Browse Prior Art Database

Timer-Controlled, Equitable Port Sharing for Short Hold Mode Invention

IP.com Disclosure Number: IPCOM000119593D
Original Publication Date: 1991-Feb-01
Included in the Prior Art Database: 2005-Apr-01
Document File: 2 page(s) / 101K

Publishing Venue

IBM

Related People

Bentley, AM: AUTHOR [+2]

Abstract

A problem exists on X.21 Short Hold Mode (SHM) communication lines when continuously busy stations do not disconnect, thereby monopolizing the port(s) and preventing both outgoing calls and incoming calls from completing. Described is a mechanism which provides fairness to all remote users of an X.21 SHM port group by using a "maximum connect timer", a "fair answer period", and a "delay for answer" timer.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Timer-Controlled, Equitable Port Sharing for Short Hold Mode Invention

      A problem exists on X.21 Short Hold Mode (SHM)
communication lines when continuously busy stations do not
disconnect, thereby monopolizing the port(s) and preventing both
outgoing calls and incoming calls from completing. Described is a
mechanism which provides fairness to all remote users of an X.21 SHM
port group by using a "maximum connect timer", a "fair answer
period", and a "delay for answer" timer.

      The X.21 Short Hold Mode architecture provides that a physical
connection be dropped when both a disconnect timer has expired and a
preselected number of SDLC Receive Ready commands has been exchanged.
The architecture also states that only interactive applications are
suitable for SHM, and only if they are characterized by infrequent
transfer of data.  Batch or continuous data transfer usage, even if
for short periods, is unsuited for SHM.  This creates the following
problems:
-    A busy station can continually transmit or receive data and fail
to meet the conditions for a SHM disconnect: the required number of
Receive Ready exchanges does not occur.  Such a controller will lock
out all other users of the port.  This can prevent both incoming and
outgoing calls for other stations from completing.
-    If the system makes continuous outgoing calls to multiple remote
controllers and in the process ties up the port or port group for an
extended time, the SHM architecture provides no period to temporarily
halt this outgoing traffic to allow incoming calls to be received.

      HIGH-LEVEL DESCRIPTION OF INVENTION:
-    Primary SDLC uses a new timer, the 'maximum connect timer', to
control how long to keep active a connection to a busy controller.
After this time, if there are other potential users of the port, the
connection is cleared, allowing potential incoming calls to complete.
Cleared outgoing calls are reestablished later, after the delay for
answer period.
-    When multiple, continuous outgoing calls keep the port or port
group busy for the 'fair answer time', but normal SHM disconnections
do occur between calls, then, after the next disconnection, further
calls out are held, and one port is kept free for the delay for
answer period to allow incoming calls to complete. With the 'connect
when free' network option, incoming calls will complete immediately
and in the order placed by the remote callers.

      1.  MAXIMUM CONNECT TIMER.  This timer controls how long the
primary SDLC station will allow a connection to any one controller.
This timer runs in addition to the SHM disconnect timer.  When
prima...