Microsoft Technology Licensing, LLC

Mark Raymond Gilbert

Rahul Nigam

Chandramouleeswaran Krishnaswamy

Parminder Pal Singh

Peter Kenneth Harwood

Jianfeng Cai

Mustafaraj M Dhrolia

Donovan Allen Isaak

Mangalam Rathinasabapathy

Esha Sharma

Neetha Sumana Tuluri

Vinod Mukundan Menon

Brian G O'Connor

Zohar Raz

Divyachapan Sridharan Padur

Andrei Nicolae

Jay Lu

Eladio Guzman

Priyanka Gundeli

Dmitry Valentinovich Kholodkov

Srigopal Chitrapu

Vadivelan Ramalingam

Bhuvnesh Kumar

Vinod Menon

A system for generating proportional diagrams. The system includes a user interface, a memory for storing executable program code, and one or more electronic processors. The electronic processors are functionally coupled to the memory and the user interface. The electronic processors are configured to receive a dataset from one or more servers, wherein the dataset comprises software service health data. The electronic processors are further configured to remove extraneous data from the dataset, extract relevant data points from the dataset, and determine one or more relational data strings within the extracted relevant data points. The electronic processors are also configured to determine a frequency of occurrence for each of the extracted relevant data points within the dataset, and generate the proportional relationship diagram by plotting the relevant data points, wherein the relevant data points are plotted based on the determined relational data string and the determine frequency of occurrence.

Visual representation of directional correlation of service health

In non-limiting examples of the present disclosure, systems, methods and devices for detecting and classifying service issues associated with a cloud-based service are presented. Operational event data for a plurality of operations associated with the cloud-based application service may be monitored. A statistical-based unsupervised machine learning model may be applied to the operational event data. A subset of the operational event data may be tagged as potentially being associated with a code regression, wherein the subset comprises a time series of operational event data. A neural network may be applied to the time series of operational event data, and the time series of operational event data may be flagged for follow-up if the neural network classifies the time series as relating to a positive code regression category.

Automated detection of code regressions from time-series data

Systems and methods for automatically detecting and mitigating errors in a cloud computing environment. One example method includes receiving, from a telemetry server, telemetry data for the cloud computing environment, detecting an error within the cloud computing environment based on the telemetry data, determining an error type for the error based on the telemetry data, determining an impact severity for the error based on the telemetry data, and when the error type is a reportable error type and the impact severity exceeds a predetermined threshold, performing a mitigation action.

Automated detection and classification of dynamic service outages

Systems and methods for automatically recovering from software regression in a cloud computing environment. One example method includes determining, with an electronic processor, that a batch software update has been applied to the cloud computing environment. The method includes, in response to determining that a batch software update has been applied, transmitting a problem request to an event listener server. The method includes receiving, from the event listener server, a problem statement including a stack trace. The method includes determining, based on the stack trace, a software feature indicator. The method includes transmitting the software feature indicator to a root cause analyzer. The method includes receiving, from the root cause analyzer, a change list indicator and a relevancy score associated with the change list indicator. The method includes performing a mitigation action based on the change list indicator when the relevancy score exceeds a relevancy threshold.

Software regression recovery via automated detection of problem change lists

A system for detecting service issues within multi-sourced service data. The system includes a memory and one or more electronic processors coupled to the memory. The electronic processors are configured to receive one or more data sets in near real time, and to enrich the dataset with one or more metadata parameters, pre-clean the data within the dataset, and determine one or more data points within the dataset that are semantically similar to each other. The electronic processors are also configured to generate a similarity score for each of the semantically similar data points, and determine one or more significant clusters within the dataset within a predefined lookback window. The electronic processors are also configured to analyze the determined significant clusters to determine the existence of one or more service issues, and generate a service alert based on the analysis determining that one or more service issues are present.

Semantic sweeping of metadata enriched service data

Disclosed herein is a system for automating the causality detection process when upgrades are deployed to different resources that provide a service. The resources can include physical and/or virtual resources (e.g., processing, storage, and/or networking resources) that are divided into different, geographically dispersed, resource units. To determine whether a root cause of a problem is associated with an upgrade event that has recently been deployed, a system is configured to use telemetry data to compute an upgrade-to-upgrade score that represents differences between two different upgrade events that are deployed to the same resource unit. The system is further configured to use telemetry data to compute an upgrade unit-to-unit score that represents differences between the same upgrade event being deployed to two different resource units. The scores can be used to output an alert, for an analyst, that signals whether a recently deployed upgrade event is the cause of a problem.

Causality determination of upgrade regressions via comparisons of telemetry data

Systems and methods for automatically reducing regression for a software payload applied to a plurality of computing platforms by a software updater. One example method includes receiving a health request associated with the payload, and retrieving, from an escalation engine, a plurality of identifiers identifying a subset of the plurality of computing platforms that have completed deployment of the payload, and determining a plurality of ULS tags associated with the payload. The method includes querying an anomaly detector for failure data, including pre and post-deployment data, for the subset corresponding to the ULS tags, detecting a potential software regression associated with the payload by comparing the pre and post-deployment data, and querying a root cause analyzer based on the potential regression. The method includes receiving an identifier identifying a potential root cause for the potential regression, and transmitting an event based on the potential regression and the potential root cause.

Health indicator platform for software regression reduction

Systems and methods relate to determining whether a particular failure of a web page feature is related to a recently implemented modification, and, when applicable, automatically taking one or more actions to return the web page to a functioning state (revert the modification). One embodiment provides a system for web page recovery. The system includes at least one server including an electronic processor and memory. The memory stores instructions that, when executed by the electronic processor, cause the system to obtain a web page log, detect a failure of a feature of a web page related to the web page log, identify one or more change packages related to the web page, determine a set of possible root causes for the failure of the feature based on the one or more change packages, select one possible root cause, and revert the one possible root cause.

Automated recovery of webpage functionality

Systems and methods for determining a severity level of a computing service. One system includes an electronic processor that is configured to receive telemetry data associated with one or more tenants of an online service providing services through a plurality of computing services and calculate, based on the telemetry data, a number of accesses of each of the plurality of computing services during a predetermined time period. The electronic processor is also configured to, for each of the plurality of computing services, assign a severity level to each computing service based on the number of accesses of each computing service during the predetermined time period. The electronic processor is further configured to, in response to detecting a failure of one of the plurality of computing services, initiate a response to the failure based on the severity level assigned to the one of the plurality of computing services.

Assigning a severity level to a computing service using tenant telemetry data

A system enables initiation of request driven peak-hour builds to override 'off-peak' patching schedules for updating server applications. An 'off-peak' patching schedule is generated to minimize disruption from installing builds of patches. Notwithstanding the &#x201c;off-peak&#x201d; patching schedule, a tenant administrator initiates request driven peak-hour builds when some performance failure occurs during peak business hours. For example, the tenant administrator may generate a service request that includes incident data that is usable to identify and/or develop a particular patch for resolving the performance failure. Based on the service request, the &#x201c;off-peak&#x201d; patching schedule is overridden to expedite an out-of-sequence installation of a particular patch. In this way, a tenant administrator that becomes aware that some performance failure is disrupting information workers during a peak usage time-range (e.g., business hours) is empowered to initiate a request driven peak-hour build to quickly resolve the performance failure during the peak usage time-range.

Enabling tenant administrators to initiate request driven peak-hour builds to override off-peak patching schedules

A method of and system for identifying one or more outlier anomalies in a computer environment is carried out by collecting data from a computing environment, identifying a plurality of anomalies in the computing environment based in part on the collected data, grouping the plurality of anomalies into one or more clusters, and classifying each of the one or more clusters based on a plurality of dimensions. The method may also include assigning a weight to each dimension of the plurality of dimensions for each of the one or more clusters, aggregating the weights assigned to each dimension to calculate a score for each of the one or more clusters, and generating a ranking for each of the one or more clusters base in part on the calculated score. After the rankings are generated, one of the clusters may be identified as an outlier anomaly based on the rankings. The plurality of dimensions and the weights assigned to each dimension may be selected by employing machine-learning models.

Identifying service issues by analyzing anomalies

In non-limiting examples of the present disclosure, systems, methods and devices for monitoring telemetry data from a cloud-based application service are presented. Telemetry data for a plurality of operations for the cloud-based application service may be analyzed, wherein the analysis comprises comparing a first time series with a second time series, and where data from the second time series relates to operations that were executed prior in time compared with execution of operations related to the first time series. One or more operational changes in the cloud-based application service may be identified based on the analysis, and at least one telemetry monitor may be dynamically configured based on the one or more operational changes that were identified.

Dynamic monitoring on service health signals

A system to reduce strain on server farm computing resources by over-riding 'off-peak' patching schedules in response to performance failures occurring on a server farm. Embodiments disclosed herein determine a patching schedule for causing builds of patches to be sequentially installed on server farms during an off-peak usage time-range. Responsive to a performance failure occurring on the server farm, embodiments disclosed herein identify a particular patch that is designed to resolve the performance failure. Then, the patching schedule is over-ridden to expedite an out-of-sequence installation of whichever build is first to include the particular patch. Because resolution of the performance failure is expedited, the impact of the performance failure on the computing resources of the server farm is reduced as compared to existing server farm patching systems.

Growing community of inventors

Issaquah, WA, United States of America

Mark Raymond Gilbert