Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!MBUNIX.MITRE.ORG!jcmorris From: jcmorris@MBUNIX.MITRE.ORG (Morris) Newsgroups: comp.lang.asm370 Subject: Re: CTCAs Message-ID: <8912032350.AA20677@brazos.rice.edu> Date: 1 Dec 89 22:00:53 GMT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: IBM 370 Assembly Programming Discussion List Distribution: inet Organization: The Internet Lines: 32 In a recent article CMH117@PSUVM.BITNET (Charles Hannum) writes: > > Exactly what *are* Channel-To-Channel Adapters (CTCAs), and what do > they (and CP COUPLE) do? > The CTCA is usually used to provide a dedicated, high-speed (typically 3 megabyte/sec) data path between mainframes. The CTCA itself (which can be either a feature on the processor or a stand-alone device) appears to be an I/O device to each processor attached to it. At the hardware level, that's all it is...in effect, a very expensive (but very fast) null modem cable. In VM you can create virtual CTCA ports with the CP DEFINE CTCA [as] cuu command; you then use the CP COUPLE command to say what is attached to the other end of the virtual cable by giving the userid and virtual address of the distant end. For example, if A and B want to communicate over a virtual CTCA, then: A issues the command: CP DEFINE CTCA AS 333 B issues the commands: CP DEFINE CTCA AS 444 CP COUPLE 444 TO A 333 and now A and B have a CTCA between them. What you do with the link is up to your programming system. It could be used for data transfer, for supervisory data transfer (e.g., passing ENQ/DEQ commands between processors which share DASD), or whatever you want. Hope this helps... Joe Morris