What is our primary use case?
Several projects benefited from the ODI data integration capabilities such as Loading OLTP system's data to Data warehouses, pulling data from cloud-based databases and loading to on-premises DW, for example: extracting data from a SQL server which is based on Azure infrastructure.
Perform data integration from cloud to cloud.
Data migration between OLTP systems.
Event-based integration such SOA, for example, BPEL performed calls of ODI scenarios based on EBS events.
How has it helped my organization?
All projects where I have implemented ODI solution highly benefited of high data volume integration.
What is most valuable?
There are several very important features that we use daily.
- All our systems can be widely integrated by ODI, such as transactional systems, our data warehouses, and B2B integration.
- ODI is really powerful for BI projects with traditional star-schema loads. For example, we integrated Salesforce to an Oracle data warehouse with it.ODI is a platform agnostic, you may have any data source and any target to integrate and ODI will enable you to connect.
- It facilitates real-time data replication from Microsoft SQL Server to an operational database sitting on Oracle Database.
- It provides for legacy systems integration. It integrates any SQL server database.
What needs improvement?
There is always room for improvement, even when something is really good, but I think ODI is one of the best ETL tools in the market. However, I've always waited for the web service feature to be improved in next versions, so at last the 12C version the web service has been improved, some of the limitations of the 11g version were solved. Also, it would be really good if Oracle considered enabling the tool to integrate with some other platforms that are deprecated simply for commercial reasons, although it looks like the cloud option offers some extra connectors (for example, in Salesforce CRM).
For how long have I used the solution?
More than five years.
What do I think about the stability of the solution?
I've never had any issues with stability.
What do I think about the scalability of the solution?
I haven't had any issues with scalability.
How are customer service and technical support?
Technical service is a 6, and Oracle needs to improve this.
Which solution did I use previously and why did I switch?
Besides ODI I have used SSIS and Power Center Informatica, both are similar and typical ETL tools, ODI up until 11g version had some differences, for example in ODI 11g we call the integration object as"Interface" wheres in Informatica it is called "Mappings" which has been implemented in the ODI 12c version.
I still could see the Oracle solution being more robust and more logical, not to mention when it comes to customization, in ODI I felt more freedom to develop and customize the built-in code and designer objects such as the models, interfaces, for example, you can run updates in the repository tables"SNP's".
Also, ODI handles high data volume more efficiently, of course, your database must be well tuned a good hardware helps a lot. The other difference is that you have more flexibility to manage the ODI metadata since everything is based on database .
How was the initial setup?
Getting the right JDK during agent config on Solaris was challenging on one occasion. It needs a Solaris expert to do this but in general when one knows the bits and pieces of the tool it should be straight forward .
What about the implementation team?
I was usually part of the implementation team to implement the product .
What's my experience with pricing, setup cost, and licensing?
oracle would better provide these kind of information .
What other advice do I have?
The setup of the environment requires experts on site, but it is very worth it since at first for customers it can be quite complex. The vendor needs to enable the customer on a frequent basis during implementation, but the results are great.