Method and arrangement for fault analysis in a multi-layer network

Abstract

The invention relates to a method for fault analysis in a multi-layer network. The method includes receiving a first set of performance measurements occurring in response to the fault and representing at least a first and a second layer in a multi-layer communications network. A probable set of network elements affected by the fault are localized. The type of fault is inferred from one or more symptoms in the performance measurements. The root cause of the failure is determined from a combination of information on probable set of network elements and inferred type of fault. The invention also relates to a network management node for carrying out the method. The node includes a root cause analyzer to detect affected network elements and fault type upon network failure, wherein affected network elements are determined in a spatial localization unit and the fault type is determined in a fault type inference unit.

1 . A method in a network management system for diagnosing a root cause of a fault in a multi-layer communications network, comprising:
 receiving a first set of performance measurements representing performance of network elements on at least a first and a second layer in the multi-layer communications network;   localizing the fault by identifying, based on the first set of performance measurements, a probable set of network elements affected by the fault;   inferring a type of fault from one or more symptoms incurred from one or more performance measurements in the first set of performance measurements; and   determining the root cause by combining information on the probable set of network elements with the inferred type of fault.   
     
     
         2 . The method according to  claim 1 , wherein localizing the fault includes:
 representing the first set of performance measurements by input events (e(i));   associating a set of network elements (e(i).elements) to each input event (e(i)); wherein a network element may be associated to a plurality of input events (e(i)); and   localizing the fault by identifying at least one network element associated to each of the input events (e(i)).   
     
     
         3 . The method according to  claim 1 , wherein inferring the type of fault includes:
 representing the first set of performance measurements by input events (e(i));   determining one or more symptoms following from a sequence of input events (e(i)); and   mapping at least one type of fault to the symptoms.   
     
     
         4 . The method according to  claim 3 , wherein the sequence of input events (e(i)) is tracked in a graph of events. 
     
     
         5 . The method according to  claim 4 , wherein tracking in the graph is initiated at a lower protocol level moving to a higher level in the multi-layer network. 
     
     
         6 . The method according to  claim 4 , wherein the graph is pre-generated based on basic domain knowledge of protocol layers, upon setting up of the multi-layer network. 
     
     
         7 . The method according to  claim 1 , further comprising: triggering a second set of performance measurements in addition to the measurements included in the first set of performance measurements. 
     
     
         8 . The method according to  claim 7 , wherein the first set of performance measurements includes performance measurements from multiple first measurement paths through the network elements, each measurement path including a subset of network elements, the second set of performance measurements representing performance measurements generated on at least a second measurement path, and wherein localizing the fault is repeated for the second set of performance measurements. 
     
     
         9 . The method according to  claim 7 , wherein the second set of performance measurements represents performance measurements of other types to those given in the first set of performance measurements and wherein the second set of performance measurements are provided as an input to the inferring the type of fault. 
     
     
         10 . The method according to  claim 7 , wherein the second set of performance measurements represents more frequent performance measurements than those given in the first set of performance measurements and wherein the second set of performance measurements is used to anticipate future service degradation. 
     
     
         11 . The method according to  claim 8 , wherein the second set of performance measurements is a combination of measurements on the at least second measurement path, measurements of other types and more frequent measurements to those given in the first set of performance measurements. 
     
     
         12 . A network management node in a multi-layer network comprising:
 a receiver configured to receive a first set of performance measurements representing performance of network elements on at least a first and a second layer in the multi-layer communications network;   a root cause analyzer arranged to detect affected network elements and fault type upon a network failure, wherein the root cause analyzer ( 320 ) further includes:
 a spatial localization unit configured to identify a probable set of network elements affected by the fault based on the first set of performance measurements, and 
 a fault type inference unit configured to infer a type of fault from one or more symptoms incurred from one or more performance measurements in the first set of performance measurements; and 
   a transmitter configured to output information from the root cause analyzer.   
     
     
         13 . The network management node according to  claim 12 , further comprising:
 an event correlator arranged to correlate events in time, space or protocol and to output the result to the root cause analyzer.   
     
     
         14 . The network management node according to  claim 12 , further comprising:
 a measurement generating unit adapted to communicate with the spatial localization unit and the fault type inference unit and adapted to initiate generation of a second set of measurements based on input from the spatial localization unit or the fault type inference unit.   
     
     
         15 . The network management node according to  claim 12 , wherein the receiver is further configured to receive a second set of performance measurements in addition to the measurements included in the first set of performance measurements. 
     
     
         16 . The network management node according to  claim 15 , wherein the first set of performance measurements includes performance measurements from multiple first measurement paths through the network elements, each measurement path including a subset of network elements, the second set of performance measurements representing performance measurements generated on at least a second measurement path, and wherein the step of localizing the fault is repeated for the second set of performance measurements. 
     
     
         17 . The network management node according to  claim 15 , wherein the second set of performance measurements represents performance measurements of other types to those given in the first set of performance measurements and wherein the second set of performance measurements are provided as an input to the fault type inference unit to infer the type of fault. 
     
     
         18 . The network management node according to  claim 15 , wherein the second set of performance measurements represents more frequent performance measurements than those given in the first set of performance measurements, and wherein the fault type inference unit uses the second set of performance measurements to anticipate future service degradation. 
     
     
         19 . The network management node according to  claim 16 , wherein the second set of performance measurements is a combination of measurements on the at least second measurement path, measurements of other types and more frequent measurements to those given in the first set of performance measurements.