Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!YKTVMH3.BITNET!PERSHNG
From: PERSHNG@YKTVMH3.BITNET ("John A. Pershing Jr.")
Newsgroups: comp.protocols.ibm
Subject: LU type 0
Message-ID: <9001171513.AA21632@lilac.berkeley.edu>
Date: 17 Jan 90 14:51:57 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: "John A. Pershing Jr." <PERSHNG%YKTVMH3.bitnet@lilac.berkeley.edu>
Organization: The Internet
Lines: 17

From an SNA perspective, there isn't an "LU Type 0" protocol.  Or, to put
it another way, an LU-0 session doesn't have any architected rules: it
is more-or-less outside of the realm of SNA, and any rules are strictly
by agreement between the two communicating LUs.  There are potentially
an infinite number of "LU-0" protocols...

There *are* a couple of commonly-used protocols that fall in the LU-0
category:  BiSync 3270s; and SLU-P.  I don't have a reference handy for
the BiSync 3270 protocols; I believe that SLU-P is described in the IMS
communications manual, probably in the ACP/TPF manuals, and probably also
in the manuals that describe IBM's financial clusters (the 3600 and 4700
series machines).  I don't think that the IBM manuals actually use the
name "SLU-P" (which means "Secondary LU -- Programmable"), but this is
the term that is used in the financial community.

      John Pershing
      IBM Research, Yorktown Heights