Abstract: A method and apparatus for determining time-varying thresholds for measured metrics are provided. With the method and apparatus, values of a given metric are captured over time. The behavior of the metric is analyzed to determine its seasonality. Correlated historical values of the metric and additional related metrics (cross-correlation) are used as inputs to a feed-forward back propagation neural network, in order to train the network to generalize the behavior of the metric. From this generalized behavior, point-by-point threshold values are calculated. The metric is monitored and the monitored values are compared with the threshold values to determine if the metric has violated its normal time-varying behavior. If so, an event is generated to notify an administrator of the error condition.

Abstract: A method is presented for correlating related transactions, such as a parent transaction that invokes a child transaction within a distributed data processing system, using a particular format for the correlation tokens. Each transaction is associated with a correlation token containing a hierarchical, three-layer identifier that includes a local transaction identifier and a local system identifier, which are associated with the local system of the child transaction, along with a root transaction identifier, a root system identifier, and a registry identifier. The local transaction identifier is unique within the local system, and the local system identifier is unique within a registry that contains a set of system identifiers. The registry is associated with a domain in which the local systems operate, and multiple domains exist within a transaction space of entities that use these correlation tokens. Correlation token pairs are analyzed to construct a call graph of related transactions.

Abstract: A method is presented for correlating related transactions, such as a parent transaction that invokes a child transaction within a distributed data processing system, using a particular format for the correlation tokens. Each transaction is associated with a correlation token containing a hierarchical, three-layer identifier that includes a local transaction identifier and a local system identifier, which are associated with the local system of the child transaction, along with a root transaction identifier, a root system identifier, and a registry identifier. The local transaction identifier is unique within the local system, and the local system identifier is unique within a registry that contains a set of system identifiers. The registry is associated with a domain in which the local systems operate, and multiple domains exist within a transaction space of entities that use these correlation tokens. Correlation token pairs are analyzed to construct a call graph of related transactions.

Abstract: A method is presented for correlating related transactions, such as a parent transaction that invokes a child transaction within a distributed data processing system, using a particular format for the correlation tokens. Each transaction is associated with a correlation token containing a hierarchical, three-layer identifier that includes a local transaction identifier and a local system identifier, which are associated with the local system of the child transaction, along with a root transaction identifier, a root system identifier, and a registry identifier. The local transaction identifier is unique within the local system, and the local system identifier is unique within a registry that contains a set of system identifiers. The registry is associated with a domain in which the local systems operate, and multiple domains exist within a transaction space of entities that use these correlation tokens. Correlation token pairs are analyzed to construct a call graph of related transactions.

Abstract: A method and apparatus for determining time-varying thresholds for measured metrics are provided. With the method and apparatus, values of a given metric are captured over time. The behavior of the metric is analyzed to determine its seasonality. Correlated historical values of the metric and additional related metrics (cross-correlation) are used as inputs to a feed-forward back propagation neural network, in order to train the network to generalize the behavior of the metric. From this generalized behavior, point-by-point threshold values are calculated. The metric is monitored and the monitored values are compared with the threshold values to determine if the metric has violated its normal time-varying behavior. If so, an event is generated to notify an administrator of the error condition.