Abstract: Methods are provided of automatically evolving a process model including process-elements and relationships between process-elements. The method comprises receiving user input defining one or more new constraints to be satisfied by the evolved process model. The method further comprises performing an optimization method to find a best solution of a cost function depending on a set of optimization conditions, over a solution space of possible consistent and/or sound process models derivable from the process model to be evolved depending on the one or more new constraints. The set of optimization conditions comprises maximization of constraints satisfaction corresponding to satisfaction of the one or more new constraints and of previous constrains from previous executions of the method. The method still further comprises updating the process model with the obtained best solution. Computer programs and systems are also provided that are suitable for performing such methods.

Abstract: Provided is a process of determining a future battery level of one or more battery-powered computing devices, the process including: accessing an event record in memory describing a scheduled event in which a user of a plurality of computing devices is scheduled to participate, inferring a subset of the plurality of computing devices to be used in that time period, determining present battery levels of the computing devices, the levels being values indicative of an amount of energy stored by batteries, determining present usage rates of battery energy, inferring battery outlooks corresponding to the scheduled event, a battery outlook being an estimated amount of energy consumption attributable to the scheduled event, and predicting future battery levels of computing devices based on at least a present battery level, a present usage rate, and a battery outlook corresponding to the scheduled event.

Abstract: First data sets associated with a job of a workflow are accessed, each first data set specifying a runtime of a job and a first plurality of feature values of a plurality of features. Feature weighting analyses are executed utilizing the first data sets to rank the plurality of features with respect to their predictive value on whether or not execution of the job is to fail. During execution of the workflow, a runtime of the job of the workflow is monitored. During execution of the job, a likelihood of failure of the job is generated based at least in part on the monitored runtime of the job and a plurality of runtimes of second data sets, the second data sets selected from the first data sets based on the rank of the plurality of features and one or more expected feature values associated with the execution of the job.

Abstract: A browser resource controller combines code metric values with a complexity analysis of rendered content to determine whether resource metric values are appropriate for a web application. The browser resource controller analyzes rendered content of a web application to generate the complexity metric values that represent the complexity of the web application. The browser resource controller also compares executable elements from the web application with exploitative code components from code repositories to determine an exploitative code risk. The browser resource controller determines a resource consumption limit for a web application based on both the exploitative code risk and the complexity metric values and compares the resource consumption limit to a detected resource consumption value.

Abstract: A digital currency driven channel assignment technique is disclosed. Each AP in a distributed network uses a channel selection manager and a distributed ledger to select channels according to a channel assignment and a digital currency associated with the distributed ledger. The digital currency incentivizes APs to make sacrifices in their channel selection for the benefit of the overall network while punishing APs that select channels selfishly and cause bandwidth interference.

Abstract: Provided is a computer system that includes a processor and a memory coupled to the processor, the memory including computer readable program code embodied therein that, when executed by the processor, causes the processor to generate a catalog that identifies a plurality of tasks that a plurality of network resources are available to perform, the network resources including Internet-of-things devices and human network resources and to generate, in response to receiving a request to perform a complex project, a solution path that includes an ordered list corresponding to selected ones of the plurality of tasks that are capable of aggregately performing the complex project, wherein the selected ones of the plurality of tasks define the solution path in an edge graph that include the plurality of tasks represented as edges therein.

Abstract: Provided is a process of determining a future battery level of one or more battery-powered computing devices, the process including: accessing an event record in memory describing a scheduled event in which a user of a plurality of computing devices is scheduled to participate, inferring a subset of the plurality of computing devices to be used in that time period, determining present battery levels of the computing devices, the levels being values indicative of an amount of energy stored by batteries, determining present usage rates of battery energy, inferring battery outlooks corresponding to the scheduled event, a battery outlook being an estimated amount of energy consumption attributable to the scheduled event, and predicting future battery levels of computing devices based on at least a present battery level, a present usage rate, and a battery outlook corresponding to the scheduled event.

Abstract: In a datacenter setting, a summary of differences and similarities between two or more states of the same or similar systems are predicted. Initially, a Long Short-Term Memory (LSTM) neural network is trained with to predict a summary describing the state change between at least two states of the datacenter. Given a set of training data (at least two datacenter states that are annotated with a state change description), the LSTM neural network learns which similarities and differences between the datacenter states correspond to the annotations. Accordingly, given a set of test data comprising at least two states of a datacenter represented by context graphs that indicate a plurality of relationships among a plurality of nodes corresponding to components of a datacenter, the LSTM neural network is able to determine a state change description that summarizes the differences and similarities between the at least two states of the datacenter.

Abstract: To aid in the root cause analysis of current system errors or anomalies, a graph-based root cause analysis software determines whether a graph representing an anomalous region of a system, referred to as a pattern, is similar to a previously stored pattern in a pattern library. The analysis software extracts a sub-graph or pattern representing components currently experiencing an anomaly from an overall system graph. The analysis software calculates a similarity score based on the comparison of the extracted pattern to patterns in the pattern library. The patterns in the pattern library represent previously encountered anomalies and include attributes, event data, expert/system administrator notes, etc., that can aid in diagnosing the current system anomaly.

Abstract: Determining a similarity between a pair of graphs or patterns can be a computationally expensive and time-consuming process. To reduce the similarity calculation costs, patterns can be simplified based on equivalent classes of components. A similarity score can be calculated between nodes of a pattern. The nodes which represent a same component type and have similar attributes will likely have a high similarity score and can be combined into a single node representing the entire class of the components. The decision to combine nodes also considers a node's topological features such as relationships and connections to other nodes. By combining equivalent nodes, the search space for mapping and determining similarity between two graphs can be reduced. Reducing the search space, exponentially reduces the number of iterations required for determining an optimal similarity score and improves the performance and scalability of the overall root cause analysis framework.

Abstract: Embodiments of the present disclosure relate to a fault tolerant root cause analysis (RCA) system that is able to handle calculation failures during runtime. Calculations (e.g., evaluation of a diagnostic model for a specific component or device) that are performed during the RCA are integrated using different resources of the system under analysis. In order to make a final diagnosis, the resources exchange messages containing calculation inputs and outputs. Calculation problems due to calculation failures in a particular resource can be resolved efficiently which reduces resource utilization and minimizes failure propagation to other parts of the system. Accordingly, the system is able to recover and output a diagnosis even if some of the resources fail or generate problems.

Abstract: A thing-sourcing project request including requirements for a thing-sourcing task that requires data input by a thing-sourcing device is received from a requestor device. A determination is made if real-time data is needed in order to complete the thing-sourcing task. In response to determining that real-time data is not needed, a determination is made if a similar thing-sourcing task has been previously completed. If not, the method determines if the thing-sourcing task can be completed using pre-existing data. If so, a data archive is searched for relevant pre-existing data that can be used to complete the thing-sourcing task. The thing-sourcing task is completed using the relevant pre-existing data, and a response to the thing-sourcing project request is transmitted to the requestor device.

Abstract: In a datacenter setting, annotations or descriptions of relevant parts or subgraphs corresponding to components in the datacenter are predicted. Given a set of training data (library of subgraphs seen in the past labeled with a textual description explaining why were they considered relevant enough to be placed in the historical database), the recurrent neural network (RNN) learns how to combine the different textual annotations coming from each relevant region into a single annotation that describes the whole system. Accordingly, given a set of input or test data (datacenter state modeled a context graph that is not annotated), the system determines which regions of the input graph are more relevant, and for each of these regions, the RNN predicts an annotation even in a previously unseen or different datacenter infrastructure.

Abstract: A method includes performing, by a processor: detecting a performance anomaly in a production computer system, generating a snapshot image of software and data that were executed on the production computer system during the performance anomaly, generating diagnostic information for the performance anomaly, communicating the diagnostic information to an experiment computer system, generating an experiment based on the diagnostic information and the snapshot image to create an experimental image, executing the experimental image on the experiment computer system to perform the experiment, and evaluating an effect of the experiment on the performance anomaly.

Abstract: Incomplete state information for nodes of a datacenter is completed utilizing historical state information. A context graph is received having a plurality of nodes that correspond to components of the datacenter, each node including properties that correspond to the represented component. It is determined that at least one of the properties for a node is incomplete. A context hash is derived from the context graph and compared to a plurality of subgraph hashes. Based upon the comparison, one or more subgraphs are identified that are similar to the context hash. Where a single similar subgraph is identified, the property information for the single subgraph is imputed to the context graph to complete the incomplete information. Where more than one similar subgraph is identified, the property information for the identified similar subgraphs is aggregated and the aggregated property information is imputed to the context graph to complete the incomplete information.

Abstract: Embodiments of the present disclosure relate to a fault tolerant root cause analysis (RCA) system that is able to handle calculation failures during runtime. Calculations (e.g., evaluation of a diagnostic model for a specific component or device) that are performed during the RCA are integrated using different resources of the system under analysis. In order to make a final diagnosis, the resources exchange messages containing calculation inputs and outputs. Calculation problems due to calculation failures in a particular resource can be resolved efficiently which reduces resource utilization and minimizes failure propagation to other parts of the system. Accordingly, the system is able to recover and output a diagnosis even if some of the resources fail or generate problems.

Abstract: Systems and methods are provided for aiding participants in meaningfully contributing to event online Q&A with minimal loss of attention to the event. As a user is inputting an event contribution fragment, probable event contribution completions are predicted. Based on the predicted event contribution completions, semantically-similar, prior-received event contributions are presented to the user. The user is permitted to act upon the similar contributions in lieu of completing and/or publishing his or her own contribution. For instance, the user may be provided the ability to vote on, comment on, or amend the semantically-similar, prior-received event contribution in lieu of completing and/or publishing his or her own contribution. Upon receiving an indication of a desired user action with respect to a semantically-similar, prior-received event contribution, the user-input event contribution fragment is discarded and the user's action on the prior-received event contribution is published in lieu thereof.

Abstract: Incomplete state information for nodes of a datacenter is completed utilizing historical state information. A context graph is received having a plurality of nodes that correspond to components of the datacenter, each node including properties that correspond to the represented component. It is determined that at least one of the properties for a node is incomplete. A context hash is derived from the context graph and compared to a plurality of subgraph hashes. Based upon the comparison, one or more subgraphs are identified that are similar to the context hash. Where a single similar subgraph is identified, the property information for the single subgraph is imputed to the context graph to complete the incomplete information. Where more than one similar subgraph is identified, the property information for the identified similar subgraphs is aggregated and the aggregated property information is imputed to the context graph to complete the incomplete information.

Abstract: In a datacenter setting, future states of nodes of a datacenter (each node representing a component of the datacenter in a context graph) are predicted. Initially, historical metrics collected from the datacenter nodes, as well as historical metrics of neighboring nodes. The metrics are aggregated into historical metric summary vector representations of the nodes which are utilized to train a future state predictor to predict future datacenter states. Once trained, metrics may be input into the future state predictor and the future state predictor may be utilized to predict a future state of one or more of the nodes of the datacenter.

Abstract: In a datacenter setting, annotations or descriptions of relevant parts or subgraphs corresponding to components in the datacenter are predicted. Given a set of training data (library of subgraphs seen in the past labeled with a textual description explaining why were they considered relevant enough to be placed in the historical database), the recurrent neural network (RNN) learns how to combine the different textual annotations coming from each relevant region into a single annotation that describes the whole system. Accordingly, given a set of input or test data (datacenter state modeled a context graph that is not annotated), the system determines which regions of the input graph are more relevant, and for each of these regions, the RNN predicts an annotation even in a previously unseen or different datacenter infrastructure.