International Business Machines Corporation

Cisco Technology, Inc.

Rohit Bahl

Stephen Williams

Debashish Ghosh

Alan Yeung

Adam R Neal

Samuel S Lee

Andrew Russell Low

Ramesh Yeevani-Srinivas

Eric O Funk

Harsh Parandekar

Keith W Campbell

Paul Clyde Sherrill

Various techniques are used to schedule computing jobs for execution by a computing resource. In an example method, a schedule is generated by selecting, for a first slot in the schedule, a first computing job based on a first priority of the first computing job with respect to a first characteristic. A second computing job is selected for a second slot in the schedule based on a second priority of the second computing job with respect to a second characteristic. The second slot occurs after the first slot in the schedule, and the second characteristic is different than the first characteristic. The first characteristic or the second characteristic includes an execution frequency. The computing jobs are executed based on the schedule.

Evolutionary modelling based non-disruptive scheduling and management of computation jobs

A method comprises, at a local participant device, establishing audio connections with remote participant devices over a network for an online voice conference. The method includes generating ultrasound signals for corresponding ones of the remote participant devices, and transmitting the ultrasound signals over corresponding ones of the audio connections. The method further includes collecting indications, transmitted by corresponding ones of the remote participant devices over the network, that indicate whether the corresponding ones of the remote participant devices detected the ultrasound signals. The method includes identifying which of the remote participant devices detected the ultrasound signals based on the indications, and localizing degraded voice quality to particular ones of the local participant device and the remote participant devices based, at least in part, on results of identifying.

Automatic discovery and localization of voice degradation faults using ultrasound techniques

A multi-cloud service mesh orchestration platform can receive a request to deploy an application as a service mesh application. The platform can tag the application with governance information (e.g., TCO, SLA, provisioning, deployment, and operational criteria). The platform can partition the application into its constituent components, and tag each component with individual governance information. For first time steps, the platform can select and perform a first set of actions for deploying each component to obtain individual rewards, state transitions, and expected returns. The platform can determine a reinforcement learning policy for each component that maximizes a total reward for the application based on the individual rewards, state transitions, and expected returns of each first set of actions selected and performed for each component. For second time steps, the platform can select and perform a second set of actions for each component based on the reinforcement learning policy for the component.

Systems and methods for orchestrating microservice containers interconnected via a service mesh in a multi-cloud environment based on a reinforcement learning policy

Differentiated sidecars in a service mesh may be provided. A first routing rule includes a first plurality of weights to be associated with a first plurality of data paths of a first microservice instance may be received. Next, first mapping between a first set of features associated with the first microservice instance and the first plurality of weights may be determined. Then a second microservice instance may be detected and a second set of features associated with the second microservice instance may be detected. A second routing rule comprising a second plurality of weights to be associated with a second plurality of data paths of the second microservice instance may be determined. The second plurality of weights may be determined such that a second mapping between the second set of features and the second plurality of weights imitates the first mapping.

Differentiated smart sidecars in a service mesh

In an embodiment, one or more non-transitory computer-readable storage media store one or more sequences of instructions, which when executed using one or more processors causes the one or more processors to perform various functions, such as accessing data stored in memory regarding a plurality of computation jobs. Such data includes, for instance, periodicity constraints that specify execution frequencies for the plurality of computation jobs, categorization data that categorizes the plurality of computation jobs into a plurality of job types, and organization data that organizes the plurality of computation jobs into a plurality of ordered arrangements. In this embodiment, there are at least a first ordered arrangement and a second ordered arrangement, each ordered arrangement comprises references to each of the plurality of computation jobs according to job type, and the job types are associated with relative priority indicia. The functions performed by the one or more processors also include populating a resulting ordered arrangement by selecting priority indicia from between respective job types in the first ordered arrangement and second ordered arrangement, and generating a schedule for execution of the plurality of computation jobs in accordance with the relative priority indicia in the resulting ordered arrangement and the periodicity constraints. The one or more processors are also configured to execute the one or more sequences of instructions to cause execution of the plurality of computation jobs according to the schedule.

In one embodiment, a system includes: a plurality of cloud nodes implemented on computing devices, the plurality of cloud nodes configured to execute computing jobs in a cloud computing environment according to a schedule, a schedule optimizer configured to: use a machine learning model to determine functional intent for job requests according to at least job execution metadata, and generate a schedule recommendation for jobs associated with the job requests, where the schedule recommendation is generated based at least in part on the functional intent; and a job executor configured to provide the schedule recommendation as the schedule to the plurality of cloud nodes, where the schedule optimizer and the job executor are instantiated in memory and executed by processing circuitry on at least one computing device.

Cloud resources optimization

Techniques are described for schema lifecycle management. In one example, a method includes dividing, with one or more processing devices, a plurality of statements of a data definition language (DDL) script into a plurality of groups, the plurality of groups configured to be executed in parallel by a persistence layer. The method further includes assigning, with the one or more processing devices, one or more of the plurality of groups to a plurality of worker applications. The method further includes executing, with the one or more processing devices, constituent statements of the plurality of statements of the DDL script within each group of the plurality of groups by the plurality of worker applications in a session.

Schema lifecycle manager

In one embodiment, a method includes receiving at a network device comprising a data transfer optimization module, input identifying a source directory comprising data and a target directory at a storage device, splitting the data into a plurality of data groups for transfer to the storage device, transmitting the data groups concurrently to the storage device on data paths wherein transfer times of the data groups are monitored, receiving at the data transfer optimization module, identification of a data group with a longest transfer time, splitting the data group with the longest transfer time, and placing the data from the data group into at least two of the data groups for transfer to the storage device. An apparatus and logic are also disclosed herein.

Method and apparatus for optimizing data transfers utilizing machine learning

Techniques are described for a system for providing expert information by verified experts in response to currently circulating assertions such as social media postings. In one example, a method includes identifying one or more authors of respective contents issued by one or more registered expert content sources as verified experts associated with tagged topics that are tagged in association with the respective contents. The method further includes detecting, in one or more media, a media content related to a topic of interest that corresponds to one or more of the tagged topics with which one or more of the verified experts are associated. The method further includes outputting the media content related to the topic of interest to a response interface made available to one or more of the verified experts associated with the one or more of the tagged topics that correspond to the topic of interest.

Management of information from verified experts

Techniques are described for using stored function instantiations, together with user queries. In one example, a method includes receiving a query from a user that contains one or more parameters. The method further includes comparing one or more parameters of the query with one or more parameters of one or more stored function instantiations. The method further includes determining a pre-existing function instantiation from the one or more stored function instantiations that have one or more parameters in common with the query. The method further includes identifying one or more parameters that are not in common between the query and the pre-existing function instantiation. The method further includes modifying the pre-existing function instantiation into a repurposed function instantiation that incorporates the one or more parameters from the query that are not in common between the query and the pre-existing function instantiation. The method further includes executing the repurposed function instantiation.

Modified data query function instantiations

A method for transforming data structures includes processing a portion of a source data structure (SDS) with a first processing resource by associating the first processing resource with a target data structure (TDS) key element located in a key position of a queue, determining whether the queue includes an SDS node element corresponding to the key position, processing the SDS node element responsive to determining that the queue includes the SDS node element corresponding to the key position, wherein the processing the SDS node element comprises retrieving the SDS node element from the queue, processing the retrieved SDS node element to generate a resultant TDS node element, outputting the resultant TDS node element to an empty key position in the queue, and mapping the resultant TDS node element to the TDS.

Mapping data structures

A technique for annotating and synchronizing on a device is provided. A client root model of fragments is marked with annotations in response to triggers received by an application of the device. Marking the client root model of the fragments with annotations includes at least: in response to the any of the fragments being extracted out and flagged, marking the client root model with annotations; in response to binding any of the fragments to a source, marking the client root model of the fragments with annotations and marking the client root model as flagged; and in response to sharing in any the of fragments, marking the client root model with annotations.

Growing community of inventors

Kanata, Canada

Rohit Bahl