OnDemand User Group

Support Forums => z/OS Server => Topic started by: jplumey on June 21, 2019, 08:15:56 AM

Title: Getting content out of CMOD z/OS
Post by: jplumey on June 21, 2019, 08:15:56 AM

This is going to sound like a ludicrous plan, but I wanted to get the community's take on the feasibility on one of my client's (proposed) plans to move several billion documents out of CMOD z/OS.

They plan on "extracting" out all of the metadata and OAM tables from DB2 z/OS, throwing it onto a Hadoop file system, and then letting their engineers figure out how to re-assemble all of their documents (TIFF, PDF, LINE).

I understand the complexities involved, and I'm not asking for a solution, I simply want to know if anyone has ever done anything like this, with or without IBM's help. Given the engineering challenges to make this happen, is it even feasible for a non-IBM, non-OnDemand engineer to do this? On a scale of 1-10 (10 being the most optimistic), how likely is something like this to succeed?

Title: Re: Getting content out of CMOD z/OS
Post by: Nolan on June 21, 2019, 10:52:29 AM

Sounds absolutely crazy but I know there are a lot of people smarter than me.   The OAM objects are only related from tables in OnDemand, the compression used on each report can be different.   If it was possible, then IBM would have delivered a more integrated connection to Records Manager.   I don't think I have ever heard IBM or an external supplier claim that it can be done.   Any migration always involves extracting the data using OnDemand tools and then sending it to a different platform.    If the engineers can prove they can rebuild one compressed OAM report without breaking any licensing agreements with IBM.  Let us know. 

My guess -1 :)

Title: Re: Getting content out of CMOD z/OS
Post by: jplumey on June 25, 2019, 07:48:47 AM

Thanks Nolan. Yes I have never seen any vendor or partner that can do this without reverse engineering the data formats, or the code behind the OnDemand tools. This should be an interesting conversation, for sure.

Title: Re: Getting content out of CMOD z/OS
Post by: Justin Derrick on June 26, 2019, 10:22:04 AM

Yes, just confirming that this plan is ridiculous.  I have almost 25 years experience with CMOD, and have a deep understanding of how CMOD objects and compression works -- and I would never do this.

I have an extraction utility that will do this job if your client is interested.

-JD.

Title: Re: Getting content out of CMOD z/OS
Post by: jplumey on June 26, 2019, 11:56:52 AM

Justin,

Thanks for the reply. I'm sure they would be interested. I think they've abandoned this idea. Now they want to figure out how long it will take to extract 4 billion documents out so we're attempting to get some real numbers as opposed to numbers that various vendors are promising. Got any rough numbers for your utility? Does it work based on load or application group?

Title: Re: Getting content out of CMOD z/OS
Post by: Justin Derrick on June 27, 2019, 05:01:59 PM

I’ll send you a DM with more info.

-JD.