |
Save the file.
Connect to SQLPlus. Execute the following command:
sqlplus /nolog
From the command prompt, execute the following command:
@Setup_Trial_CDA.sql
Navigate to the <$ConsolidationMgmtRoot>/<Study-name>/Logs directory to review the log.
If any errors are found, make necessary corrections and re-execute Setup_Trial_CDA.sql.
Follow the instructions in this section for each InForm study that is to be consolidated for consumption by OHSCDA.
On the consolidation schema database server, based on the network architecture as described in Section 2.2.3.2, "Creating a Consolidation Management Root Directory", create or reuse a directory within <$ConsolidationMgmtRoot> /<Study-name> to hold the files for consolidation of this study.
Copy or reuse the contents of InForm_consolidate.zip into the <$ConsolidationMgmtRoot>/<Study-name> directory.
Set your current directory to the <$ConsolidationMgmtRoot>/<Study-name>/DestCDA directory.
Edit CONFIGDESTCDA.sql to set specific values of configuration parameters for this study. Make the following changes:
Note:
If you prefer not to create the TNSNAMES entries, you can provide the complete connection string instead of the TNS name when specifying the remote databases below.The trial parameters mentioned in Table 2-6 must match with the ones defined in Table 2-5.
Table 2-6 Destination Configuration Parameters
| Parameter | Set value to |
|---|---|
|
Slash |
Use \ for Windows or / for Linux |
|
Parameters for configuring the source schema and database |
|
|
trial_name |
The name of the study. |
|
inform_release |
For InForm 4.6, enter R46. For later releases, enter R55. |
|
trial_tnsname |
Name in the local TNSNAMES file of the source database for the study. Can also supply entire TNS entry for the source database as a string. |
|
trial_schema_user |
Username of the InForm source schema. |
|
trial_global_name |
Global name of the source schema database. You can get this by connecting to the source schema and running the following query: SELECT * FROM GLOBAL_NAME |
|
trial_ro_user |
This is OHSCDA reporting object schema user name and is defaulted to trail_name prefixed with 'cda'. |
|
trial_ro_password |
This is OHSCDA reporting object schema password and is defaulted to trail_name prefixed with 'cda'. You can change this password. |
|
Parameters for configuring the target schema and database |
|
|
dest_tnsname |
Name of the target database in the TNSNAMES file on the source database server for the study. Can also supply entire TNS entry for the target database as a string. |
|
dest_schema_owner |
This defaults to and is same as trial_ro_user parameter. |
|
dest_schema_password |
This defaults to and is same trial_ro_password parameter. |
|
dest_consolidated_user |
This is the consolidation schema username. Default is RXI_INFORM. |
|
dest_consolidated_password |
This is the consolidation schema password. Defaults is RXI_INFORM. You can change this password. |
|
dest_dba_user |
Database username who has dba privileges on the consolidation database. |
|
dest_dba_password |
Password for the database username on the consolidation database. |
|
dest_ts |
Name of a tablespace in the target schema database to which the consolidation and access account will be given access. |
|
dest_ts_temp |
Name of a temporary tablespace in the target schema database to which the consolidation and access account will be given access. |
|
consolidate_job_interval |
The interval between executions of the program to consolidate new and modified data from this study. A value of 1 results in an interval of one day. A value of 1/24 results in an interval of one hour. |
Save the file.
Connect to SQLPlus. Execute the following command:
sqlplus /nolog
From the command prompt, execute the following command:
@Setup_Dest_CDA.sql
Navigate to the <$ConsolidationMgmtRoot>/<Study-name>/Logs directory to review the log.
If any errors are found, make necessary corrections and re-execute Setup_Dest_CDA.sql.
If InForm is one of the source systems, connect to the InForm Consolidation database and execute the cda_cdb_automerge_ddl.sql script located in the OCDA_HOME/oracle.pharma.ocda.standard/DataModels/ directory.
After you have initiated at least one consolidation and created Auto Merge tables, perform the following steps before running ETL against the consolidation schema:
Connect to the Consolidation schema database as SYSTEM.
Execute script OCDA_INFORM_VIEWS.sql located in the OCDA_HOME/oracle.pharma.ocda.standard/DataModels/ directory.
As mentioned earlier, initial consolidations, especially of large source schemas, should be done in isolation to avoid contention for resources. But once the initial consolidation of a schema is done, subsequent consolidations of that schema are incremental and involve smaller numbers of records (incremental consolidations are limited to new or modified Subjects, CRFs, and Queries accumulated since the previous consolidation).
Once initial consolidation of a source schema is complete, it is useful to adjust the scheduling of its incremental consolidations so that they start near to the time of other incremental consolidation of other source schemas. The closer in time consolidations start across studies, the smaller the time period represented in the consolidation schema.
To adjust the time when a study's consolidation job executes, perform the following steps:
Connect as the source schema owner.
Determine the consolidation job ID using the following query for InForm 4.6:
select job from user_jobs where what like 'REFRESH_PROCS.REFRESH_WITH_STREAMS_DELAY%';
or the following query for InForm 5.5, 6.0, and 6.1:
select job from user_jobs where what = 'CDA_PROCS.CONSOLIDATE';
Change the time for the next execution of the job by using the dbms_job.next_date procedure.
Note:
The next_date parameter is an Oracle date-time in which you can specify both the date and time for the next execution.Remember to follow the call with a commit. For example,
begin
dbms_job.next_date (
job => 12345,
next_date => to_date('21-AUG-2013@01:01:00','DD-MON-YYYY@HH24:MI:SS')
);
end;
commit;
/
This causes job 12345 to next be executed at 1:01 AM on 21-AUG-2013.
You can also change the interval between subsequent executions of the job using the dbms_job.interval procedure. Set the interval parameter to the string 'SYSDATE + <desired-interval>', replacing <desired-interval> with the number of days, or the fraction of a day to elapse before the next execution.
begin dbms_job.interval( job => 12345, interval => 'SYSDATE + 1/2' ); end; commit; /
This sets the interval between executions of job 12345 to half a day or every 12 hours.
This section describes procedures you must perform for certain OHSCDA sources. Please review each section and determine if it applies to the source(s) you are using for OHSCDA. If the section does apply, carry out the procedure. If the section is optional, decide whether it is appropriate for your requirements and act accordingly.
Different source applications handle the deletion of records in different ways. Oracle Clinical allows deletions of records in a way that avoids orphans, and has built-in auditing of these deletions. Siebel Clinical also allows deletions, properly cascading to maintain referential integrity, but does not have built-in auditing of deletions. InForm does not provide any means to hard-delete records in its transactional database. So, it does not need to be concerned neither with maintaining integrity in the face of deletions nor with auditing deletions.
OHSCDA's ETL consults the audit trails to determine if source records have been deleted, and then soft-deletes those records in the warehouse.
OHSCDA defines an optional process that you can use to manage hard deletion of records in Siebel Clinical. You create table and triggers in the source system to handle deletion of records. For a description of this process, see Oracle Clinical Development Analytics Administrator's Guide.
Since no deletions are supported through the InForm UI, OHSCDA makes no provision for detecting them. If you need to delete records from an InForm transactional table at the backend and want the OHSCDA warehouse to be consistent with that change, you must identify the warehouse records that arose from the InForm records that are to be deleted. Also, you must remove them from the consolidation schema and the warehouse.
Depending on how you use OHSCDA, you may prefer to have reports that give information only on currently active studies. If so, and if source studies are completed or possibly deleted, the recommended approach for removing such a study from OHSCDA is to use OHSCDA's access control capability. That is, enable Study and Study-site access controls, and then exclude the study and its sites from the list of studies (and sites) available to OHSCDA users. Aside from being simpler than identifying and deleting records from the warehouse, this approach has the advantage of being reversible if the decision to remove the study from the warehouse is reconsidered.
If you do not wish to use access control to avoid display of deactivated or deleted studies, there is an alternative solution. In the OHSCDA warehouse, every dimension table, and every base fact table, has a column named DELETE_FLG. OHSCDA's OBIEE repository is designed to treat records where DELETE_FLG equals 'Y' as if they were not present in the warehouse. OHSCDA's SIL also disregards records with DELETE_FLG = 'Y' when calculating values for aggregate tables (those with names ending with _A). So you can cause records to be effectively removed from the warehouse by setting their DELETE_FLG to 'Y'. Thus, if a study has been removed from a source database, you could identify all dimension and fact records for that study, and set their DELETE_FLG to 'Y'. This causes them to be forever invisible to OHSCDA's analytics.
This solution could be used, but only with great caution, for a study that is considered to be deactivated, but still remains in the source. Once marked as deleted, warehouse records for a study will not get their DELETE_FLG set back to 'N', but OHSCDA creates new warehouse records if there are changes or additions in the source for that study. So, if the study is truly inactive in the source, soft-deleting warehouse records for the study by setting DELETE_FLG to 'Y' will work. But any changes or additions to the study would cause new records to be loaded into the warehouse for the study, and it would re-appear in analyses.
To optimize source system integration with OHSCDA, perform the following steps:
For OC 4.6, navigate to Informatica Relational Connections > OC Connection Object > Attribute in Informatica Workflow Manager.
Set Connection Environment SQL value as:
For Oracle Database 11.2.0.2:
alter session set optimizer_features_enable='11.2.0.2'
For Oracle Database 11.2.0.3:
alter session set optimizer_features_enable='11.2.0.3'
For Oracle Database 11.1.0.7:
alter session set optimizer_features_enable='11.1.0.7'
Create the following indexes:
Function based index on DISCREPANCY_ENTRIES - NVL ("MODIFICATION_TS", "CREATION_TS")
Function based index on DISCREPANCY_ENTRY_REVIEW_HIST - NVL ("NEXT_STATUS_TS", "CREATION_TS")
Function based index on RECEIVED_DCIS - NVL ("MODIFICATION_TS", "RECEIVED_DCI_ENTRY_TS")
Function based index on RECEIVED_DCMS - NVL ("MODIFICATION_TS", "RECEIVED_DCM_ENTRY_TS")
|
 Copyright © 2010, 2016, Oracle and/or its affiliates. All rights reserved. Legal Notices |
|