Data Applications → Install Oracle EBS for Incorta On-Premises
About the Oracle EBS Data Application for Incorta On-Premises
The Incorta Oracle E-Business Suite (EBS) data application provides a new, faster, more effective option for analyzing and understanding your data, with no data modeling or extract, transform, load (ETL) required. You will get instant access to up-to-date, consolidated enterprise resource planning (ERP) data and can build all of the high-performance reports you need on your own.
With an extensive library of prebuilt content, Oracle EBS users gain immediate access to rich content areas and sample dashboards that are easily customizable and logically grouped by department and persona to help answer the most-pressing business questions quickly.
In addition, the Oracle EBS data applications expedite your organization’s migration from legacy reporting tools, and drastically speed your Oracle EBS-based reporting and analytics project’s time to value. They also let you more quickly add new data sources to your Oracle EBS analysis.
The following are the Oracle EBS modules for which data applications are available in the on-premises version of the Incorta Direct Data Platform™:
Financial Modules
- Accounts Receivable Analytics
- Fixed Assets
- General Ledger
- Payables
- Purchasing PO Module
- Revenue & Billing
Supply Chain Planning Modules
Business Flows
- Order to Cash
- Procure to Pay
Other Modules
- Discrete Manufacturing
- EAM (Enterprise Asset Management)
- EBS HR (Human Resources)
- Projects
The Supply Chain Planning Modules apply only to customers using Oracle Advanced Supply Chain Planning (ASCP) or Value Chain Planning (VCP).
Each data application module represents a 1:1 mapping of the corresponding tables and relationships in Oracle EBS. Also included is prebuilt metadata that can be used to load data for each module from the Oracle database and view it in prebuilt dashboards.
The Incorta on-premises EBS data application supports Oracle EBS version R12.x.
Install the Oracle EBS Data Application
The Oracle EBS data application requires the following ZIP file:
FullInstallEBSVer<VERSION_NUMBER>.zip
The Oracle EBS data application requires the import of a tenant file to an Incorta Cluster. A CMC Administrator can perform the tenant import. A systems administrator with root access to the host can copy the tenant data files, configure Spark, and configure the necessary datasources.
Here are the steps to install the Oracle EBS data application:
- Contact Incorta Support to obtain the Oracle EBS data application ZIP file. Download and unzip the file.
- Within the unzipped Oracle EBS data application file, locate the
ebsVer<VERSION_NUMBER>.zip
file.
Import the EBS Data Application Tenant
- Import the ebsVer<VERSION_NUMBER>.zip tenant:
- Sign in to the CMC.
- In the Navigation bar, select Clusters.
- In the cluster list, select a Cluster name.
- In the canvas tabs, select Tenants.
- Select + → Import Tenant.
- Drag and drop the
ebsVer_<VERSION_NUMBER>.zip
to the Click or drag a file here to upload panel in the Import a tenant to the cluster dialog. - Select Next.
- Verify or enter the tenant properties.
- Verify or enter the tenant email properties.
- Select Create.
Tenant Properties
Property | Control | Description |
---|---|---|
Name | text box | Enter the tenant name. Select Check to determine if a Tenant already exists with the name entered. |
Username | text box | Enter the username for the Super User |
Password | text box | Enter the password for the Super User |
text box | Enter the email address for the Super User | |
Path | text box | Enter the shared storage path for tenant related data |
Pause scheduled jobs | toggle | Enable this property if the imported tenant will have all scheduled jobs paused on import |
Tenant Email Properties
Property | Control | Description |
---|---|---|
Sender’s Username Auth | toggle | Enable this property if the email requires username authentication |
System Email Username | text box | Enable Sender’s Username Auth to configure this property. Enter the username for the system email. |
System Email Address | text box | Enter the email address for the system email |
System Email Password | text box | Enter the password for the system email address |
SMTP Host | text box | Enter the Simple Mail Transfer Protocol (SMTP) host for the system email |
SMTP Port | text box | Enter the SMTP port number |
Share Notifications | toggle | Enable this property to share notifications |
Copy the EBS Data Application Tenant Data Files
The EBS data application package contains the following CSV data files that you must copy to the Tenant data folder:
contentTypeLookup.csv
Date_US.csv
Month_US.csv
Period.csv
Quarter.csv
Here are the steps to perform a secure copy of the CSV files using scp:
INCORTA_NODE_HOST=100.101.102.103cd ~/Downloads/FullInstallEBSVer5/Datascp -i ~/.ssh/host_pemkey.pem contentTypeLookup.csv Date_US.csv Month_US.csv Period.csv Quarter.csv incorta@${INCORTA_NODE_HOST}:/tmp/
Secure shell into the host.
ssh -i ~/.ssh/host_pemkey.pem incorta@${INCORTA_NODE_HOST}
Copy the CSV data files to the /<INCORTA_INSTALLATION_PATH>/Tenants/<TENANT_NAME>/data
directory in bash shell.
sudo su incortaINCORTA_INSTALLATION_PATH=/home/incorta/IncortaAnalyticscp /tmp/contentTypeLookup.csv /tmp/Date_US.csv /tmp/Month_US.csv /tmp/Period.csv /tmp/Quarter.csv $INCORTA_INSTALLATION_PATH/Tenants/bp_oracle_ebs/data
Configure Spark for the EBS Data Application
Allocate 10 GB or more to the Spark Worker process, and restart Spark:
- Edit the
<INCORTA__INSTALLATION_PATH>/IncortaNode/spark/conf/spark-env.sh
file with vim.
vim /home/incorta/IncortaAnalytics/IncortaNode/spark/conf/spark-env.sh
Set the SPARK_WORKER_MEMORY value to 10g (or greater)
SPARK_WORKER_MEMORY = 10gRestart Spark
cd /home/incorta/IncortaAnalytics/IncortaNode./stopSpark.sh./startSpark.shCheck the Spark configuration using the Spark Master UI:
http://<HOST_NAME>:9091
.
Configure the EBS Data Application Datasources
- Sign in to the Incorta Direct Data Platform™.
- In the Navigation bar, select Data.
- In the Context bar, select the External Data Source tab.
Edit the EBS Data Source
- Select Edit (pen icon) to the right of the EBS datasource.
- Enter the EBS connection properties for your instance:
- Username
- Password
- Connection Pool: 10
- Connection String:
jdbc:oracle:thin:@<HOST>:<PORT>/<DATABASE>
Edit the VCP Data Source (for ASCP customers only)
- Select Edit (pen icon) to the right of the VCP datasource.
- Enter the VCP connection properties for your instance:
- Username
- Password
- Connection Pool: 10
- Connection String:
jdbc:oracle:thin:@<HOST>:<PORT>/<DATABASE>
Edit the Incorta (Metadata) Data Source
- Select Edit (pen icon) to the right of the Incorta datasource.
- Enter the EBS connection properties for your instance:
- Username
- Password
- Connection Pool: 10
- Connection String:
jdbc:mysql://<HOST>:<PORT>/<DATABASE>
Edit the SparkSQL Data Source
Select Edit (pen icon) to the right of the SparkSql datasource.
Enter the SparkSQL connection properties for your instance:
- Username
- Password
- Connection Pool: 10
- Connection String:
jdbc:postgresql://<HOST>:5442/<TENANT_NAME>
NoteThe port is the Data Store (DS) port defined in the CMC.
Configure the Oracle EBS Data Application
To configure the Oracle EBS data application, update the corresponding Session Variables and perform a full load of the EBS COMMON Schemas.
Update EBS Data Application Session Variables
- Sign in to the Incorta Direct Data Platform™.
- In the Navigation bar, select Schema.
- In the Schema Manager, select the Session Variables tab.
- Edit the query for the following variables with information for your EBS instance, by selecting Edit (pen icon) to the right of each variable:
Default_Book_Type_code
Default_Fiscal_Year
Default_Ledger
Default_Period
Default_Period_2
Default_Start_Date
Perform a Full Load of the EBS COMMON Schemas
- Sign in to the Incorta Direct Data Platform™.
- In the Navigation bar, select Schema.
- In the Schema Manager, select the Schemas tab.
- Select each of the schemas below. In the Action bar, select Load → Full Load.
EBS_ITEM_COMMON
EBS_CAL_COMMON
EBS_FND_COMMON
EBS_HR_COMMON
EBS_FIN_COMMON
EBS_PARTY_COMMON
To configure a specific EBS module, refer to the list of Financial Modules or Supply Chain Planning Modules and select the module for configuration details.
Monitor your Incorta host disk space as you load the EBS schemas. For large tables, to limit full data loads, add a SQL clause WHERE CREATION_DATE > <DATE>
.
EBS Data Application Security Groups
The EBS data application tenant includes the following security groups:
Feature | Supported |
---|---|
ReportAdmins | Analyze User, Schema Manager, User, User Manager |
ReportUsers | Individual Analyzer, User |
Additional Information
Flexfields
EBS has two types of flexfields, descriptive flexfields (DFF) and key flexfields (KFF). To include flexfields, you will need to either create custom tables or customize the SQL of existing tables.
For example, the out of the box EBS_FIN_COMMON schema for Chart of Accounts consists of three segment tables: GL_ACCOUNT_SEG, GL_COST_CENTER_SEG, and GL_BALANCING_SEG. To include KFF, open these tables in the Schema Designer and customize the SQL to support your implementation. You can also create additional segment tables as required.
Table Placeholders
Some tables have 1=0 in the SQL because they are not used in the data applications, and as a result, data is not loaded into the table. If these tables are needed for your implementation, you can modify the SQL to load the table.
FND Lookups
You can support FND_LOOKUPS by using a lookup() function in a formula field so that you do not need to join to the table. For example:
lookup(EBS_FND_COMMON.FND_LOOKUP_VALUES.MEANING, EBS_FND_COMMON.FND_LOOKUP_VALUES.LOOKUP_TYPE,'TRANSFER STATUS', EBS_FND_COMMON.FND_LOOKUP_VALUES.LOOKUP_CODE,EBS_PA.PA_COST_DISTRIBUTION_LINES_ALL.TRANSFER_STATUS_CODE , EBS_FND_COMMON.FND_LOOKUP_VALUES.LANGUAGE,'US')
EBS Hierarchies
For EBS hierarchies like Account or Cost Center, use EBS_GL.GL_SEG_VAL_NORM_HIERARCHY. For segments without multiple parents, use one of the following two approaches:
- Incorta self join hierarchy
- Flattened data
For segments that have multiple parents, further analyze the top nodes for the given segment. You can create a limited number of top nodes as separate attributes or hierarchies within Incorta.
Calendars
The Gregorian calendar is supported with a join to the EBS_CAL_COMMON.GREGORIAN_CALENDAR table. This table is loaded from the Date_US.csv
file.
The fiscal calendar is supported with a join to the EBS_CAL_COMMON.GL_DATE table.
Multi-Currency
Out of the box, amounts in Entered and Accounted currency are supported. To support other global currencies, such as USD, for the multi-ledger use case, use the EBS_FIN_COMMON.GL_DAILY_RATES table and customize it as required. Perform a lookup to this table to get the exchange rate and multiply it with the base amount. For an example, refer to one of the following fields in the CollectionsManager.Payments
schema:
Amount Due Original (AR Global)
Amount Due Original (Revenue Global)
Amount Due Remaining (AR Global)
Amount Due Remaining (Revenue Global)
Multiple Child Tables with a Common Parent
The Accounts Payable module has an example use case of multiple child tables with a common parent. AP_INVOICE_PAYMENTS_ALL is a child of AP_INVOICES_ALL. There is no direct path from AP_INVOICE_DISTRIBUTIONS_ALL to AP_INVOICE_PAYMENTS_ALL. As a result, it is not possible to report Payment Amount with Invoice Number and PO ID in one Incorta report. To resolve this issue, create a Materialized View that has the INVOICE_ID (key) and the aggregated payment amounts from the payments table, and make it the parent table of AP_INVOICES_ALL.