Abstract: A system and method for the analysis of log data is presented. The system uses SuperMinHash based locality sensitive hash signatures to describe the similarity between log lines. Signatures are created for incoming log lines and stored in signature indexes. Later similarity queries use those indexes to improve the query performance. The SuperMinHash algorithm uses a two staged approach to determine signature values, one stage uses a first random number to calculate the index of the signature value that is to update. The two staged approach improves the accuracy of the produced similarity estimation data for small sized signatures. The two staged approach may further be used to produce random numbers that are related, e.g. each created random number may be larger than its predecessors. This relation is used to optimize the algorithm by determining and terminating when further created random numbers have no influence on the created signature.

Abstract: A system and method is proposed for estimating the contribution of components of a distributed computing environment to the generation of economically relevant values, like e.g., revenue numbers. Agents are deployed to the computing environment that trace executed transactions and that monitor components used to execute those transactions. The transaction trace data also contains data about the origin/user of transactions, which may be used to group transactions corresponding to particular interactions of individual users with the monitored application into visit data. Data describing economically relevant activities of transactions, like the purchase of goods, are also observed by agents and reported in trace data. Functional dependencies described in transaction trace data and resource related dependencies derived from component monitoring data are used to identify functionality and components that contributed to the generation of business value.

Abstract: A system and method for the distributed analysis of high frequency transaction trace data to constantly categorize incoming transaction data, identify relevant transaction categories, create per-category statistical reference and current data and perform statistical tests to identify transaction categories showing overall statistically relevant performance anomalies. The relevant transaction category detection considers both the relative transaction frequency of categories compared to the overall transaction frequency and the temporal stability of a transaction category over an observation duration. The statistical data generated for the anomaly tests contains next to data describing the overall performance of transactions of a category also data describing the transaction execution context, like the number of concurrently executed transactions or transaction load during an observation period.

Abstract: A system and method for the estimation of the cardinality of large sets of transaction trace data is disclosed. The estimation is based on HyperLogLog data sketches that are capable to store cardinality relevant data of large sets with low and fixed memory requirements. The disclosure contains improvements to the known analysis methods for HyperLogLog data sketches that provide improved relative error behavior by eliminating a cardinality range dependent bias of the relative error. A new analysis method for HyperLogLog data structures is shown that uses maximum likelihood analysis methods on a Poisson based approximated probability model. In addition, a variant of the new analysis model is disclosed that uses multiple HyperLogLog data structured to directly provide estimation results for set operations like intersections or relative complement directly from the HyperLogLog input data.

Abstract: A system and method for the estimation of the cardinality of large sets of transaction trace data is disclosed. The estimation is based on HyperLogLog data sketches that are capable to store cardinality relevant data of large sets with low and fixed memory requirements. The disclosure contains improvements to the known analysis methods for HyperLogLog data sketches that provide improved relative error behavior by eliminating a cardinality range dependent bias of the relative error. A new analysis method for HyperLogLog data structures is shown that uses maximum likelihood analysis methods on a Poisson based approximated probability model. In addition, a variant of the new analysis model is disclosed that uses multiple HyperLogLog data structured to directly provide estimation results for set operations like intersections or relative complement directly from the HyperLogLog input data.

Abstract: A system and method is disclosed for the automated identification of causal relationships between a selected set of trigger events and observed abnormal conditions in a monitored computer system. On the detection of a trigger event, a focused, recursive search for recorded abnormalities in reported measurement data, topological changes or transaction load is started to identify operating conditions that explain the trigger event. The system also receives topology data from deployed agents which is used to create and maintain a topological model of the monitored system. The topological model is used to restrict the search for causal explanations of the trigger event to elements of that have a connection or interact with the element on which the trigger event occurred. This assures that only monitoring data of elements is considered that are potentially involved in the causal chain of events that led to the trigger event.

Abstract: A system and method is disclosed for the automated identification of causal relationships between a selected set of trigger events and observed abnormal conditions in a monitored computer system. On the detection of a trigger event, a focused, recursive search for recorded abnormalities in reported measurement data, topological changes or transaction load is started to identify operating conditions that explain the trigger event. The system also receives topology data from deployed agents which is used to create and maintain a topological model of the monitored system. The topological model is used to restrict the search for causal explanations of the trigger event to elements of that have a connection or interact with the element on which the trigger event occurred. This assures that only monitoring data of elements is considered that are potentially involved in the causal chain of events that led to the trigger event.

Abstract: A system and method for the distributed analysis of high frequency transaction trace data to constantly categorize incoming transaction data, identify relevant transaction categories, create per-category statistical reference and current data and perform statistical tests to identify transaction categories showing overall statistically relevant performance anomalies. The relevant transaction category detection considers both the relative transaction frequency of categories compared to the overall transaction frequency and the temporal stability of a transaction category over an observation duration. The statistical data generated for the anomaly tests contains next to data describing the overall performance of transactions of a category also data describing the transaction execution context, like the number of concurrently executed transactions or transaction load during an observation period.

Abstract: A system and method is disclosed for identifying and evaluating the business relevant impact of observed operating anomalies of monitored components of computing environments like data centers or cloud computing environments. The disclosed technology uses end-to-end transaction trace, availability and resource utilization data in combination with topology data received from agents deployed to the monitored computing environment. An abnormal operating condition is localized within a topological model of the monitored environment and has a defined temporal extent. On detection of an anomaly, affected transaction traces are selected that used the topology entity on which the anomaly was observed while the anomaly existed. Those transactions are then traced backwards, until a topology entity is reached that represents an entry point of monitored system.

Abstract: A system and method for the distributed analysis of high frequency transaction trace data to constantly categorize incoming transaction data, identify relevant transaction categories, create per-category statistical reference and current data and perform statistical tests to identify transaction categories showing overall statistically relevant performance anomalies. The relevant transaction category detection considers both the relative transaction frequency of categories compared to the overall transaction frequency and the temporal stability of a transaction category over an observation duration. The statistical data generated for the anomaly tests contains next to data describing the overall performance of transactions of a category also data describing the transaction execution context, like the number of concurrently executed transactions or transaction load during an observation period.

Abstract: A system and method for the creation of locality sensitive hash signatures using weighted feature sets is disclosed. The disclosed methodology takes advantage of discretization mechanisms commonly used in computer systems to model the influence of the feature weights on the calculated hash signature. Pseudo random numbers required for the signature calculation are created in ascending order, which enables the signature generation mechanism to identify and avoid the unnecessary creation of pseudo random numbers to improve the performance of the signature calculation process. Further, hierarchic, tree-search like algorithms are used during the processing of signature weights to further decrease the number of required random numbers. The features of the Poisson Process model, like its ability to provide random numbers in ascending order and the split- and mergeability of Poisson Processes are used to further improve the performance of the signature calculation process.

Abstract: A system and method is disclosed that analyzes a set of historic transaction traces to identify an optimized set of transaction clusters with the highest transaction frequency. The transaction clusters are defined according to multiple parameters describing the execution context of the analyzed transactions. The transaction clusters are described by coordinates in a multidimensional, hierarchical classification space. Descriptive statistical data is extracted from historic transactions corresponding to previously identified transaction clusters and stored as reference data. Transaction trace data from currently executed transactions is analyzed to find a best matching historic transaction cluster. The current transaction traces are grouped according to their corresponding historic transaction cluster.

Abstract: A system and method is disclosed for the automated identification of causal relationships between a selected set of trigger events and observed abnormal conditions in a monitored computer system. On the detection of a trigger event, a focused, recursive search for recorded abnormalities in reported measurement data, topological changes or transaction load is started to identify operating conditions that explain the trigger event. The system also receives topology data from deployed agents which is used to create and maintain a topological model of the monitored system. The topological model is used to restrict the search for causal explanations of the trigger event to elements of that have a connection or interact with the element on which the trigger event occurred. This assures that only monitoring data of elements is considered that are potentially involved in the causal chain of events that led to the trigger event.

Abstract: A system and method for the analysis of log data is presented. The system uses SuperMinHash based locality sensitive hash signatures to describe the similarity between log lines. Signatures are created for incoming log lines and stored in signature indexes. Later similarity queries use those indexes to improve the query performance. The SuperMinHash algorithm uses a two staged approach to determine signature values, one stage uses a first random number to calculate the index of the signature value that is to update. The two staged approach improves the accuracy of the produced similarity estimation data for small sized signatures. The two staged approach may further be used to produce random numbers that are related, e.g. each created random number may be larger than its predecessors. This relation is used to optimize the algorithm by determining and terminating when further created random numbers have no influence on the created signature.

Abstract: A system and method for the distributed analysis of high frequency transaction trace data to constantly categorize incoming transaction data, identify relevant transaction categories, create per-category statistical reference and current data and perform statistical tests to identify transaction categories showing overall statistically relevant performance anomalies. The relevant transaction category detection considers both the relative transaction frequency of categories compared to the overall transaction frequency and the temporal stability of a transaction category over an observation duration. The statistical data generated for the anomaly tests contains next to data describing the overall performance of transactions of a category also data describing the transaction execution context, like the number of concurrently executed transactions or transaction load during an observation period.

Abstract: A system and method for the estimation of the cardinality of large sets of transaction trace data is disclosed. The estimation is based on HyperLogLog data sketches that are capable to store cardinality relevant data of large sets with low and fixed memory requirements. The disclosure contains improvements to the known analysis methods for HyperLogLog data sketches that provide improved relative error behavior by eliminating a cardinality range dependent bias of the relative error. A new analysis method for HyperLogLog data structures is shown that uses maximum likelihood analysis methods on a Poisson based approximated probability model. In addition, a variant of the new analysis model is disclosed that uses multiple HyperLogLog data structured to directly provide estimation results for set operations like intersections or relative complement directly from the HyperLogLog input data.

Abstract: A method is disclosed that estimates causal relationships between events based on heterogeneous monitoring data. The monitoring data consists in transaction tracing data, describing the execution performance of individual transactions, resource utilization measurements of infrastructure entities like processes or operating systems and network utilization measurement data. A topology model of the monitored environment describing its entities and the communication activities of these entities is incrementally created. The location of occurred events in the topology model is determined. The topology model is used in conjunction with a domain specific causality propagation knowledge base to calculate the possibility of causal relationships between events. Different causality determination mechanisms, based on the type of involved events are used to create graphs of causal related events.

Abstract: A method is disclosed that estimates causal relationships between events based on heterogeneous monitoring data. The monitoring data consists in transaction tracing data, describing the execution performance of individual transactions, resource utilization measurements of infrastructure entities like processes or operating systems and network utilization measurement data. A topology model of the monitored environment describing its entities and the communication activities of these entities is incrementally created. The location of occurred events in the topology model is determined. The topology model is used in conjunction with a domain specific causality propagation knowledge base to calculate the possibility of causal relationships between events. Different causality determination mechanisms, based on the type of involved events are used to create graphs of causal related events.

Abstract: A system and method is disclosed that analyzes a set of historic transaction traces to identify an optimized set of transaction clusters with the highest transaction frequency. The transaction clusters are defined according to multiple parameters describing the execution context of the analyzed transactions. The transaction clusters are described by coordinates in a multidimensional, hierarchical classification space. Descriptive statistical data is extracted from historic transactions corresponding to previously identified transaction clusters and stored as reference data. Transaction trace data from currently executed transactions is analyzed to find a best matching historic transaction cluster. The current transaction traces are grouped according to their corresponding historic transaction cluster.