Intel SIU520 SS7 User Manual
Intel in
Communications
Building Fault-tolerant SS7
Systems Using the
Intel
®
NetStructure™
SIU520 SS7 Signaling Gateway
Application Note
Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway Application Note
Table of Contents
Abstract 1
Introduction 1
Overview of SIU Operation 2
Circuit-switched API Operation 2
Transaction-based API Operation 3
Management Interface 3
Potential Points of Failure 3
Failure of SS7 Links 4
Failure of Routes 5
Failure of Power Supply 6
Failure of Signaling Interface Unit 6
Routing Architectures of a Dual-resilient SIU System 7
Dual SIU architecture for Circuit-switched Applications 10
Dual SIU architecture for Transaction-based Applications 13
Failure of IP Subnetwork 15
Failure of Application 15
Configuring a Dual SIU Pair 17
Hardware Requirements 17
System Configuration 18
Changes to the config.txt Parameter File 18
Configuring the Inter-SIU Link 18
Routing Configuration 19
Circuit Group Configuration 19
Example Configuration 19
Run-time Operations of a Dual-resilient SIU System 20
Connecting a Host to Two SIUs 20
Communicating with Both SIUA and SIUB 21
Transferring Control of a Circuit Group between SIUs 21
Activating and Deactivating Circuit Groups 21
System Initialization 21
Failure Detection 21
Transferring the Circuit Group 22
Re-synchronization of the Circuit Sate Information 22
Recovery of the Failed Unit 23
Transferring Control Back 23
Circuit Group Conflict 23
Appendix A: Frequently Asked Questions 24
Appendix B: For More Information 24
Appendix C: Abbreviations 24
Application Note Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway
Table of Figures
Figure 1 Structure of the Intel®NetStructure™ SIU520 SS7 Signaling Gateway 2
Figure 2 Integrating the SIU520 3
Figure 3 SIU Connected to Adjacent Node with Two Links in Link Set 4
Figure 4 SIU520 Connected to Mated STP Pair Providing Route Resiliency 5
Figure 5 Dual SIU Architecture 6
Figure 6 Transmit Routing to a Single Destination 7
Figure 7 Dual-resilient SIUs Connected to a Mated STP Pair in a Straight Line 8
Configuration
Figure 8 Dual-resilient SIUs Connected to a Mated STP Pair in a Crossed Link 8
Configuration
Figure 9 Transmit Routing Through Mated STPs 9
Figure 10 Normal Routing for Circuit Group 0 When Controlled by SIUA 10
Figure 11 Routing When All Local Links Have Failed, Group 0 Controlled by SIUA 11
Figure 12 Routing Following Failure of SIUA 12
Figure 13 Two Different Architectures for a TCAP Processing SIU System 13
Figure 14 Message Flow on a Dual-resilient System Running 14
the SS7 Stack up to TCAP
Figure 15 Dual LAN Operation on the SIU520 15
Figure 16 TCAP Dialogue Groups Example 16
Figure 17 Inter-SIU Link over Crossed E-1/T-1 Cable 17
Figure 18 Inter-SIU Link Set over V.11 18
Figure 19 Example Configuration to an Adjacent SSP/SCP 19
Figure 20 Example Configuration to an Adjacent STP Pair 20
Abstract
In order to achieve five-nines (99.999%) reliability and a
high degree of fault tolerance in an SS7 environment
using Intel
®
NetStructure™ SIU520 signaling gateways,
an SS7 end point spread over two signaling interface
units (SIUs) and multiple application servers can be
configured and deployed. Splitting the protocol
processing functionality of a signaling point by
implementing an SS7 node over two SIUs isolates the
hardware processors on the chassis from each other.
This separation lets one processor continue if the other
fails, allowing the system to remain in service. Distributing
application processing of a signaling point on multiple
application servers not only increases the total capacity
of a system, but also offers a higher level of fault
tolerance in the user application space.
Intel NetStructure SS7 products are designed for this
dual-processor approach and provide the architecture for
splitting a point code over two active SS7 protocol
engines. Using this technique, the links in an SS7 link set
can be spread between two separate chassis when Intel
NetStructure SS7 boards are installed in each.
This document describes the features of the SIU520 SS7
signaling gateway that are available to build SS7
solutions and reach the five-nines requirements of
telco-grade service platforms.
Introduction
This application note describes the architecture of the
Intel NetStructure SIU520 signaling gateway, reviews the
most common potential points of failure of an SS7
system based on this product, and explains methods that
can mitigate each of these potential failure points. This
document also explains in detail the configuration and
run-time operation considerations of a dual-resilient
SIU520-based system.
Because of the high expectation of service reliability by
the users of public telephone networks, equipment
manufacturers and system integrators demand high
levels of fault tolerance and availability, often citing the
five-nines for availability (requiring a system to be
operational for 99.999% of the time).
These systems need to continue to offer service even
when partial hardware or software failure has occurred.
There are several well-known methods of achieving this
type of reaction to partial failure in the signaling
component of communications networks, including:
■
Multiple signaling paths (SS7 links and link sets) to
each end point
■
Distribution of these paths through independent interfaces and cabling
■
Distribution of the processing of SS7 terminations at a
single signaling point between multiple processing
cards in a single SIU
■
Physical isolation and duplication of the SS7 interface
for a single signaling point on independent protocol
engines sharing a single point code
■
Splitting the functionality of the application layer
between multiple application servers
The first method can be achieved by implementing multiple links (64 Kb/s or 56 Kb/s channels) between two
adjacent inter-communicating points. (By definition, these
links will all be in the same link set.) The last two can be
accomplished by using two independent, but co-operating, SIU520s relaying the SS7 signaling to a distributed
application layer split over multiple application hosts.
Note: Readers should be familiar with Signaling System
7 (SS7) concepts. They should also be aware that the
information contained in this application note is provided
as a complement to the Intel NetStructure SIU520
Developer’s Manual; hence, an understanding of the
terms defined in the developer’s manual is assumed.
Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway Application Note
1
Overview of SIU Operation
The Intel NetStructure SIU520 SS7 signaling gateway is
an SS7 network access product that provides a resilient
interface to SS7 networks via a TCP/IP local area network (LAN). As shown in Figure 1, SIU520 software
includes SS7 protocol layers, as well as a configuration
and management module. The SIU520 supports multiple
SS7 signaling links within the same pulse code modulation (PCM) trunk interface or over multiple PCM trunks.
The SIU examines the timeslots carrying the SS7 information and processes them accordingly, outputting this data
to the LAN using TCP/IP. Similarly, it takes commands
from the TCP/IP LAN and converts those to SS7 signals
for transmission to the SS7 network. In the receive direction, information is conveyed to the user application in
structured messages placed in a sequential queue.
For both circuit- and transaction-related operations, the
SIU520 provides the ability to automatically distribute
signaling information between a number of physically
independent application platforms, thus providing a
degree of fault tolerance within the application space.
The SS7 signaling may be presented from the network
multiplexed in a timeslot on an E-1 (2.048 MB/s) or T-1
(1.544 MB/s, also known as DS1) bearer or by a V.35
(56/64 kb/s) synchronous serial interface.
For telephony operation (using a telephony layer 4 protocol
such as ISUP or TUP), if the SS7 signaling is multiplexed
onto a PCM bearer, the voice circuits may be passed
transparently through the SIU to the application platform
that terminates the physical circuits (see Figure 2).
Circuit-switched API Operation
The message-based application programming interface
(API) operates transparently over TCP/IP Ethernet, using
software modules provided by Intel. For circuit-switched
(telephony) applications, each application platform terminates and hence controls a fixed range of physical circuits, or circuit identification codes (CICs). CICs are configured in groups of up to 32, each group normally
equating to all the circuits in a single E-1 or T-1 trunk.
Each group is terminated on a fixed application platform
or host processor, enabling the SIU to automatically
direct API messages to the correct platform.
Application Note Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway
2
Figure 1. Structure of the Intel®NetStructure™ SIU520 SS7 Signaling Gateway
Application
#0
Configuration
and
Management
Application
#1
API Layer/Ethernet Driver
MAP or INAP or IS41
TCAP
SCCP
MTP Levels 1-3
Application
#N
Ethernet
ISUP TUP
SIU 520
Transaction-based API Operation
TCAP-based applications can be distributed on multiple
application hosts using two different methods which are
explained in details further in this document (see “Failure
of Applications”, pg. 15). These methods imply running
TC-user application parts (such as GSM-MAP, INAP, or
IS-41) on each application host. When running any
application part above TCAP on the SIU520 itself, the
product allows operation of a single host application.
Management Interface
The SIU520 constantly monitors the state of its physical
connections, PCM trunk inputs, the communication
channel via TCP/IP Ethernet to the host processors and
reports status information to an application process
running on a user-defined host. The host elected to
receive management messages can be selected by
sending an API_MSG_COMMAND management request.
(See the Intel NetStructure SIU520 Developer’s Manual
for more information.) Host 0 is used by default.
Potential Points of Failure
In this section, the most critical points of failure of an Intel
NetStructure SIU520 SS7 signaling gateway-based system are reviewed. The list of potential points of failure
include:
■
SS7 links
■
SS7 routes
■
Power supply
■
Signaling interface unit
■
IP subnetwork
■
Application host
For each of these points of failure, a solution is provided
and implementation details are given.
Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway Application Note
3
Figure 2. Integrating the SIU520
E-1 or T-1 Trunks,
Voice Circuits
CT Application Platform
Only
}
SIU520
E-1 or T-1 Trunks
SS7
Information
CT Application Platform
with SS7 Channel
and Voice Circuits
}
Ethernet
Failure of SS7 Links
Problem — With a single link to the adjacent signaling
point, service is disrupted if the link goes down for some
reason (i.e., layer 1 alarm, congestion).
Solution — Link resiliency is achieved by using multiple
links between a local point code and an adjacent point
code. By definition, such links are said to belong to the
same link set, which can contain up to 16 links. Ideally,
the links of a link set should not be carried over a unique
physical medium (such as an E-1 or T-1 trunk) but,
instead, should be split over independent physical trunks.
Link failure management is a standard MTP3 operation
and is not an SIU520-specific feature. In other words,
failure between links in a same link set will happen in a
completely transparent way for the user application.
Details — In an SIU520 system configuration, two
commands are used in config.txt to configure link sets
and links: MTP_LINKSET and MTP_LINK. In this
example, two SS7 links are defined between local point
code 0x100 and adjacent point code 0x200 on time
slot 16 of PCM ports 1 and 2. (Also see Figure 3.)
Application Note Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway
4
Figure 3. SIU Connected to Adjacent Node with Two Links in Link Set
A
* MTP_LINKSET <linkset_id> <adjacent_spc> <num_links> <flags> <local_spc> <ssf>
MTP_LINKSET 0 0x200 2 0x0000 0x100 0x08
* MTP_LINK <link_id> <linkset_id> <link_ref> <slc> <bpos> <blink> <bpos2>
*<stream> <timeslot> <flags>
MTP_LINK 0 0 0 0 0 0 0 0 16 0x0006
MTP_LINK 1 0 1 1 0 1 0 1 16 0x0006
) Load sharing between link 0 and link 1 under normal conditions
SIU520
Point Code
0x100
Link Set id 0
B) Traffic sent over link 1 under failure of link 0
SIU520
Point Code
0x100
Link id 0, slc 0
Link id 1, slc 1
SSP/SCP
Point Code
0x200
SSP/SCP
Point Code
0x200
Failure of Routes
Problem — With a single route to a destination point
code (DPC), service can be disrupted if all the links of the
link set used to reach that signaling node fail. Route
failover is a standard MTP3 operation which does not
require any specific action from the user application.
Solution — To eliminate this single point of failure, an
alternative link set can be provided in the SIU520 system
configuration to reach the same DPC. Route failover is a
standard MTP3 operation which does not require any
specific action from the user application.
Note: When an alternative route to a given DPC is
defined in an SIU520 configuration file, a choice must be
made between two different traffic modes: load sharing
or failover. In load-sharing mode, traffic sent towards the
remote signaling point is shared between the two link
sets. In failover mode, all traffic sent towards the remote
signaling point will normally be sent using the primary link
set, unless this link set fails, in which case the traffic will
use the alternative link set. See the Intel NetStructure
SIU520 Developer’s Manual for more information on the
selection of traffic mode in the MTP_ROUTE command.
Details — This example (see Figure 4) shows two link
sets (each containing one link) being used in load-sharing
mode to reach destination point code 0x400.
Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway Application Note
5
Figure 4. SIU520 Connected to Mated STP Pair Providing Route Resiliency
* MTP_LINKSET <linkset_id> <adjacent_spc> <num_links> <flags> <local_spc> <ssf>
MTP_LINKSET 0 0x200 2 0x0000 0x100 0x08
MTP_LINKSET 1 0x300 2 0x0000 0x100 0x08
* MTP_LINK <link_id> <linkset_id> <link_ref> <slc> <bpos> <blink> <bpos2>
*<stream> <timeslot> <flags>
MTP_LINK 0 0 0 0 0 0 0 0 16 0x0006
MTP_LINK 1 0 1 1 0 1 0 1 16 0x0006
MTP_LINK 2 1 0 0 0 2 0 2 16 0x0006
MTP_LINK 3 1 1 1 0 3 0 3 16 0x0006
* MTP_ROUTE <dpc> <linkset_id> <user_part_mask> <flags> <second_ls> <pc_mask>
MTP_ROUTE 0x400 0 0x0020 0x0003 1 0x00000000
A) Load sharing between link set 0 and link set 1 under normal
Link Set id 0
STPA
Link id 0, slc 0
SIU520
L
Point Code
0x100
Link Set id 1
B) Traffic sent over link set 1 under failure of STP
Link Set id 0
SIU520
Point Code
0x100
Link Set id 1
ink id 1, s
Link id 0, slc 0
L
ink id 1, slc 0
Point Code
0x200
lc 0
STPB
Point Code
0x300
STPB
Point Code
0x300
SSP/SCP
Point Code
0x400
SSP/SCP
Point Code
0x400
Failure of Power Supply
Problem — Ensuring that the unit survives the loss of
one power supply and making it possible to replace a
failed power supply without affecting the availability of the
system.
Solution — The Intel NetStructure SIU520 SS7 signaling
gateway can be optionally configured with a redundant
and hot swappable power supply.
Details — Please refer to Intel NetStructure SIU520/Intel
NetStructure SG430 Hardware User Manual, Issue 3 to
obtain part numbers for redundant power supplies for the
SIU520.
Failure of Signaling Interface Unit
Problem — Since the SIU520 provides an SS7 interface
to a distributed application, it is usually deployed for
high-density service platforms. Should the SIU of a single
SIU-based system fail, many resources (telephony circuits to TCAP dialogues) would become unavailable and
would cause major service disruption.
Solution — A major feature of the SIU architecture is
that two units can be configured to operate as a single
entity, sharing the same local SS7 point code. In normal
operation, signaling can be shared between two units. In
the event of a failure, signaling is maintained by the
remaining unit.
Details — In a dual configuration, one unit is assigned
as SIUA, the other as SIUB. Under normal operation, the
application uses both the resources of SIUA and SIUB
(see Figure 5).
The distributed layer 4 management is achieved using a
LAN connection and allows SS7 messages for any transaction or call to be received by either unit, regardless of
the unit that is actually processing the call or transaction.
The distributed MTP3 functionality is achieved using a
specially configured inter-SIU link set, containing one or
more signaling links. Transmit messages from each SIU
are load shared between links that connect to the local
SIU, if these are available. If all local network-facing SS7
links have failed, transmit messages are relayed to the
Application Note Building Fault-tolerant SS7 Systems Using the Intel®NetStructure™ SIU520 SS7 Signaling Gateway
6
Figure 5. Dual SIU Architecture
Application
Ethernet
SIUA SIUB
API Layer/Ethernet Driver
MAP or INAP
or IS41
TCAP
SCCP
MTP Levels 1-3
ISUP TUP
SS7
Distributed Layer 4
Management
Distributed MTP3
Management
Link Set
Ethernet
API Layer/Ethernet Driver
MAP or INAP
or IS41
TCAP
SCCP
MTP Levels 1-3
SS7
ISUP TUP