And this has been the way it has worked from the first availability of a PARM disk. The addition of the PARM disk and SYSTEM CONFIG was a huge leap forward in flexibility.
Regards, Richard Schuh (also neither a historian nor a philosopher) > -----Original Message----- > From: The IBM z/VM Operating System > [mailto:ib...@listserv.uark.edu] On Behalf Of John Franciscovich > Sent: Monday, November 09, 2009 1:09 PM > To: IBMVM@LISTSERV.UARK.EDU > Subject: Re: SYSTEM CONFIG file and slots > > >The question begs to be asked, Why is it done this way? (cue > >historians and philosophers). > > > >It seems a little restrictive and not in the spirit of VM, > that being > >flexibility. > > This behavior is documented in the Usage Notes for the > CP_OWNED statement in the CP Planning and Administration manual. > > A spool file can span spool volumes, and we need a way to > identify the volume where each piece of the file is located. > > Internally, the metadata for a spool file includes an array > of pointers to individual data blocks on disk that comprise > the spool file. Each pointer, called an "ASA", includes the > slot number, which identifies the volume where the data block > resides. The array of pointers also resides on disk, and is > also pointed to by an ASA. This is used to locate and > restore the files during IPL. > > While this may seem restrictive in one sense, it also > provides flexibility, allowing: > - The device addresses to change > - The volsers to change > These make it easy to copy a system and its spool. > > John Franciscovich (not a historian or philosopher, just a > developer) z/VM Development >
