Ca, Inc.

David Cosgrove

Timothy Pirozzi

Christopher Burke

John Murdough

Jason Normandin

Tim Diep

Novel solutions for monitoring and analyzing networks in terms of the volatility of various devices. Some solutions consider a weighted set of metrics in determining such volatility. Evaluation of devices against peers in view of these factors can produce insight about network conditions.

Monitoring network volatility

A performance anomaly detector obtains a performance metric value from a performance measurement of operation of a resource, and determines a modified Z-score based on the. Occurrence of an anomaly event for the resource is detected responsive to determining that: 1) the modified Z-score exceeds a Z-score anomaly threshold; and 2) a threshold number of other modified Z-scores, which are determined for the other performance metric values in the sequence and are within a sequence window range of the performance metric value, also exceed the Z-score anomaly threshold. A remedial action for the resource is initiated responsive to detecting the occurrence of the anomaly event for the resource.

Detecting computer system anomaly events based on modified Z-scores generated for a window of performance metrics

A method may provide performance management for a data communication network including a plurality of network elements. The method may include defining a cluster of the network elements for a performance metric, and defining one of the network elements of the cluster as a proxy network element for the cluster. A proxy baseline of the performance metric for the cluster may be calculated based on performance metric data for the proxy network element, and the performance metric for each of the network elements of the cluster may be monitored using the proxy baseline of the performance metric. Related systems and computer program products are also discussed.

Methods providing performance management using a proxy baseline and related systems and computer program products

Methods include receiving at a receiving device a plurality of reports, each corresponding to at least one item and comprising data associated with one or more performance metrics. The methods further include identifying events for each report corresponding to at least one item using the data in the report. In addition, the methods include determining a report score for each report based on a number and type of the identified events. The methods also include outputting the report scores.

Growing community of inventors

Portsmouth, NH, United States of America

David Cosgrove