Last month, along with my colleague Tim Wilson, I met with a couple of customers in China. I wrote a previous blog post based on this business trip that recalled the importance of using the right terminology for the tool adoption.
Customer under the impression that distributed and mainframe teams required separate CLM / RTC servers (i.e. JTS)
During a discussion with another customer, it came up that the person in charge of SCM administration (for distributed) was under the impression that distributed and mainframe teams required separate CLM / RTC servers (i.e. JTS).
Since the tool itself does not impose such separation, I made the point this would rather be an organizational decision rather than a technical one. I felt like this was pretty new to the customer.
To clarify things, I initiated a blackboard session to highlight the different options wrt. the deployment topologies, their advantages as well as their implications for the future. Some core messages were:
- The RTC server(s) can run on a various environments (including Windows, Linux, AIX, Unix, zLinux, z/OS, IBMi, etc.).
- The fact that RTC EE is used for developing for z/OS does NOT imply that the RTC server shall run on z/OS. This was already stressed in other resources from IBM colleagues,
- You have the options of using a single JTS or separate JTSes. In both cases, you can associate multiple CCM instances to your JTS(es). Choices should be guided either based on performance considerations or, again, on organizational considerations.
- When considering options for using multiple JTSes:
- It’s very important to note that, once separated, they could NOT be merged later.
- Two separate servers require doubled effort for administration, upgrade, etc.
- The reporting tools should be adapted accordingly (e.g. Insight/RPE instead of RRDI/RRDG)
At the end, the customer could have a better understanding of the CLM/RTC EE deployment options for his organization.
For the little story, in this customer situation, SCM administrators for both distributed and mainframe developments were supposed to meet with each others after our visit to discuss the best deployment option. Yet another good effect of ‘Enterprise Modernization’ brought by the adoption of the RTC EE tooling !
Note: in this article, I insisted on some aspects of RTC/CLM deployment. For a complete picture, the Deployment wiki is definitely a good read.