Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Format 5 Procedure Correlation Identification Algorithm

IP.com Disclosure Number: IPCOM000102578D
Original Publication Date: 1990-Dec-01
Included in the Prior Art Database: 2005-Mar-17
Document File: 2 page(s) / 55K

Publishing Venue

IBM

Related People

Baker, RW: AUTHOR

Abstract

Format 5 is a new Procedure Correlation Identification (PCID) with bits 0, 1, 14 and 15 set on to distinguish new PCIDs from previous formats. The existing method for creating this PCID causes duplicate PCID's within networks which use recommended naming conventions. These collisions occur because the first word of the PCID was stored as a 30-bit remainder (HASH1 value). The first word of the PCID was then modified by turning bits 0, 1, 14 and 15 on. If bit 14 and/or 15 are the only unique bits between two 30-bit remainders, turning those bits on results in a collision.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 53% of the total text.

Format 5 Procedure Correlation Identification Algorithm

       Format 5 is a new Procedure Correlation Identification
(PCID) with bits 0, 1, 14 and 15 set on to distinguish new PCIDs from
previous formats.  The existing method for creating this PCID causes
duplicate PCID's within networks which use recommended naming
conventions.  These collisions occur because the first word of the
PCID was stored as a 30-bit remainder (HASH1 value).  The first word
of the PCID was then modified by turning bits 0, 1, 14 and 15 on.
If bit 14 and/or 15 are the only unique bits between two 30-bit
remainders, turning those bits on results in a collision.

      It is obvious that the intent of the architecture was to
minimize the possibility of collisions in the PCID generation.  In
practice, however, since customers use naming conventions which are
sequential, analysis of one customer's input resulted in 101
sequentially numbered SSCP names only generating 50 unique PCIDs.

      The new algorithm in this proposal does not change the format
of the PCID but is less prone to the problem of the original
algorithm.  The HASH1 value, used by this algorithm, is the remainder
obtained when dividing the 64-bit results of the Exclusive OR of the
NETID (network name) and CPNAME (system services control point name)
by a 28-bit prime number (rather than the original 30-bit prime
number).  This will result in a 28-bit remainder which can be stored
in the first word of the PCID.  The fi...