Browse Prior Art Database

Fast Class Based Service State Calculation in an RPR

IP.com Disclosure Number: IPCOM000144141D
Published in the IP.com Journal: Volume 7 Issue 1A (2007-01-25)
Included in the Prior Art Database: 2007-Jan-25
Document File: 2 page(s) / 318K

Publishing Venue

Siemens

Related People

Juergen Carstens: CONTACT

Abstract

In transport networks it is important to know if a set of services which runs over a RPR infrastructure is still available after a network outage took place. Evaluating the ability of the RPR to pass traffic even if some failure occurs is hence an important feature of the management system. Therefore, a service state mechanism has to be provided at the management system level. Therefore, it is proposed to calculate in advance the impact on the services of each class in the ring if one service fails.( According to the standard IEEE 802.17 all service classes are shown in Figure 1.) First, the bandwidth associated with the subclasses A0, A1, BCIR and BEIR is specified by the management system. Then, the maximum amounts of bandwidth that can be used by the traffic of the classes A,B,C are defined and used by the management system in order to validate the creation of new services. Based on the obtained values, a set of rules is proposed to mark a service of a class as disabled or as affected.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 2

S

Fast Class Based Service State Calculation in an RPR

Idea: Pereira Goncalo, PT-Alfragide

In transport networks it is important to know if a set of services which runs over a RPR infrastructure is still available after a network outage took place. Evaluating the ability of the RPR to pass traffic even if some failure occurs is hence an important feature of the management system. Therefore, a service state mechanism has to be provided at the management system level.

Therefore, it is proposed to calculate in advance the impact on the services of each class in the ring if one service fails.( According to the standard IEEE 802.17 all service classes are shown in Figure 1.) First, the bandwidth associated with the subclasses A0, A1, BCIR and BEIR is specified by the management system. Then, the maximum amounts of bandwidth that can be used by the traffic of the classes A,B,C are defined and used by the management system in order to validate the creation of new services. Based on the obtained values, a set of rules is proposed to mark a service of a class as disabled or as affected.

1) A service is marked as disabled if a service endpoint can not be reached physically from another endpoint (e.g. if the physical connection is broken).

2) A service is marked as disabled if some endpoint is affected. (e.g. the physical port is affected)

3) If the span is affected (the span is always affected as a whole even one ringlet is affected), it is not possible to specify which services of a certain class are affected. But if the available ammount of bandwidth of a given traffic class decreases to a level which can not meet the services, then each node has to ensure that all services of a class are affected proportionally. Following procedure is proposed to realize:

1. In case of a single failure in a span connecting two nodes:

(a) All services using this span have to be diverted to the rest of the ring

(b) First, the remaining bandwidth is allocated to all class A services. If the allocated bandwidth is sufficient for meeting the services, none of the services has to be marked as affected. If the bandwidth is not sufficient, then the services of class C have to be overrun and then they are marked as affected. If the resulting allocated bandwidth is also not sufficient, the services of class B have to b...