Proposal: Mainframes

Mainframes do not exist "alone" (as opposed to desktop systems, embedded solutions and other all-in-one boxes), but rely on a solid infrastructure available on-site, such as networking, SAN fabric and perhaps even dedicated power lines.

Does this proposal include questions about planning, maintenance and resolving connectivity issues between mainframe internal cards and external devices, perhaps with specific model numbers?

I was actually working on a proposal like this and found that @rbanffy had started this one. As an IBMer I'm very frustrated that we have a lot of academic information (manuals galore) but very limited applied knowledge available in the eco-system. My desire for this proposal is that it allows for all manner of questions from scaling, capacity planning, sysplex administration, installation howtos, etc. so that people that are working on the mainframe (in my parlance at least that is generally z/OS but should include VSE, TPF, z/VM, ...) can get and give practical howtos and less of a RTFM which on z/OS at least could take years.

Having been a practitioner for several years before joining IBM (pre-dating StackExchange Community) I felt that this kind of practical approach was something that would benfit those in the field, and, perhaps more importantly in some respects, demystify the mainframe so others can take advantage of all the assets that reside on these mysterious boxes :)

I'd say yes. Mainframes have unique characteristics and tooling that set them apart from the "usual" stuff (x86 clusters, real or virtual, and bigger x86 or SPARC and soon ARM boxes). I started this proposal because I myself want to learn more about it and there are very few ways to get on-board.

We certainly need more questions that are not covered by on StackOverflow. System administration is just as relevant to mainframes as programming them and would attract a bigger audience.

Given that area focus is on existing marketed systems, questions regarding specific hardware models are within the scope.

The [mainframe] tag on ServerFault is moribund, StackOverflow is for coding, and ibm-main is for discussion and solutions (where SE sites are quite emphatically the latter without the former). The way the voting on sample questions is going, it seems the site resulting from this proposal would be "ServerFault for Mainframes (HLASM okay too)."

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .