Table of Contents:

Definition and Overview

1. Introduction

2. Softswitch Architectural Overview

3. Network Element Failure Scenarios in the Network Office

4. High-Availability Techniques for the Network Office

5. Software Failure Analysis

6. Conclusions

Self-Test

Glossary

PDF of this tutorial

This section identifies possible network element failure scenarios, including the following:

• Element overload
• Element faults: hardware and software
• Network faults

Element Overload

If the call load offered to a softswitch element is greater than the processing capacity of the platform on which the element is running, then a dangerous overload condition is reached. Any and all elements running on the affected platform could malfunction and service availability could be adversely impacted.

Element Faults

There are two categories of element faults: hardware faults and software faults. When a softswitch computing platform or related device malfunctions, this is deemed to be a hardware fault. Hardware faults can prevent the correct execution of tasks requested by other elements of the system, resulting in a potential "daisy chain" of failures. A variety of events are classified as hardware faults:

• Central processing unit (CPU) failure on a hardware platform that prevents its software from executing correctly
• Memory failure on the hardware platform that prevents its software from executing correctly
• Hard disk or other component failure on the hardware platform that prevents it from loading its software and from executing its functions
• Signaling system 7 (SS7) interface (V.35 or T1/E1) failure on the hardware platform that breaks down the communication between the platform and the public switched telephone network (PSTN) signaling network
• Integrated services digital network (ISDN) interface (T1/E1) failure on the hardware platform that breaks down the communication between the platform and the PSTN signaling network

Software faults are identified as events in the software operation of the system that cause loss of system performance or correct operation. Events such as the following are classified as software faults:

• Operating system malfunction such as a system "crash"
• Memory overrun, where the system does not have sufficient memory to allocate to the software operations that are being executed
• Deadlocked processes or threads
• Abnormal termination of processes or threads
• An abnormal state of the software preventing proper performance, from which the system cannot exit without external assistance

Network Faults

Network faults are identified as breakdowns of connectivity in the network that prevents the distributed softswitch from functioning properly. In some cases, it may be very difficult for the system to distinguish between network faults and element faults. Examples include the following:

• Ethernet interface failure on a hardware platform that breaks down communication between the platform and the network
• Disconnection or severing of a cable, attached to the Ethernet interface that prevents the hardware platform from communicating with the network
• Failure of a path-critical router or Ethernet hub
• Disconnection or severing of a cable, attached to the PSTN connector(s) that prevents the hardware platform from communicating with the PSTN
• Extreme congestion on the data network that results in the loss of ability for the softswitch to transmit data in a timely fashion onto the data network
• Persistent corruption of data received from the network
• Loss of data received from the network. This would typically manifest itself as dropped packets
• Corruption of data received from the PSTN