Abstract: State of the art approaches of exception management are reactive, manual, and intuition-driven. Command center teams often react to exceptions. They also use BI tools which only provide statistical observation, but fail to mine domain-aware insights and actionable recommendations. The disclosure herein generally relates to analyzing process exceptions, and, more particularly, to a method and system for generating a data model to analyze and predict process exceptions. The system generates a data model by using information on rules, associated properties and exceptions, as training data. The data model is further used to process information on different rules to identify exceptions, and then to generate recommendations in response to identified exceptions.

Abstract: Data inaccuracy and insufficiency are critical aspects to be analyzed to improve batch predictions, specifically in context of SLA jobs as they are foremost in affecting deliverables. Embodiments of the present disclosure provide a method and system for enhancing batch predictions by localizing jobs contributing to time deviation and generating fix recommendations by fixing data inaccuracy and insufficiency. The term fix recommendation refers to recommending a list of plausible fixes to identified causes that reduce batch prediction errors enhancing accuracy of predictions. The localization is performed by bottom-up traversing of a batch graph representing a batch process, if the batch process has a Service level Agreement (SLA) job, by narrowing down to the SLA job that has end time inaccuracies. The localization enables identifying the origin or real contributors and root cause analysis is performed for the localized jobs to generate effective fix recommendations by fixing data inaccuracy and insufficiency.

Abstract: This disclosure relates generally to method and system for predicting batch processes. Conventional batch schedulers provide a single point of control for defining and monitoring background executions in a distributed network. The method of the present disclosure obtains a set of batch jobs from one or more users to generate a set of batch graphs by deriving a metadata. Further, a set of batch models is generated for the set of batch graphs. The set of batch models includes at least one of a forecasting model, a job-job regression model, and a job-workload regression model. Further, a batch job schedule is generated for the set of batch graphs to predict a revised batch job schedule with a real time feed and the set of batch models. Additionally, a proactive notification is sent to each user alarming one or more unexpected delays indicating the revised batch job schedule.

Abstract: A trouble ticket may contain various details of an observed issue and information of the actual issue which is hidden in the trouble ticket description along with other information. Embodiments herein describe an algorithm of algorithms (AoA) layer approach for extracting issues from trouble ticket descriptions. The method identifies the type of trouble ticket whether it is machine generated or user generated trouble tickets and accordingly selects the appropriate algorithm for issue extraction. It selects a relevant algorithm for the machine generated trouble tickets, and user generated trouble tickets based on trouble ticket data properties, historical performance, and a user feedback. It identifies right mix of columns from a service catalog to be used for matching based on presence of weighted keywords in the input data and identifies high-weighted keywords from input data descriptions and compares them with selected columns from catalog to generate accurate matches with their similarity score.

Abstract: A system for smart alerts in a batch system for an IT enterprise. The method includes alert configuration by identifying recent steady state of a batch job and deriving schedules for the steady state. The normal behaviour is then computed within the schedules. The method further includes aggregating the one or more alerts by identifying correlated group of alerts by pruning of one or more jobs and alerts, detecting correlations between the two or more alerts and deriving causality of the grouped alerts. The method finally includes predicting of future alerts of a batch job.

Abstract: A method for managing alerts generated in an enterprise environment is described. The method comprises discarding, by a processor, one or more duplicate alerts based on one or more predetermined rules, wherein a duplicate alert is an instance of an alert occurring within a duplicate alert arrival time associated with the alert. The method further comprises obtaining an alert correlation signature associated with the alert from a database, wherein the alert correlation signature comprises information associated with one or more alerts correlated to the alert. The method further comprises, obtaining, from the database, a temporal signature associated with the alert, wherein the temporal signature indicates a probable time of occurrence of the alert. Further, the alert, the alert correlation signature, and the temporal signature are transmitted to a computing device of a user of the enterprise environment.

Abstract: Systems and methods for generating a blueprint for an enterprise is provided. The system obtains information pertaining to an enterprise from data sources, extracts entities from the information, and identifies relationships across the entities, performs conflict resolution check on the information received from the data sources based on pre-defined rules, identifies one or more missing links in the information. The system further computes a completeness score for the information based on at least one of the conflict resolution check and the missing links, and generates a blueprint of the information and the one or more entities for the enterprise. Authentic data sources are identified from the data sources based on at least one of assigned weight, and one or more instances of, and contribution by, the data sources in the information. The completeness score is computed based on the instances and contribution of the sources in the information.

Abstract: Described herein are a method and a system for extracting issues based on ticket mining. In one implementation, a plurality of tickets comprising descriptions of the issues in computing systems are received. The received descriptions are then cleaned by removing unwanted details. Upon cleaning, the clean descriptions are mapped with descriptions stored in service catalog data to obtain unmapped clean descriptions. In an example, the unmapped clean descriptions include one of user-generated descriptions, system-generated descriptions, and both the user-generated descriptions and the system-generated descriptions. For the user-generated descriptions; the issues are extracted by pre-processing the user-generated descriptions, determining keywords from the processed unmapped clean descriptions, constructing n-grams of keywords from the extracted keywords, and extracting the n-grams of keywords as the issues present in the computing systems.

Abstract: System and method for analyzing and prioritizing issues for automation are disclosed. In an embodiment, a method for addressing issues associated with automation, and more particularly to, system and method for analyzing and prioritizing issues for automation is disclosed. The problem of analyzing and prioritizing issues for automation that result in benefits is disclosed. The present disclosure proposes to select only those issues for automation that are expected to last long enough to repay cost invested in their automation. Further, the proposed approach prioritizes the issues on the basis of benefits realized by automation of an issue with respect to cost of implementing automation and savings in the cost of resolution of the issue due to automation. Furthermore, the proposed approach compares and contrasts two case studies and demonstrates how the inherent differences about the domain get captured through this approach and are reflected in various steps of automation prioritization.

Abstract: Systems and methods for generating a blueprint for an enterprise is provided. The system obtains information pertaining to an enterprise from data sources, extracts entities from the information, and identifies relationships across the entities, performs conflict resolution check on the information received from the data sources based on pre-defined rules, identifies one or more missing links in the information. The system further computes a completeness score for the information based on at least one of the conflict resolution check and the missing links, and generates a blueprint of the information and the one or more entities for the enterprise. Authentic data sources are identified from the data sources based on at least one of assigned weight, and one or more instances of, and contribution by, the data sources in the information. The completeness score is computed based on the instances and contribution of the sources in the information.

Abstract: System and method for prioritizing and planning issues in automation is disclosed. The system and method prioritizes the issues by considering various cost and benefit factors associated with automation of the issues. The issues that are best suited for automation are identified and a plan for automation prioritization of the issues is presented. An approach to evaluate day-zero automation potential is disclosed. Further, a comprehensive blueprint of each issue is built in the form of an operations graph. The blueprint of each issue is analyzed and the issue is evaluated on various aspects of cost, quality, and risk. Further, a systematic plan for automation is made and a business case is generated.

Abstract: A system for smart alerts in a batch system for an IT enterprise. The method includes alert configuration by identifying recent steady state of a batch job and deriving schedules for the steady state. The normal behaviour is then computed within the schedules. The method further includes aggregating the one or more alerts by identifying correlated group of alerts by pruning of one or more jobs and alerts, detecting correlations between the two or more alerts and deriving causality of the grouped alerts. The method finally includes predicting of future alerts of a batch job.

Abstract: A method for managing alerts generated in an enterprise environment is described. The method comprises discarding, by a processor, one or more duplicate alerts based on one or more predetermined rules, wherein a duplicate alert is an instance of an alert occurring within a duplicate alert arrival time associated with the alert. The method further comprises obtaining an alert correlation signature associated with the alert from a database, wherein the alert correlation signature comprises information associated with one or more alerts correlated to the alert. The method further comprises, obtaining, from the database, a temporal signature associated with the alert, wherein the temporal signature indicates a probable time of occurrence of the alert. Further, the alert, the alert correlation signature, and the temporal signature are transmitted to a computing device of a user of the enterprise environment.

Abstract: This disclosure relates generally to computing devices, and more particularly to transition of IT operations. In one embodiment, a method and system is provided for generating an efficient transition plan for IT operations while addressing aspects such as coverage, risk, time, and cost. The IT operations are modeled through graphs and use well-defined problems in graph theory to build solutions. Heavy hitter issues are identified to maximize coverage. To minimize risk, severity of an issue is determined, wherein the severity is based on the instability caused or penalties associated with the issue. Further, transition time is minimized by finding issue-communities for parallel transition by finding maximum cliques. Yet further, the bin-packing algorithm is used to optimize the teams of resolvers and thus minimize cost. Finally, a transition plan is generated by systematically identifying issue communities for transition using the minimum hitting set and minimum vertex cover problem.

Abstract: Described herein are a method and a system for extracting issues based on ticket mining. In one implementation, a plurality of tickets comprising descriptions of the issues in computing systems are received. The received descriptions are then cleaned by removing unwanted details. Upon cleaning, the clean descriptions are mapped with descriptions stored in service catalog data to obtain unmapped clean descriptions. In an example, the unmapped clean descriptions include one of user-generated descriptions, system-generated descriptions, and both the user-generated descriptions and the system-generated descriptions. For the user-generated descriptions; the issues are extracted by pre-processing the user-generated descriptions, determining keywords from the processed unmapped clean descriptions, constructing n-grams of keywords from the extracted keywords, and extracting the n-grams of keywords as the issues present in the computing systems.

Abstract: Systems and Methods for detection and localization of performance faults in data centers are described. In one embodiment, a method comprises identifying a performance fault in a data center upon detection of the performance fault at any of a plurality of monitors in the data center, wherein the plurality of monitors are placed at monitor nodes, amongst a plurality of nodes, in the data center. Further, the method comprises evaluating a fault vector for the data center upon identification of the performance fault, wherein the fault vector is evaluated based on a fault indicator corresponding to each of the plurality of monitors. Based on the comparison of the fault vector with signatures of each of the plurality of nodes, one or more faulty nodes, amongst the plurality of nodes in the data center, are determined as likely root cause of the performance fault.

Abstract: Systems and Methods for detection and localization of performance faults in data centers are described. In one embodiment, a method comprises identifying a performance fault in a data center upon detection of the performance fault at any of a plurality of monitors in the data center, wherein the plurality of monitors are placed at monitor nodes, amongst a plurality of nodes, in the data center. Further, the method comprises evaluating a fault vector for the data center upon identification of the performance fault, wherein the fault vector is evaluated based on a fault indicator corresponding to each of the plurality of monitors. Based on the comparison of the fault vector with signatures of each of the plurality of nodes, one or ore faulty nodes, amongst the plurality of nodes in the data center, are determined as likely root cause of the performance fault.