International Business Machines Corporation

Rong Nickle Chang

Christopher Ward

Chunqiang Tang

Melissa J Buco

Laura Zaihua Luan

Edward Cholchin So

Tao Tao

Chun Zhang

Zon-Yin Shae

Chang-Shing Perng

Fausto Bernardini

Jean-Marc Berthaud

Byung Chul Tak

Joel D Dalsky

Shih-Chung Fang

Lee Tsiao

Philip Shi-lung Yu

Wei Tan

Joel Leonard Wolf

Jarir Kamel Chaar

Arup Acharya

Vinod Muthusamy

Dilip Dinkar Kandlur

Karthik Rajagopal Gomadam

Perng Chang-shing

Fei Wang

Hubertus Franke

Nitin Gaur

Geetika Tewari Lakshmanan

Hidayatullah Habeebullah Shaikh

Jim Alain Laredo

Yada Zhu

Eric William Brown

John Martin Prager

John Erik Wittern

Kumar Bhaskaran

Xiaohui Gu

Guerney Douglass Holloway Hunt

Ching-Farn Eric Wu

Meir M Laker

Hamed A Ellozy

Yexi Jiang

Yew-Huey Liu

Sunjit Tara

Giovanni Mariani

Klaus Sonnenleiter

Mihwa Choi

Maoyuan Qu

Robert Calhoun

Hongyi Bian

Xiaohui c/o Ibm Uk Ltd Gu

A method includes: receiving, by a processor set, a request for a web-based service; generating, by the processor set, a feature vector including values based on parameters of the request; generating, by the processor set, an endpoint selection vector including plural probabilities corresponding to plural endpoints, wherein the endpoint selection vector is generated using the feature vector with a machine learning model; selecting, by the processor set, one of the plural endpoints based on the plural probabilities; and invoking, by the processor set, the selected endpoint.

Continuously improving API service endpoint selections via adaptive reinforcement learning

A deep-learning neural network can be trained to model a probability distribution of the asset-price trends for a future time period using a training data set, which can include asset-price trends of a plurality of assets over a past time period and a latent vector sampled from a prior distribution associated with the asset-price trends of a plurality of assets. The training data set can represent a time series data. A portfolio optimization can be executed on the modeled probability distribution to estimate expected risks and returns for different portfolio diversification options.

Generative network based probabilistic portfolio management

A computer-implemented method and system for authenticating API is provided. An API invocation request associated with a user is received. An API operation and the shareable API key includes validating API key credentials of the shareable API key associated with the API invocation request. There is an additional validation of user credentials of the user associated with the API invocation request. It is determined whether the user having the validated user credentials is authorized to use the shareable API key to invoke the API operation. The API operation is executed in response to determining the user having validated user credentials is authorized to use the shareable API key to invoke the API operation. The authentication integrates validation of the user and the shareable API key, and determines whether a user is a subscriber of a multi-tenant subscription service.

Authenticating API service invocations

A method and system for managing at least one service level of a service provided by a service provider to a customer of the service provider under a service level agreement. The service level agreement is a contract between the service provider and the customer. Measurement data and at least one adjudication element associated with a service level of the at least one service level are adjudicated in accordance with the at least one adjudication element such that the adjudication generates a respective adjudicated data point for each data point of the measurement data. The measurement data includes at least one data point measured on a respective resource that provides the service. A modification history chain is generated by: initially establishing the modification history chain as being empty; and adding an identification of each adjudicated element applied to each data point of the measurement data to the modification history chain.

Managing service levels provided by service providers

A virtual storage module operable to run in a virtual machine monitor may include a wait-queue operable to store incoming block-level data requests from one or more virtual machines. In-memory metadata may store information associated with data stored in local persistent storage that is local to a host computer hosting the virtual machines. The data stored in local persistent storage replicates a subset of data in one or more virtual disks provided to the virtual machines. The virtual disks are mapped to remote storage accessible via a network connecting the virtual machines and the remote storage. A cache handling logic may be operable to handle the block-level data requests by obtaining the information in the in-memory metadata and making I/O requests to the local persistent storage or the remote storage or combination of the local persistent storage and the remote storage to service the block-level data requests.

Scalable cloud storage architecture

Network based service composition with variable distance conditions may be provided. A workflow definition may be received and a network topology may be built based on the workflow definition. Relational data may be received and a network instance may be built based on the network topology and the distance measurement. A plurality of network instances may be built, for example, for different distance conditions. One or more paths may be determined for a pair of services based on one or more of the network instances.

Network based service composition with variable conditions

A method and system for managing at least one service level of a service provided by a service provider to a customer of the service provider under a service level agreement. The service level agreement is a contract between the service provider and the customer. Measurement data and at least one adjudication element associated with a service level of the at least one service level are retrieved in an electronic format. The measurement data and the at least one adjudication element are stored in a datastore. The measurement data includes at least one data point measured on a respective resource that provides the service. The retrieved measurement data are adjudicated for the service level in accordance with the at least one adjudication element such that the adjudication generates a respective adjudicated data point for each data point of the measurement data.

Embodiments are directed to a method of developing a workflow. The method includes receiving a proposed workflow component at a local computing device of a workflow development system. The workflow development system generates a simulation stub derived from historical workflow execution logs that are relevant to the proposed workflow component. The workflow development system executes the simulation stub, which is an approximation of the proposed workflow component.

Dynamic and collaborative workflow authoring with cloud-supported live feedback

A computer-implemented method for determining resources utilized by a service request in a data processing system. The method includes determining monitored relationship types from monitoring data, determining relationship domains, determining intra-domain relationships from relationships that are internal to the relationship domains and determining cross-domain relationships from the intra-domain relationships that are linked between pairs of the relationship domains, and determining resources utilized by the service request from the intra-domain and cross-domain relationships. The domains are derived from one of the relationship types that is monitored by a single monitoring application.

System and method for determining and optimizing resources of data processing system utilized by a service request

A method for evaluating the performance of an application when migrated from a first environment in which the application is currently executing to a second, different environment includes generating a virtual application that mimics the resource consuming behavior of the application, executing the virtual application in the second environment, and evaluating the performance of the virtual application in the second environment.

Modeling and evaluating application performance in a new environment

High level characteristics are extracted from a virtual machine provisioning request stream for a cloud computing environment. A plurality of prediction models are generated for a plurality of different types of virtual machines. A regression-correlation ensemble is applied to appropriate ones of the prediction models to predict future virtual machine demand for the cloud computing environment.

Cloud provisioning accelerator

Techniques are presented for managing execution of processes on a data processing system The data processing system comprises process instances that are each an execution of a corresponding process. Each process instance comprises activity instances. Business impacts are determined for the process instances, the activity instances, or both. Order of execution of the activity instances is managed by allocating resources to activity instances in order to achieve an objective defined in terms of the business impacts. In another embodiment, requests are received for the execution of the processes. For a given request, one or more of the operations of assigning, updating, aggregating, and weighting of first business impacts associated with the given request are performed to create second business impacts associated with the given request. Additionally, requests can be modified. Modification can include changing the process requested or process input as deemed appropriate, combining related requests into a single request, or both. Unmodified requests and any modified requests are managed.

Execution order management of multiple processes on a data processing system by assigning constrained resources to the processes based on resource requirements and business impacts

A method to organize nodes into an overlay network and to disseminate multicast messages within this overlay both through an efficient tree embedded in the overlay and through gossips exchanged between overlay neighbors. Regardless of the size of the system, this invention incurs a constant low overhead on each node.

Apparatus, system and method for reliable, fast, and scalable multicast message delivery in service overlay networks

System and method for determining and optimizing resources of a data processing system utilized by a service request

Growing community of inventors

Pleasantville, NY, United States of America

Rong Nickle Chang