WO2008093209A2

WO2008093209A2 - Network performance indicators

Info

Publication number: WO2008093209A2
Application number: PCT/IB2008/000194
Authority: WO
Inventors: Aamir Sattar; Peter Wenzel
Original assignee: Nortel Networks Limited
Priority date: 2007-01-29
Filing date: 2008-01-29
Publication date: 2008-08-07
Also published as: WO2008093209A3

Abstract

A method of monitoring network performance, comprising providing a first network performance indicator bearing on a performance of a network from an end-to-end perspective. Then, providing a second network performance indicator bearing on a performance of a node within the network, wherein the second network performance indicator is related to the first network performance indicator. The first and second network performance indicators are correlated to identify and isolate a problem within the node or network.

Description

NETWORK PERFORMANCE INDICATORS

[0001] This application claims the benefit of U.S. provisional application 60/886,991 filed January 29, 2007, the disclosure of which is incorporated herein by reference in its entirety.

Background of the Invention

[0002] Today's networks are very distributed and can comprise of several call processing nodes. A single call typically traverses across several nodes within the network. It is extremely challenging to isolate problems based on nodal alarms that are typically single event driven.

[0003] Alarm based fault management has been used to attempt to solve this problem. This methodology is very effective when it comes to identifying a failure on the node, but it is extremely challenging to ascertain how a particular failure impacts the end user from a service point of view. Moreover, contemporary service providers operate networks having multiple nodes, each generating thousands of alarms every hour. Thus, it is very difficult for the network operator to process and filter the alarm information in "near real time" to identify the alarms that are truly indicative of a service problem.

Summary of the Invention

[0004] The invention provides a way to monitor network performance in near- real time, providing:

• An indication of the end-to-end (E2E) service • A framework for quick and efficient fault isolation

• Complimentary to fault management, not a replacement

• A way to identify both nodal and network issues based upon traffic patterns, not call-by-call events

• An enabler for proactive network monitoring and trending. • Means to proactively detect and react to network service degradation [0005] Network Performance Indicators (NPIs) provide a methodology of performing near real-time monitoring of complex, distributed networks. Network Performance Indicators are useful as following:

• An indication of the end-to-end service • A framework for quick and efficient fault isolation

• Complimentary to fault management, not a replacement

• A way to identify both nodal and network issues based upon traffic patterns, not call-by-call events in "near real time"

• An enabler for proactive network monitoring and trending. • Means to proactively detect and react to network service degradation

[0006] The present invention provides two levels of NPIs, one at solution level (E2E) and other on the node level. Nodes then have a network and nodal view as well. Combined together these NPIs provide a mechanism for fault identification and isolation. [0007] A method of monitoring network performance, comprising providing a first network performance indicator bearing on a performance of a network from an end-to-end perspective. Then, providing a second network performance indicator bearing on a performance of a node within the network, wherein the second network performance indicator is related to the first network performance indicator. The first and second network performance indicators are correlated to identify and isolate a problem within the node or network. [0008] Those skilled in the art will appreciate the scope of the present invention and realize additional aspects thereof after reading the following detailed description of the preferred embodiments in association with the accompanying drawing figures.

Detailed Description of the Preferred Embodiments

[0009] The embodiments set forth below represent the necessary information to enable those skilled in the art to practice the invention and illustrate the best mode of practicing the invention. Upon reading the following description in light of the accompanying drawing figures, those skilled in the art will understand the concepts of the invention and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure and the accompanying claims. [0010] The present invention is accomplished by: • Defining and implementing solution level NPIs (network performance indicators) for a particular solution. These indicators are typically based on

CMs from the nodes.

• Implementing nodal level NPIs on all the nodes (nodal and network view).

• Providing a means to display the NPI information graphically in network operations centers.

[0011] Figure 1 illustrates the NPI solution of the present invention and nodal framework according to one embodiment of the present invention. [0012] Figure 2 shows the NPI solution of the present invention and a nodal diagram. The capability to correlate data between the E2E NPIs and nodal NPIs is key to practicing the present invention.

[0013] First the NPIs - Network Performance Indicators, are defined. They are typically metrics built upon counters/OMs which describe the traffic flow on a large scale. [0014] There are multiple types of NPIs, as shown in Figure 3: • E2E NPIs - Describe the behavior of the service from an end-to-end perspective

• Product NPIs - Describe how the product is performing, as it relates to those E2E NPI. Multiple Product NPIs exist to isolate the problem:

• Nodal Product NPIs - Isolate problems with that product • Network Product NPIs - Isolate problems with the network, from the product's viewpoint

• Secondary & Tertiary NPIs - Additional NPIs utilized for fault isolation / debugging

[0015] As illustrated in Figure 4, in order to utilize NPIs to identify, isolate, and recover the issue, the E2E NPIs indicate that there is a problem in the network, and product NPIs indicate where the problem is. The E2E & Product NPIs facilitate quick drill-down to the root cause of the disruption.

[0016] Figure 5 illustrates several events that may impact network performance, and their respective distinct fingerprints. [0017] As shown in Figure 6, the present invention allows us to assess and/or explain NPI changes over upgrades to the network. This includes product behavioral changes and OM logic changes.

[0018] The present invention also allows proactive performance monitoring.

As illustrated in Figure 7, trending enables proactive identification of issues and engineering bottlenecks in the network. Proactive trending can identify issues before they impact service.

[0019] Figure 8 illustrates an example of a scenario where performance monitoring is integrated into network activities. Performance monitoring may be integrated into all network activities, including but not limited to: • Software upgrades

• Patch applications

• Engineering activities

• Feature activation

• Routine maintenance • Network Changes

NPIs integrated into all operations ensures seamless network activities. [0020] The present invention allows for real time performance monitoring, which is the next step in network monitoring. An example is illustrated in Figure 9. Real time performance monitoring includes: • A methodology of performing real-time monitoring of complex, distributed networks

• An indication of the end-to-end service

• A framework for quick and efficient fault isolation

• Complimentary to fault management, not a replacement • A way to identify both nodal and network issues

• Based upon traffic patterns, not call-by-call events • An enabler for proactive network monitoring and trending

[0021] Today's network topologies offer new challenges, as illustrated in Figure 10:

• Networks are more distributed • Shared packet infrastructures replace TDM links

• New and more complex services are running on these networks

• The scope of a single switch has expanded.... [0022] This creates some new monitoring challenges:

• A single call can be impacted by any number of nodes, interfaces, routing issues, etc.

• Alarm correlation to isolate the source of a problem is difficult

• Avoid alarm overload - If there are too many alarms, they will be ignored

• Do not want to debug all issues via call-by-call alarming

• Generating alarms for suspected remote issues • There is a need to understand how the end-to-end service is behaving

• Alarms are inherently nodal in nature

[0023] Performance monitoring according to the present invention addresses these challenges in the following ways:

• Describes how the end-to-end service is behaving • Identifies all impacts to service, independent of alarming

• Does not require an alarm for every possible scenario

• Avoids alarm overload

• Can identify problems on external nodes / networks

• Is a methodology for understanding network behavior • Provides a framework for navigating trough the network to identify the fault location

• Is ideal for understanding performance of packet based networks

• Enables both reactive and proactive monitoring (via trending)

• Allows for proactive capacity bottleneck avoidance • Identification of problematic network components Network Operations Center (NOC)-based performance monitoring according to the present invention identifies issues that fault management cannot. [0024] Those skilled in the art will recognize improvements and modifications to the preferred embodiments of the present invention. All such improvements and modifications are considered within the scope of the concepts disclosed herein and the claims that follow.

Claims

ClaimsWhat is claimed is:

1. A method of monitoring network performance, comprising: • providing a first network performance indicator bearing on a performance of a network from an end-to-end perspective; and • providing a second network performance indicator bearing on a performance of a node within the network, wherein the second network performance indicator is related to the first network performance indicator, wherein the first and second network performance indicators are correlated to identify and isolate a problem within the node or network.