What is your favorite Feature of ODI 12c? (Compared to 10g or 11g)
If you have used ODI 10g or 11g or both you know that there is the big difference between those versions and ODI 12c. What is your favorite new feature of ODI 12c? Is it Reusable Mappings? Maybe its just the fact that its flow based?
There are a lot of great new features in ODI 12c, but one of my favorites is the deployment specifications on the physical layout of the mapping. These deployment specifications allow you to use the same logical flow for your mapping with multiple different physical deployments. For example, you may need to perform an initial load of data from source to target once. Then, every subsequent run of the mapping would be an incremental update, only inserting/updating rows that are new or changed. In this case, there are different Knowledge Modules that must be used, one for the full load and a different one to perform the incremental update logic. With deployment specifications, we can use the exact same logic in the mapping and create 2 different physical implementations without duplicating the mapping. In past versions of ODI this would have been two mappings, both with the same logic that would need to be maintained twice, with lots of room for error!
I think the main difference between the older versions and the 12c is the GUI.
Maybe for a new user the 12c is more friendly, but for me that works with odi since 10g, everything in the 12c takes 2 to 3 times more then in the older version.
The 12c has a lot of features as default that the older version needs to customize to have it, but in the end you can do everything in all versions, the only thing that changes is the way you do it.
For example, in the older version if you want to load data in parallel using interfaces, you need to create a java variable to get the session number and concat it with the I$_ prefix in the topology. In 12c this is enable by default.
This is just a example, but everything will be like this.
Then I think the best thing you can do is install the 12c version and give it a try.
in the end everything will be resume to the GUI.
hope this can help you. If you want more info about odi, please take a look in my blog.
Hi,
I am using ODI from long. I used 10g, 11g & 12c. When I started my work on ODI lower versions i.e. 10g & 11g, I feel helpless sometime. But after releasing of 12c. It proves me wrong. There are lots of features which you may get any other ETL tool but to implement the solution,could be difficult. I like Declarative flow based design, Multiple target, extensive Big data support, Multithreading target load, and most important OWB integration.
Data Quality Software Development Manager at Yapı Kredi Bank
Real User
2016-05-18T14:02:22Z
May 18, 2016
Oracle Data Integrator is a tool that can talk or learn how to talk with any database or operating system in it's own language... This is the power of ODI.!
While ODI 10G and 11G had a declarative based design, the main difference between ODI 12c and prior versions is being "declarative flow based design". It is both flow based design with a declarative approach, so that both traditional ELT developers and E-LT developers can build easily. With holding the status of "easy learning", I believe Oracle Data Integrator will still be on top.
ETL Developer at a comms service provider with 10,001+ employees
Real User
2016-05-16T15:11:07Z
May 16, 2016
Hi Brian,
comparing 11 and 12c version of ODI I can tell that Reusable Mappings are really something it help to me reduce some codding in well organized way.
I try also multiple targets as I need similar functionality but finally not use it as I need target to be file and DB table and seems it is not possible multiple targets to use different technology. Probably this will be beneficial if it is developed in next version.
Fact that mapping are flow based make the transformation more compact and easy for reading, which is important for big transformation flows.
Solution Manager at a tech services company with 501-1,000 employees
Consultant
2016-05-16T21:42:34Z
May 16, 2016
I would say the new flow based approach is the main change. Easier to read the data flow to understand the logic of transformation instead of creating/opening many interfaces and try to follow all.
Reusable mapping is also very useful just like multiple target load. Despite i did not use yet, big data support can be very handy for the teams working on that.
Business Process and Strategy Specialist Advisor at NTTData
Consultant
Top 5
2016-05-16T20:14:20Z
May 16, 2016
I'm a long time user of ODI and specifically ODI12c. If that makes me a addict; then sign me up for the meeting. I love the reusable mapping but the best part of this 12c is that its now Flow based. I would also have to agree with Michael Rainey on all the points that he brought to light in his response to my question. I mean it just makes other integration tools look like child's play.
Oracle Data Integrator (ODI) is a data integration software solution that provides a unified infrastructure to streamline data and application integration projects. It uses a powerful design approach to data integration, which separates the declarative rules from the implementation details. The solution is based on a unique ELT (Extract Load Transform) architecture, eliminating the need for a standalone ETL server and proprietary engine.
Oracle Data Integrator Features
ODI has many...
There are a lot of great new features in ODI 12c, but one of my favorites is the deployment specifications on the physical layout of the mapping. These deployment specifications allow you to use the same logical flow for your mapping with multiple different physical deployments. For example, you may need to perform an initial load of data from source to target once. Then, every subsequent run of the mapping would be an incremental update, only inserting/updating rows that are new or changed. In this case, there are different Knowledge Modules that must be used, one for the full load and a different one to perform the incremental update logic. With deployment specifications, we can use the exact same logic in the mapping and create 2 different physical implementations without duplicating the mapping. In past versions of ODI this would have been two mappings, both with the same logic that would need to be maintained twice, with lots of room for error!
Hi how are you?
I think the main difference between the older versions and the 12c is the GUI.
Maybe for a new user the 12c is more friendly, but for me that works with odi since 10g, everything in the 12c takes 2 to 3 times more then in the older version.
The 12c has a lot of features as default that the older version needs to customize to have it, but in the end you can do everything in all versions, the only thing that changes is the way you do it.
For example, in the older version if you want to load data in parallel using interfaces, you need to create a java variable to get the session number and concat it with the I$_ prefix in the topology. In 12c this is enable by default.
This is just a example, but everything will be like this.
Then I think the best thing you can do is install the 12c version and give it a try.
in the end everything will be resume to the GUI.
hope this can help you. If you want more info about odi, please take a look in my blog.
Thank you
There are a lot of nice new features that I mentioned in two blog posts :
www.rittmanmead.com
www.rittmanmead.com
But I think my favorite is the auto-completion when writing expressions/code. So much time spared with it!
Hi,
I am using ODI from long. I used 10g, 11g & 12c. When I started my work on ODI lower versions i.e. 10g & 11g, I feel helpless sometime. But after releasing of 12c. It proves me wrong. There are lots of features which you may get any other ETL tool but to implement the solution,could be difficult. I like Declarative flow based design, Multiple target, extensive Big data support, Multithreading target load, and most important OWB integration.
Regards,
Himanshu
Oracle Data Integrator is a tool that can talk or learn how to talk with any database or operating system in it's own language... This is the power of ODI.!
While ODI 10G and 11G had a declarative based design, the main difference between ODI 12c and prior versions is being "declarative flow based design". It is both flow based design with a declarative approach, so that both traditional ELT developers and E-LT developers can build easily. With holding the status of "easy learning", I believe Oracle Data Integrator will still be on top.
Hi Brian,
comparing 11 and 12c version of ODI I can tell that Reusable Mappings are really something it help to me reduce some codding in well organized way.
I try also multiple targets as I need similar functionality but finally not use it as I need target to be file and DB table and seems it is not possible multiple targets to use different technology. Probably this will be beneficial if it is developed in next version.
Fact that mapping are flow based make the transformation more compact and easy for reading, which is important for big transformation flows.
I would say the new flow based approach is the main change. Easier to read the data flow to understand the logic of transformation instead of creating/opening many interfaces and try to follow all.
Reusable mapping is also very useful just like multiple target load. Despite i did not use yet, big data support can be very handy for the teams working on that.
I'm a long time user of ODI and specifically ODI12c. If that makes me a addict; then sign me up for the meeting. I love the reusable mapping but the best part of this 12c is that its now Flow based. I would also have to agree with Michael Rainey on all the points that he brought to light in his response to my question. I mean it just makes other integration tools look like child's play.