Isolating the sources of faults/potential faults within computing networks

Patent metadata

Field	Value
Publication number	US-9667476-B2
Application number	US-201514846915-A
Country	US
Kind code	B2
Filing date	Sep 7, 2015
Priority date	Jan 13, 2015
Publication date	May 30, 2017
Grant date	May 30, 2017

Abstract

The health of a computing network is checked to determine whether there is a current fault in the network or a prediction of a potential fault in the network. The fault can be any type of fault, including a link failure, a failure in the transmitter of data over the link, a failure in the receiver of the data, or any other type of failure. If a fault or potential fault is indicated, a deterministic approach is provided for finding the source of the fault or potential fault within the network. The deterministic approach uses current values for selected operational parameters of components within the network, as well as historical data to determine the source.

First claim

What is claimed is:

1. A computer-implemented method of isolating sources of alerts within computing networks, said computer-implemented method comprising:
obtaining, by a control entity executing on a processor of a computing environment, one or more values for one or more operational parameters of a network of the computing environment;
detecting by the control entity an alert associated with the health of the network;
obtaining, by the control entity based on detecting the alert, at least one current value for at least one operational parameter of the one or more operational parameters; and
determining a source for the alert, wherein the determining the source for the alert comprises using at least a portion of the obtained one or more values for the one or more operational parameters and the obtained at least one current value for the at least one operational parameter to determine the source of the alert.

2. The computer-implemented method of claim 1 , wherein the obtaining the one or more values for the one or more operational parameters comprises retrieving by the control entity from an input/output subsystem coupled to the control entity the one or more values for the one or more operational parameters, wherein the input/output subsystem periodically fetches the one or more values for the one or more operational parameters from one or more components of the network using a read diagnostic parameter extended link service.

3. The computer-implemented method of claim 2 , wherein the periodically fetching is staggered over time.

4. The computer-implemented method of claim 1 , wherein the determining the source for the alert comprises:
determining that a selected operational parameter of the one or more operational parameters has a predefined relationship with respect to a selected threshold; and
based on determining that the selected operational parameter has the predefined relationship with respect to the selected threshold, using a selected equation related to the selected operational parameter to determine the source.

5. The computer-implemented method of claim 4 , wherein the selected operational parameter comprises receive power and the predefined relationship comprises below the threshold.

6. The computer-implemented method of claim 5 , wherein the selected equation comprises a receive power to transmit power ratio providing a power ratio.

7. The computer-implemented method of claim 6 , wherein the network includes a cable plant and a transmitter, and wherein the using the selected equation comprises:
determining whether the power ratio is below a specified margin;
based on the power ratio being below the specified margin, indicating a problem in the cable plant; and
based on the power ratio being equal or above the specified margin, indicating a problem in the transmitter.

8. The computer-implemented method of claim 1 , further comprising requesting information from at least one identified component of the network, based on one or more actions being performed based on determining the source.

9. The computer-implemented method of claim 1 , wherein the alert comprises one of a fault in the network or a predicted fault in the network.

10. The computer-implemented method of claim 1 , wherein the network comprises a plurality of ports, the at least a portion of the obtained one or more values for the one or more operational parameters comprises values representing speeds of multiple ports of the plurality of ports, and wherein the determining the source for the alert comprises comparing the values representing speeds of the multiple ports.

Assignees

• IBM

Inventors

• Benner Alan F
• Colonna Christopher J
• Flanagan John R
• Riedy Dale F
• Yudenfriend Harry M

Classifications

• H04L41/0631

  using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis · CPC title

• H04L41/0677 Primary

  Localisation of faults · CPC title

• H04L43/16

  Threshold monitoring · CPC title