GTE Omni Si Database Technical Practices Issue 1 Manual

							TL-130500-1001
PD-200
9.0 Described in this paragraph are the basic operating
BASIC OPERATING
principles of the packet transport system used to support the
PRINCIPLES
data system and the Digital Featurephone.
Call Path Concepts
9.1 Simultaneous independent paths (logical links) can be
established between various devices within the data system.
This lets a single card simultaneously communicate with several
other cards in the system by multiplexing transmission across the
packet transport system. For a description of logical links usage,
see Table 9.1.
Table 9.1 Logical Links
SVR 5210END POINTS
RemoteRemote
Processor < -- >Processor
Remote
Universal Control
Processor 
< -- >Board
RemoteAdministrative and
Processor 
< -- >Maintenance Processor
Universal
Administrative and
Control
< -- >Maintenance Processor
BoardLOGICAL LINK USAGE
Exchange of subscriber data
Exchange of call control
information
Exchange of system control
information
Exchange of system control
information and call accounting
information
For exchange of subscriber data, the UCB- and 
ADMP-resident software controls allocation of logical links. Since each
remote processor has a unique packet line address, call routing
for data connections can be defined as the relationship of an
X.121 address to a packet line address. Two such packet line
addresses (source and destination) are required to set up a call
between remote processors. The data connection logical link
between two remote processors is the path that user data 
mini-packets will use. Each data connection logical link supports one
or more virtual circuits, or logical connections.
The data connection logical links carry the virtual circuits. These
are distinct from the call control logical links between the UCB
card and a remote processor. This call control logical link is the
only one over which call setups, clears, and other controlling
commands go. Data call control is based on a two-path
concept. All calls are internally known (after setup) by their call
identification. This call identification is dynamically mapped to a
call control block on the controlling UCB card. A call control
block contains all necessary source and destination mapping
information about a specific data call.
8/87s-133 
						

							TL-130500-1001PD-200 Data System9.2 The PD-200 Data System functions as an independent data
switch within the system.It provides interconnections among
asynchronous DTE (Data Terminal Equipment), DCE (Data
Communication Equipment), and packet mode X.25 
DTEs. The
data system also supports interfaces to X.25 
PDNs (Public Data
Networks). All data devices not collocated with the switch
connect directly to interface circuitry that allows the device to
transmit data over a single pair of wires. Asynchronous 
DTEs or
DCEs interface APMs (Asynchronous Packet Managers) or
DFP/APMs. The X.25 DTEs and X.25 PDNs interface SPMs(Synchronous Packet Managers).
Each APM and SPM that interfaces with the packet transport
system communicates over a single-pair wire using 
mini-packet protocol. The wire pair terminates on a VPLC (Voice
Packet Line Card) or VPLC2 which connects to an LPB (Local
Packet Bus) within the backplane of the switch. High-speed
hardware components within the packet transport system accept
mini-packets from their source VPLC cards and switch them to
their destination without additional processing. The end points
run mini-packet protocol, not the switching hardware within the
packet transport system.
Data call-processing software runs on the UCB card configured
with data call-processing software. This card is responsible for
all of the non-real-time critical functions, such as command
interpretation, call setup, and call takedown. Except for
command information, user data does not pass through the 
UC6card. It goes directly between the end points (APM-to-APM,
APM-to-SPM, or SPM-to-SPM).
Administrative software resides on the ADMP cards. The ADMP
has high bandwidth access to the system disk where event and
data accounting information are kept. The ADMP contains
software which provides a simulated X.25 end point. This allows
access to the data administrative functions, such as Recent
Change and maintenance. via virtual circuits. Any device
connected to the PD-200 Data System can be used as a data
maintenance console as long as enough security information is
given.
Software oaded devices that have a LED (Light Emitting Diode)
(UCB cards, ADMP cards, APM, and SPM) use the LED to signal
hardware malfunction. A card that passes ROM self-tests
flashes this LED at approximately 60 IPM. A card that fails 
self-tests flashes this LED at approximately 120 IPM. When these
devices are loaded with the appropriate software, the LED is lit
steadily.
The VPLC and VPLC2, which are not software-loaded cards,
also have an LED. This LED is off until the card is initialized.
Then the LED lights steadily as long as the card is operational.
S-l 34
8187SVR 5210 
						

							TL-130500-1001Data Interfaces9.3 The PD-200 Data System supports the following data
devices:
l ASCII asynchronous data terminals at speeds up to 19.2 Kbps
l Host computers emulating asynchronous data terminals at
speeds up to 19.2 Kbps
l A limited set of full-duplex modems (private line, 
auto-answer, and auto-dialer) providing an asynchronous, serial,
ASCII, single-speed interface
l X.25 
DTEs at speeds up to 64 Kbps with HDLC framing
l X.25 
PDNs at speeds up to 64 Kbps with HDLC framing
Signaling9.4 For asynchronous devices, user commands (call setup, call
disconnect, etc.) are entered via the device keyboard by using a
command language based on and similar to 
CCITTRecommendation X.28. The X.25 devices use X.25 control
packets to provide signaling.
Asynchronous Call9.5 This paragraph presents a summary of a call setup as seen
Originationat an asynchronous terminal. It details the connection being
established between a terminal and an X.25 host, both directly
connected to the PABX:
l DTR (Data Terminal Ready) is raised by the terminal
connected to the originating APM.
l A    is entered
on the terminal keyboard. This informs the PD-200 Data
System of the correct operating speed and parity for the
terminal and that the terminal is to run full duplex. For a 
half-duplex application,  < hyphen > -=I carriage
return> should be used (this only applies to ports marked
autobaud in the port table).
l The herald is received, and on 
autobaud devices the user is
prompted for the “Terminal Type“.
l After receiving the command prompt, the user enters the
connect command on the keyboard, followed by the X.121
address to which he wants to connect. Mnemonic addressing
can also be used.
l An X.25 incoming call packet is sent to the X.25 host or
asynchronous port, informing it of the incoming call. The host
can accept or reject the call.
e If the connection cannot be established, or if the X.25 end point
rejects the call, a message indicating the reason (error, busy,
etc.) is displayed on the originating terminal. (These
messages are suppressed if X.3 parameter 6 is set to a zero.)
SW 52108187S-l 35 
						

							TL-130500-10010 A connected message is displayed on the originating terminal
when the connection has been made. (This message is
suppressed if X.3 parameter 6 is set to a zero.)
l The call indicator on the originating APM will be lit.
Call Disconnect9.6 Either party can disconnect a data call. Asynchronous
devices disconnect a data call by dropping 
DTRDSR, or by
escaping to the network command mode and entering the
disconnect command on the keyboard. (This option is not
allowed if X.3 parameter 1 is set to zero.) The X.25 devices
disconnect a data call either by issuing a call clear packet, or by
dropping data set signals or restarting the X.25 line. This will
cause all calls on this line to be disconnected and is considered
an abnormal condition. Either way, the call indicator on an APM
will be extinguished when a connection is terminated.
Protocol Conversion9.7 The data system provides an asynchronous to X.25 protocol
conversion for each asynchronous port. Recommendation X.25
allows a data device to transmit data as a series of messages or
packets, and to simultaneously support multiple virtual circuits to
a number of different devices. The following description
assumes that an X.25 host connected to an SPM is
communicating with a terminal connected to an APM.
When the terminal user enters data from the terminal, data is
transmitted to an APM asynchronously, byte by byte, where it is
converted into X.25 data packets. This is done by collecting the
asynchronous characters to form X.25 data packets to send to
the VPLC. Special software in the APM, referred to as PAD
(Packet Assembler/Disassembler) software, converts the
asynchronous data to X.25.
Once the X.25 packet (X.25 level 3) is created at the APM, it is
broken down by the APM into a series of mini-packet protocol
packets. These are transmitted, via the wire pair, to the VPLC.
The VPLC places the mini-packets onto the LPB; then they are
switched by the packet transport system hardware to their
destination. Mini-packet protocol is used as the link level (level
2) protocol to ensure that the packet is delivered without errors or
duplication to its destination. At the SPM, the mini-packet
protocol packets are collected to reconstruct the original X.25
data packet that was formed
a: the APM. Next the SPM envelops
the X.25 level 3 packet information in level 2 frames, using HDLC
framing. Since the X.25 host computer directly accepts X.25
data packets, the PAD function is not required here.
Asynchronous interfaces are supported for speeds up to 19.2
Kbps with a standard EIA RS-232-C connector, whereas
synchronous interfaces are supported at speeds up to 64 kbps
operating with X.25 protocol and connected via EIA RS-232-C
(for up to 19.2 Kbps links) or V.35 (for higher-speed links).
S-l 36
8187SVR 5210 
						

							TL-130500-1001Data Device9.8 This paragraph describes addressing for data devices.
Addressing
Numbering Plan9.8.1 To plan a data numbering scheme, the system users must
first be grouped in the following categories.
l A rotary hunt group (89 users maximum) 
- asynchronous
l 89 individual users (maximum) of the non-rotary hunt group 
-asynchronous
l 1 pair of primary/secondary X.25 lines 
- synchronous
l 1 pair of load-sharing X.25 lines 
- synchronous
l 1 single X.25 line 
- synchronous
. 1 ADMP
Each group is identified by a unique address (server number).
The addresses are used to complete the data base sheets for
Record Code RT.
The addressing used by the data system is the X.121 addressing
scheme.
CCITT X.1 21 Addressing9.8.2 The CCITT has defined an international numbering plan for
public data networks known as Recommendation X.121 (Figure
9.1). Under this plan, all network addresses are composed of a
14-digit number. The first four-digit block is called a DNIC(Data Network Identification Code). One or more 
DNIC codes
are assigned to each country by the CCITT. The remaining 10
digits are called a Network Terminal Number; this is subdivided
into an 8-digit server number and a 2-digit sub-port number.
The synchronous (X.25 
DTEs) device is normally assigned a
12-digit X.1 21 address, while an asynchronous device is
assigned a 
14-digit X.1 21 address. The format of the complete
X.121 address is 
DNIC SSSSSSSS PP (Figure 9.1).
SVR 5210
~~:D;N~;l 1:’ X,;;; ‘Li;;Ti,
Figure 9.1International Numbering Plan for Public
Data Network Addressing
8/87S-l 37 
						

							TL-130500-1001Data Network9.8.3 In the United States, the DNlCs are assigned to a network
Identification CodesNI = Network Identifier digit by the Federal Communications
Commission. The GTE-Telenet Public Network has been
assigned 
DNIC 3110 (see Figure 9.2). Some currently assigned
DNIC codes are as follows:
DNICNETWORKCOUNTRY
3110Telenet
3106Tymnet
2341IPSS
2080NTI
2081TranspacUSA
USA
England
France
France
In the PD-200 Data System, it is suggested that 
DNIC 3110 be
used in the X.121 address. For ports which connect to other
public data networks, the user must obtain the appropriate X.121
address from the public data network vendor.
S-l 38
j3110~xxxxxxxx;xx i
IIII
II/  DDC
NI +--
DNIC iSERVER
-4 SUB
INUMBER
i
i PORT ;
I
II
II
Figure 9.2DNIC for TelenetNOTE: The data system cannot be configured without an X.121
address. This is supplied by a PDN (Public Data Network).
Telenet is recommended for usage by the PD-200 Data System.
8187SVR 5210 
						

							TL-130500-1001Server Number9.8.4 The server number (Figure 9.3) is an eight-digit number
which identifies an X.25 host, terminal, or network. A server
number may also identify a collection of asynchronous
lines/devices that are grouped into a set. The lines/devices in
this set can either be accessed individually or as a group. A
main pilot number provides access to a rotary hunt group. One
set may contain a maximum of 89 terminals in the PD-200 Data
System.
i3110j20200100; xx jIIIII1
i DDCNI i+-SERVER-i SUB :I
NUMBER
I
DNIC ;j PORTj
IIIIII
IIII
II
I
Figure 9.3Example of DNIC Server Number
NOTE: This example represents a user served by Telenet in
Washington, D.C., which has only one area code (202). Telenet
uses the area code as the first three digits of the server number
Sub-port Number9.8.5 This is a two-digit number that is used to identify a
specific asynchronous terminal with the set (Figure 9.4). The
sub-port is not used with X.25 synchronous lines/devices. The
sub-port number may range in value from 01 to 89 for each
server number.
/1
i3110j20200100;01 i
IIII
IIII/  DDCNI 
i+-SERVER-I SUB iI
INUMBER
I
DNIC ;j PORTj
IIIIIItI
II
IFigure 9.4 
’ Example of DNIC Server Number and
Sub-PortUsers of asynchronous terminals have the option of specifying
complete or partial X.1 21 addresses when setting up a virtual
circuit. If the address specified begins with a zero, then a
complete address will be assumed. The address used will be
the same as that which the user entered, with the leading zero
removed. If the address does not begin with a zero, a partial
address will be assumed. In this case, the called X.121 address
is calculated by placing the user-entered number into the server
field right justified and padded to the left with zeros. The DNIC
field will be the same as the 
DNIC associated with the originating
SVR 5210
8187s-1 39 
						

							TL-130500-1001
terminal. If a port number has been specified, then it will be
used; otherwise, the port number will be set to zero.
Sub-addresses 90-99 are reserved for internal administrative
functions. The ADMP uses an address which must have a
unique server number and its support number must be 00. No
other restrictions are imposed on the number of digits or the
values of the digits comprising the address.Routing9.9 All routing of X.25 data packets is based upon the virtual
circuit concept. When a data call request packet is received (or
created) by the data switching software, a path will be established
(if possible) between the data port from which the call request
was received and another port whose identity is determined from
a routing table. All subsequent data packets associated with the
virtual circuit will follow the same path between the two ports.
This path will be maintained by the data switching software until
the virtual call is cleared.
Routing decisions are based upon a table of partial X.121
addresses. This table holds up to 127 entries. Each entry
contains a partial X.121 address made up of actual 
binary-coded digits, wild-card digits, and an indication of how many
significant digits of that address are to be used in routing
decisions. Thus, many destination addresses may be
represented by a single routing table entry. This entry can
contain one or more wild cards, or a single specific address may
be represented in the table. A single address needs all 14 digits
specified where a wild card does not. The routing tables also
contain a primary port identification, primary port controlling DCP,
an optional secondary port identification, a secondary port
controlling DCP, and an indicator of how the secondary port is to
be used in conjunction with the primary port (load shared based
on number of virtual circuits, or when the primary port is busy or
out of service).
NOTE: The ADMP must have its own routing table entry.
Rotary Hunt Group9.9.1 A group of lines/devices that share the same server
number are called a rotary hunt group (Figure 9.5). The following
describes the operation of a rotary hunt group: When a call is
placed to the pilot number of a rotary hunt group, the system will
automatically scan through the defined set of ports in the group
looking for the first available non-busy line. Upon finding one,
the system will make the connection to that specific port and
inform the remote user of its choice.
Since only 1 server number is 
!ieeded for each group of 89
users, this can be used when interfacing multiple lines into an
asynchronous host, group of printers, or group of modems. The
PD-200 Data System automatically assigns ports upon demand.
Also, even though the ports have been defined as members of a
rotary group, a specific port can still be called by using the port
number. If the specific port is busy, the system informs the user.
The system does not select an alternate address (Figure 9.6).
8187SVR 5210 
						

							TJ -130500-l 001
i31 10i2020010o/01;
13110120200100;02;
13110;20200100;03;II1IIj  DDC
NI i-SERVER-i SUB iI
DNIC !NUMBERI PORT;I
I
Figure 9.5Rotary Hunt Group ExampleNOTE: Example of a rotary hunt group with 3 members (89
maximum).
/  
i3 1 1 0 i2 02 0 0 1OO~Ol~lI  
:IIIIi I/  ;  
D”DcN,CN’ j-SERVER-i SUB 1
INUMBER
II; PORT ]IIIIII1III
II
Figure 9.6EAample of Rotary Hunt Group Numbering
Plan
iNOTE: This example represents a rotary hunt group access to a
group of printers. Any 
DFP/APM can access the group by dialing
the X.121 address (DNIC and server number, 311020200100)
however, if a certain printer is needed, the sub-port would have
to be dialed as well (example 31102020010001).
Load Sharing9.9.2 Load sharing on the data system when used in conjunction
with the 
SPMs and the NIC PCBs provides equal distribution of
calls over X.25 paths to a set of destinations (Figure 9.7 and 9.8).
rii
13 1 1 0 i2 0 2 0 0 1 0 0;III
I3
III/  DDC
NI lo-SERVER-i SUB iI
I
DNIC /NUMBER
Ij PORT II
II
I
I
II
I
IIII
I
3gure 9.7Example of Load Sharing Numbering PlanNOTE: In a load sharing application, this X.121 address would
be shared by two synchronous users.
SVR 5210
8187S-l 41 
						

							 rTL-130500-1001
HOST COMPUTERrlyult: a0Luau 31 lcll II ly cAclIIIfJlt:NOTE: The 
APMs are able to access the host computer via the
SPMs. This application provides an automatic back-up
connection should one of the links fail. It is recommended to put
load sharing 
SPMs on different VPLCs.
8187SVR 5210