Splunk Inc.

Maxime Petazzoni

Steven Karis

Gergely Danyi

Scott Stewart

Matthew William Pound

Charles Smith

Mayank Agarwal

Joseph Ari Ross

Steven Flanders

Sahinaz Safari Sanjani

Rajesh Raman

Arijit Mukherji

Phillip Liu

Eric Wohlstadter

Chengyu Yang

Timothy Matthew Robin Williamson

Sakshi Garg

Sonya Chang

A method of generating metrics data associated with a microservices-based application comprises ingesting a plurality of spans and mapping an ingested span of the plurality of spans to a span identity, wherein the span identity comprises a tuple of information identifying a type of span associated with the span identity, wherein the tuple of information comprises user-configured dimensions. The method further comprises grouping the ingested span by the span identity, wherein the ingested span is grouped with other spans from the plurality of spans comprising a same span identity. The method also comprises computing metrics associated with the span identity and using the metrics to generate a stream of metric data associated with the span identity.

Generating span related metric data streams by an analytic engine

A method of computing real-time metrics for automated workflows includes aggregating a set of ingested spans into a set of traces. The method further includes executing a set of rules to determine a set of workflows associated with the set of traces, wherein each workflow of the set of workflows is associated with a respective trace of the set of traces, and wherein each workflow is operable to group together activity associated with a client process within a respective trace. The method also includes assigning a name to each workflow based on the rules and computing real-time metrics for each of the workflows.

Configuring automated workflows for application performance monitoring

According to embodiments, a method for virtual partitioning of data includes receiving a data stream comprising a plurality of traces, each trace comprising a plurality of spans from a plurality of users. The method also includes assigning the plurality of traces of the data stream to a plurality of virtual partitions based on each user of the plurality of users, each virtual partition of the plurality of virtual partitions comprising data of a user of the plurality of users. The method also includes scheduling at least a subset of the plurality of virtual partitions to at least one user partition of a shared topic, the at least one user partition comprising data from at least one virtual partition of at least one user of the plurality of users. The method also includes indexing each user partition of the shared topic based on each user and each virtual partition.

Virtual partitioning of a shared message bus

A system processes data stream language expressions that combine result data streams from multiple data stream language sub-expressions. The system determines a set of fixed dimensions based on static analysis of the data stream language sub-expression. The system determines a union set representing a union of the sets of fixed dimensions. The system determines at execution time of the data stream language expression, a plurality of sets of data streams. Each set of data stream corresponds to a data stream language sub-expression from the plurality of data stream language expressions. The system correlates data streams across the plurality of sets of data streams based on the union set. The system determines result data streams for the data stream language expression by combining data values of correlated data streams.

Combining data streams generated by instrumented software using metadata correlation

A method of identifying a root cause of a failure for a trace within a microservices-based application includes determining if a root span of the trace is an error span resulting in an error experienced by a user at a front end of the microservices-based application. If the root span of the trace is an error span, the method analyzes a plurality of spans comprising the trace to determine if the trace comprises at least one leaf error span. If the trace comprises a single leaf error span, the method attributes the root cause of the failure in the trace to a service associated with the single leaf error span. If the trace comprises multiple leaf error spans the method attributes the root cause of the failure in the trace to a service associated with a leaf error span of the multiple leaf error spans comprising a latest starting timestamp.

Identifying the root cause of failure observed in connection to a workflow

A method of rendering a graphical user interface (GUI) comprising an application topology graph for a microservice architecture comprises generating a plurality of traces from a first plurality of spans generated by instrumented services in the architecture and generating generate a second plurality of spans for uninstrumented services using information extracted from the first plurality of spans. The method further comprises grouping the second plurality of spans with the plurality of traces. Subsequently, the method comprises traversing the traces and collecting a plurality of span pairs from the plurality of traces, wherein each pair of the span pairs is associated with a call between two services. The method also comprises aggregating information across the plurality of span pairs to reduce duplicative information associated with multiple occurrences of a same span pair from the plurality of span pairs. Finally, the method comprises rendering the application topology graph using the aggregated information.

Application topology graph for representing uninstrumented objects in a microservices-based architecture

Application topology graph for representing instrumented and uninstrumented objects in a microservices-based architecture

Generating metric data streams from spans ingested by a cloud deployment of an instrumentation analytics engine

A method of diagnosing anomalous patterns from metrics data associated with a microservices-based application comprises aggregating a plurality of ingested spans into a plurality of streams of metric data. The method also comprises performing computations on a stream of metric data from the plurality of streams of metric data to identify an anomalous pattern. Further, the method comprises generating an alert in response to the anomalous pattern and querying a data set using metadata associated with the alert to retrieve additional information pertaining to the anomalous pattern.

Application performance monitoring (APM) detectors for flagging application performance alerts

Graphical user interface (GUI) for representing instrumented and uninstrumented objects in a microservices-based architecture

Growing community of inventors

San Mateo, CA, United States of America

Maxime Petazzoni