Abstract: Systems and methods for data anomaly detection include recommending one or more algorithms from a set of algorithms to process received time series data, wherein the one or more algorithms are recommended based at least in part on a type of workload for processing the received time series data. Assisted parameter tuning is provided for a detected anomaly alert and calibration, and the received time series data is processed based on a user selected algorithm that is parameter tuned.

Abstract: Examples described herein generally relate to receiving a query context for service events occurring on one or more networks, determining, based on the query context, a set of service events occurring on the one or more networks, querying multiple layers of a multiple-layer relational graph to determine one or more other service events having a defined relationship with the set of service events at one or more of the multiple layers, where the multiple layers include a configuration layer, an observation layer, and learned layer, defining relationships between services or service events, and indicating, via a user interface and in response to the query context, the one or more other service events.

Abstract: Examples described herein generally relate to managing a knowledge graph by providing, to an agent and based on a request from the agent, an output of a number of signals having an indicated correlation in a knowledge graph, receiving, from the agent, additional correlation information for at least a portion of the number of signals and/or additional signals, and storing, in the knowledge graph, the additional correlation information.

Abstract: Systems and methods are provided for optimizing data structures to improve the data retrieval through the use of bucketing techniques. A number of objects within an environment is drastically reduced utilizing bucketing techniques. Within the buckets, items are sequentially organized such that location is quicker. Items, or keys, are aligned with the same hash value together in a bucket and a mapping of the hash value to the offset of the first key occurrence in that bucket. This guarantees each lookup operation is only two random read accesses. Systems and methods provided herein control the pressures on a system for garbage collection and minimize memory usage with minimal impacts on performance.

Abstract: Examples described herein generally relate to receiving a query context for service events occurring on one or more networks, determining, based on the query context, a set of service events occurring on the one or more networks, querying multiple layers of a multiple-layer relational graph to determine one or more other service events having a defined relationship with the set of service events at one or more of the multiple layers, where the multiple layers include a configuration layer, an observation layer, and learned layer, defining relationships between services or service events, and indicating, via a user interface and in response to the query context, the one or more other service events.

Abstract: Systems and methods for data anomaly detection include recommending one or more algorithms from a set of algorithms to process received time series data, wherein the one or more algorithms are recommended based at least in part on a type of workload for processing the received time series data. Assisted parameter tuning is provided for a detected anomaly alert and calibration, and the received time series data is processed based on a user selected algorithm that is parameter tuned, thereby resulting in more efficient and reliable anomaly detection.

Abstract: Examples described herein generally relate to receiving a query context for service events occurring on one or more networks, determining, based on the query context, a set of service events occurring on the one or more networks, querying multiple layers of a multiple-layer relational graph to determine one or more other service events having a defined relationship with the set of service events at one or more of the multiple layers, where the multiple layers include a configuration layer, an observation layer, and learned layer, defining relationships between services or service events, and indicating, via a user interface and in response to the query context, the one or more other service events.

Abstract: Systems and methods for data anomaly detection include recommending one or more algorithms from a set of algorithms to process received time series data, wherein the one or more algorithms are recommended based at least in part on a type of workload for processing the received time series data. Assisted parameter tuning is provided for a detected anomaly alert and calibration, and the received time series data is processed based on a user selected algorithm that is parameter tuned.

Abstract: Examples described herein generally relate to managing a knowledge graph by providing, to an agent and based on a request from the agent, an output of a number of signals having an indicated correlation in a knowledge graph, receiving, from the agent, additional correlation information for at least a portion of the number of signals and/or additional signals, and storing, in the knowledge graph, the additional correlation information.

Abstract: Examples described herein generally relate to receiving a query context for service events occurring on one or more networks, determining, based on the query context, a set of service events occurring on the one or more networks, querying multiple layers of a multiple-layer relational graph to determine one or more other service events having a defined relationship with the set of service events at one or more of the multiple layers, where the multiple layers include a configuration layer, an observation layer, and learned layer, defining relationships between services or service events, and indicating, via a user interface and in response to the query context, the one or more other service events.

Abstract: Systems and methods for data anomaly detection include recommending one or more algorithms from a set of algorithms to process received time series data, wherein the one or more algorithms are recommended based at least in part on a type of workload for processing the received time series data. Assisted parameter tuning is provided for a detected anomaly alert and calibration, and the received time series data is processed based on a user selected algorithm that is parameter tuned, thereby resulting in more efficient and reliable anomaly detection.

Abstract: Systems and methods are provided for optimizing data structures to improve the data retrieval through the use of bucketing techniques. A number of objects within an environment is drastically reduced utilizing bucketing techniques. Within the buckets, items are sequentially organized such that location is quicker. Items, or keys, are aligned with the same hash value together in a bucket and a mapping of the hash value to the offset of the first key occurrence in that bucket. This guarantees each lookup operation is only two random read accesses. Systems and methods provided herein control the pressures on a system for garbage collection and minimize memory usage with minimal impacts on performance.

Abstract: A method and a processing device may be provided for state management of an operating system and applications. A framework may be provided for separating behaviorless state information from code or instructions for executing a method. Applications may have instances of state information derived from, or completely different from, instances of state information of an operating system. Instances of state information for an application may be layered over corresponding instances of state information of the operating system, such that the application and the operating system may have different views of the instances of the state information. At least one policy may be defined, which may include rules for resolving conflicts, information for providing a merged view of data from multiple repositories, default values for instances of data, as well as other information. In various implementations, referential integrity of state information may be guaranteed.

Abstract: A method and a processing device may be provided for state management of an operating system and applications. A framework may be provided for separating behaviorless state information from code or instructions for executing a method. Applications may have instances of state information derived from, or completely different from, instances of state information of an operating system. Instances of state information for an application may be layered over corresponding instances of state information of the operating system, such that the application and the operating system may have different views of the instances of the state information. At least one policy may be defined, which may include rules for resolving conflicts, information for providing a merged view of data from multiple repositories, default values for instances of data, as well as other information. In various implementations, referential integrity of state information may be guaranteed.

Abstract: A method and a processing device may be provided for state management of an operating system and applications. A framework may be provided for separating behaviorless state information from code or instructions for executing a method. Applications may have instances of state information derived from, or completely different from, instances of state information of an operating system. Instances of state information for an application may be layered over corresponding instances of state information of the operating system, such that the application and the operating system may have different views of the instances of the state information. At least one policy may be defined, which may include rules for resolving conflicts, information for providing a merged view of data from multiple repositories, default values for instances of data, as well as other information. In various implementations, referential integrity of state information may be guaranteed.

Abstract: A method and a processing device may be provided for state management of an operating system and applications. A framework may be provided for separating behaviorless state information from code or instructions for executing a method. Applications may have instances of state information derived from, or completely different from, instances of state information of an operating system. Instances of state information for an application may be layered over corresponding instances of state information of the operating system, such that the application and the operating system may have different views of the instances of the state information. At least one policy may be defined, which may include rules for resolving conflicts, information for providing a merged view of data from multiple repositories, default values for instances of data, as well as other information. In various implementations, referential integrity of state information may be guaranteed.

Abstract: A method and a processing device are provided. A group of documents may be selected from multiple documents of a search result. Associations among the selected group of documents may be determined and indicated. An indication of ones of the associations that are of interest and/or others of the associations that are of no interest may be received. A new search result may be presented, including one or more documents satisfying some or all of the associations of interest and none of the associations of no interest. In some embodiments, a document may be selected from a search result and characteristics of the document may be determined. A search result may be presented, which may include one or more documents having none or some of the characteristics of the selected document. A visual indication of a strength of an association of a document may be provided.