Abstract: In exemplary aspects of optimizing data centers, historical data corresponding to a data center is collected. The data center includes a plurality of systems. A data center representation is generated. The data center representation can be one or more of a schematic and a collection of data from among the historical data. The data center representation is encoded into a neural network model. The neural network model is trained using at least a portion of the historical data. The trained model is deployed using a first set of inputs, causing the model to generate one or more output values for managing or optimizing the data center with respect to design and risk aspects.

Abstract: Recommending analytic tasks based on similarity of datasets is disclosed. One example is a system including a data processor, a matching module, and a recommendation module. The data processor receives an incoming dataset via a processing system, and generates a feature vector for the incoming dataset. The matching module determines similarity measures between the generated feature vector and representative feature vectors for a plurality of datasets in a data repository, and selects at least one dataset of the plurality of datasets based on the similarity measures. The recommendation module identifies at least one analytic task associated with the selected dataset, and recommends, to a computing device via the processing system, the at least one analytic task to be performed on the incoming dataset.

Abstract: Systems and methods are provided for optimizing parameters of a system across an entire stack, including algorithms layer, toolchain layer, execution or runtime layer, and hardware layer. Results from the layer-specific optimization functions of each domain can be consolidated using one or more consolidation optimization functions to consolidate the layer-specific optimization results, capturing the relationship between the different layers of the stack. Continuous monitoring of the programming model during execution may be implemented and can enable the programming model to self-adjust based on real-time performance metrics. In this way, programmers and system administrators are relieved of the need for domain knowledge and are offered a systematic way for continuous optimization (rather than an ad hoc approach).

Abstract: Systems and methods are provided for optimizing parameters of a system across an entire stack, including algorithms layer, toolchain layer, execution or runtime layer, and hardware layer. Results from the layer-specific optimization functions of each domain can be consolidated using one or more consolidation optimization functions to consolidate the layer-specific optimization results, capturing the relationship between the different layers of the stack. Continuous monitoring of the programming model during execution may be implemented and can enable the programming model to self-adjust based on real-time performance metrics. In this way, programmers and system administrators are relieved of the need for domain knowledge and are offered a systematic way for continuous optimization (rather than an ad hoc approach).

Abstract: In exemplary aspects of optimizing data centers, historical data corresponding to a data center is collected. The data center includes a plurality of systems. A data center representation is generated. The data center representation can be one or more of a schematic and a collection of data from among the historical data. The data center representation is encoded into a neural network model. The neural network model is trained using at least a portion of the historical data. The trained model is deployed using a first set of inputs, causing the model to generate one or more output values for managing or optimizing the data center using supplemental indicators.

Abstract: In exemplary aspects of managing, monitoring and maintaining computing systems and devices such as edge data centers (EDCs), probabilistic models such as dynamic Bayesian networks (DBNs) are generated. The DBNs can define individual and collective systems such as EDCs. The DBNs are built by generating or estimating the model structure and model parameters. The model can be deployed, for instance, to identify actual or potentially anomalous behavior within the individual or collective systems defined by the model. The model can also be deployed to predict anomalous behavior. Based on the results of the model, corrective measures can be taken to remedy the anomalies, and/or to optimize the impact therefrom.

Abstract: Examples herein involve preforming a simulation of a simulated model using precomputed results of the simulation with predetermined values for a parameter set of the simulated model. In examples herein, a test sample set is selected from a sample subsets repository, and using the test sample set, determining results of a simulation of the simulated model for the test parameters.

Abstract: In exemplary aspects of optimizing data centers, historical data corresponding to a data center is collected. The data center includes a plurality of systems. A data center representation is generated. The data center representation can be one or more of a schematic and a collection of data from among the historical data. The data center representation is encoded into a neural network model. The neural network model is trained using at least a portion of the historical data. The trained model is deployed using a first set of inputs, causing the model to generate one or more output values for managing or optimizing the data center with respect to design and risk aspects.

Abstract: In exemplary aspects of optimizing data centers, historical data corresponding to a data center is collected. The data center includes a plurality of systems. A data center representation is generated. The data center representation can be one or more of a schematic and a collection of data from among the historical data. The data center representation is encoded into a neural network model. The neural network model is trained using at least a portion of the historical data. The trained model is deployed using a first set of inputs, causing the model to generate one or more output values for managing or optimizing the data center using supplemental indicators.

Abstract: One embodiment is a method that receives a description of services desired by a service requestor and then crawls web sites to extract information on services offered by service providers. The extracted information is used to transform the description of services desired by the service requestor into an improved description of services.

Abstract: Systems and methods are provided for performing a fast simulation using test parameter vectors as inputs. The method includes retrieving precomputed samples from a sample repository stored in a non-volatile memory, the precomputed samples being precomputed using a simulated model, predetermined parameter vectors, and random inputs; storing respective subsets of the precomputed samples in local memories of a plurality of respective hardware processors; storing the test parameter vectors in the local memories of the hardware processors; at each of the hardware processors, selecting a subset of the precomputed samples stored in the local memory of the hardware processor based on the test parameter vectors, computing test samples by executing the simulated model using the test parameter vectors and the random inputs; and combining the subset of the precomputed samples and the test samples to produce a simulation result.

Abstract: Systems and methods are provided for performing a fast simulation using test parameter vectors as inputs. The method includes retrieving precomputed samples from a sample repository stored in a non-volatile memory, the precomputed samples being precomputed using a simulated model, predetermined parameter vectors, and random inputs; storing respective subsets of the precomputed samples in local memories of a plurality of respective hardware processors; storing the test parameter vectors in the local memories of the hardware processors; at each of the hardware processors, selecting a subset of the precomputed samples stored in the local memory of the hardware processor based on the test parameter vectors, computing test samples by executing the simulated model using the test parameter vectors and the random inputs; and combining the subset of the precomputed samples and the test samples to produce a simulation result.

Abstract: Recommending analytic tasks based on similarity of datasets is disclosed. One example is a system including a data processor, a matching module, and a recommendation module. The data processor receives an incoming dataset via a processing system, and generates a feature vector for the incoming dataset. The matching module determines similarity measures between the generated feature vector and representative feature vectors for a plurality of datasets in a data repository, and selects at least one dataset of the plurality of datasets based on the similarity measures. The recommendation module identifies at least one analytic task associated with the selected dataset, and recommends, to a computing device via the processing system, the at least one analytic task to be performed on the incoming dataset.

Abstract: Identifying a subset of signifiers to analyze can include determining a set of distance metrics between a first signifier and each of a plurality of second signifiers, identifying a subset of the plurality of second signifiers to analyze based on the set of distance metrics using a computing device, and determining a relation between the subset of the plurality of second signifiers and the first signifier based a subset of the set of distance metrics.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to provide candidate services for an application. An example method includes determining a plurality of candidate services for a cloud application, determining an indication that a first candidate service from the plurality of candidate services is more relevant to the cloud application than a second candidate service based on a first prediction score corresponding to the first candidate service and a second prediction score corresponding to the second candidate service; presenting the first candidate service and the second candidate service to a user based on the first prediction score and the second prediction score; and adjusting a first weight corresponding to the first candidate service and a second weight corresponding to the second candidate service based on whether the first candidate service or the second candidate service is selected for inclusion in the cloud application.

Abstract: Clustering signifiers in a semantics graph can comprise coarsening a semantics graph associated with an enterprise communication network containing a plurality of nodes into a number of sub-graphs containing supernodes; partitioning each of the number of sub-graphs into a number of clusters; and iteratively refining the number of clusters to reduce an edge-cut of the semantics graph, based on the number of clusters.

Abstract: To model a computing entity, information relating to transactions associated with the computing entity is received. The received information forms a collection of information. The collection is segmented into a plurality of segments, and at least one anomalous segment is identified. A model of the computing entity is built.

Abstract: Building a semantics graph for an enterprise communication network can include extracting a first signifier and a second signifier from the enterprise communication network, determining a semantic proximity of the first signifier and the second signifier using an engine executing computer readable instructions, and building the semantics graph, wherein the first signifier and the second signifier are represented as nodes connected by an edge representing the semantic proximity of the first signifier and the second signifier.

Abstract: One embodiment collects performance data for an application server that processes transactions received from a client computer to a database server. An application log is created from the performance data and used for capacity planning in a multi-tiered architecture.

Abstract: One embodiment is a method that builds a standardized web form that includes information extracted from multiple web forms retrieved over a web from different service providers. The standardized web form is used to retrieve price quotes from the different service providers.