During the “Ask the Experts” session at the Serena Mainframe Virtual User Group meeting earlier this month, a few questions about the newest version of ChangeMan ZMF went unanswered due to the lack of time. Mark Levy, Serena Software Mainframe product manager, and I went through them and provide the answers here.
Q: For ChangeMan ZMF v6, the product moved from DB2 to linear VSAM. What are the benefits and risks that you have seen?
A: In ChangeMan ZMF v6, the Impact Analysis function was moved from being based on DB2 to a z/OS data space, backed by linear VSAM. This was done for several reasons: performance, ability to better manage Java components, and the ability to expose Impact Analysis XML services, which could not be done under the DB2 implementation. In doing so, we have provided a utility which will allow users to offload this data back into a DB2 table in case they have SQL that they would like to continue to run against this data. This move eliminates the requirement for DB2 for dynamic IA updates. Now everyone has the benefit of real-time updates to their Impact Analysis data, even if they do not have DB2.
Q: Is there a plan to manage data through ChangeMan ZMF also?
A: ChangeMan ZMF has the ability to manage data components throughout the lifecycle, assuming that these are components that do not get built, compiled, have external dependencies that would require audit or impact analysis. However, ZMF is primarily used for managing objects that relate to code (src, load, JCL, DBRM etc.)
Q: Is the Eclipse plug-in part of ChangeMan ZMF?
A: The Eclipse plug-in is an option that is part of the ChangeMan ZMF Client Pack, which also includes an Eclipse Plugin for Rational Developer and ChangeMan ZDD.
Q: What exactly is in ChangeMan HANA? What are the exciting features?
A: HANA is the code name for the next major release of ChangeMan ZMF. The proposed features are still in the early stages of research.
Q: Are you going to integrate PeopleSoft systems?
A: There are no current plans for a PeopleSoft implementation at this time. With ChangeMan ZMF’s web and XML services, customers or Serena Professional Services have the ability to do these types of integrations.
Q: What platforms will ChangeMan ZMF support for managing and deploying WebSphere?
A: ChangeMan ZMF will support the deployment of WebSphere applications and components on z/OS.
If you have additional questions, feel free to post a comment below or visit the ChangeMan ZMF page.
|Al Slovacek joined Serena in 1995 and has functioned in a variety of technical and management roles including ChangeMan ZMF development. He currently serves as Vice President of Mainframe Products. Prior to Serena, Al was a developer of a variety of mainframe products ranging from z/OS performance management, storage management, VSAM optimization and compression. Prior to this he was a Systems Programmer specializing in mainframe performance analysis, tuning, installation and configuration.|
Just following-up on the 1st question asked in relation to the DB2/VSAM move. It’s not entirely clear to me what extent this move helped with scalability of the solution .Can you expand on that a bit ? Also was Impact Analysis the only functional area concerned by this move ?Thanks in advance.
In and of itself, the move toward a linear VSAM structure didn’t effect scalability one way or the other. But this change was interwoven with changes to the key structure of the history file (also shifted to linear VSAM) which was necessary to allow for unlimited component history (recall that prior to v6, only 24 levels of component history were kept). The unlimited amounts and the addition of user meta-data fields required us to rework the record keys in the component history that had an impact on the way Impact Analysis data is managed.
So in addition to the other benefits mentioned above (performance, ability to expose out IA services, reduced dependency on DB2) the work done in concert with Component History enabled unlimited levels and additional user meta data.
Another (unanswered) question from that same VUG: How to find the best values for the ASID-parm (in the STC) and what customizations can you do to prevent s069 abends. Typical scenario: final approve, using XML, some hundreds of pkgs; This might cause some hundreds of Xnode DSNs to be recreated; Sooner or later (depending on your ASID-values) the S069 abends will appear (and because of their nature they are not easy to discover …).
Best values, like best practices evolve over time, and in this case depend on the environment. In the ASID parm, the first parameter is the maximum number of ASID’s that we will create at one time, the second is the number of minutes until it basically gives up and fails. The S069 abend would indicate that a ASCRE to create an address space failed. There are various reasons that this could fail, such as no available ASID’s left or storage issues like SQA shortage, etc. The maximum number of ASID’s on the system is determined by PARMLIB settings in IEASYSxx. So with all of these variables in place, the best guidance is to tune and experiment in your own environment. What may work well for you may not in another environment, which is why we provided the levers to control the amount of address spaces, and wait times.
when is the next version of Changeman ZMF 8.0? coming ?
We are targeting end of 2012, but is still in the planning stage, and will likely be the 7.2 release.