Browse Prior Art Database

Useful Manipulations of SNA LU Names

IP.com Disclosure Number: IPCOM000120823D
Original Publication Date: 1991-Jun-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 4 page(s) / 165K

Publishing Venue

IBM

Related People

Peters, ML: AUTHOR

Abstract

Disclosed are a method of assigning SNA network identifiers that permits the detection of unregistered NETIDs, and a method of algorithmically manipulating the characters of an SNA LU name to generate a unique binary string with certain useful properties. Method of Increasing the Probability that a NETID is Unique

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 36% of the total text.

Useful Manipulations of SNA LU Names

      Disclosed are a method of assigning SNA network
identifiers that permits the detection of unregistered NETIDs, and a
method of algorithmically manipulating the characters of an SNA LU
name to generate a unique binary string with certain useful
properties.
Method of Increasing the Probability that a NETID is Unique

      SNA networking protocol requires each network service access
port (logical unit or LU) within a network to have a unique name.
Generally, a customer enterprise assumes responsibility for assigning
unique LU names within that enterprise.  If two independent
enterprises that have assigned LU names without coordination later
wish to merge their networks together or exchange data
electronically, LU name conflicts may occur.  These conflicts may
cause session failure, sessions with the wrong partner, inability to
properly manage the interconnected networks, and other difficulties.

      The first step towards solving this problem was to prefix each
LU name with a network identifier (NETID) that is the same throughout
a given network.  However, historically SNA had no administrative
controls to guarantee the selection of unique NETIDs by customer
enterprises. Therefore, since two enterprises could independently
select the same NETID, that approach did not fully solve the original
problem.

      The second step was establishment of a central NETID naming
authority (the Registry).  However, the Registry had no way to ensure
that the NETIDs it assigned or registered did not duplicate NETIDs
chosen before the Registry's existence.  Assigning registered NETIDs
using the algorithm disclosed here enables registered NETIDs to be
distinguished from unregistered NETIDs with a high degree of
probability.
Algorithm for Assigning Registered NETIDs

      This algorithm encodes a checksum in the NETID string in a
manner consistent with the LU-name format.  A valid LU name consists
of an optional 1- to 8-character string called a network identifier,
concatenated with a period if a network identifier is present,
concatenated with a 1- to 8-character string called an unqualified LU
name.

      The characters in the NETID and unqualified LU name are drawn
from the upper-case English alphabet letters A through Z and the
numerals 0 through 9, each character encoded as an 8-bit binary value
using the EBCDIC character set, collectively referred to as "IBM
symbol set 1134 encoded using code page 500". The period is also
encoded using EBCDIC.  Character set 1134 is available on all SNA
products, regardless of national language.  The first character of
the NETID and unqualified LU name may not be a digit.

      The following procedure creates a new NETID that can be
distinguished from unregistered NETIDs with a high degree of
probability:
   1. Assign the first six characters of the new NETID from character
set 1134, taking care that the six-character string does not
dupl...