There are many ways to extract data from Oracle SaaS cloud to local on-Premise or On Cloud data warehouse. In this blog I have highlighted one such approach ( at very high level) , for additional details feel free to reach me at inquiry@bizinsightinc.com.
Step #1
Data Extraction
- SCM Cloud
- Financial Cloud
For data extraction from SCM/Financial cloud I have configured BICC , it is very easy to configure BICC and it's part of your SaaS ( no additional cost involve in that) , and configurations are very quick
- Metadata as well sample data for each extract is available.
- At any point you can Switch back to full load ( just in case there is any data corruption in your warehouse).
For Delivery of extracts , I have configured UCM server , with File age = 3 days.
On completion BICC extracts are will creates 2 files in UCM server
- Manifest.mf
- Data files in .zip formart
For WMS Cloud
For WMS cloud , I have configured Data Extracts ( as part of Job Schedules) and configured SFTP Server as destinations for data files.
On completion data extracts for WMS are creating .csv file in SFTP server
Step#2
Data files Consumption
On successful completion BICC extracts would create files in UCM server. Once BICC extracts are completed , I have a scheduled OIC job , and on execution this OIC job will
- Process and download all the latest manifest.mf from UCM server
- Download all the .zip files ( listed in manifest.mf) from UCM server
- unzip zip files ( to extract .csv files) in SFTP serever
- Archive the .zip and .mf files
- No such step is required for WMS extracts , as WMS always send .csv files to SFTP server
Step#3
Data File Load
This is the last step. As part of this step I have multiple OIC jobs ( one for each Module , DOO/AP/AR/WMS/PO/Product etc).
These are scheduled jobs and on execution these jobs will pick the respective files and load data into data warehouse . For example
OIC job for DOO will
- Pick Order Header/Order Line/Fulfillment line /EFF files from SFTP server.
- Read these files
- Load data into Specific DOO table created in datawarehouse
Step#4
To handle errors at various level I have design and develop error handling mechanism at OIC level.
No comments:
Post a Comment