-
公开(公告)号：US20230176939A1
公开(公告)日：2023-06-08
申请号：US17541453
申请日：2021-12-03
Applicant: International Business Machines Corporation
Inventor： Joshua M. Rosenkranz , Pranita Sharad Dewan , Mudhakar Srivatsa , Praveen Jayachandran , Chander Govindarajan , Priyanka Prakash Naik , Kavya Govindarajan
IPC: G06F11/07
CPC classification number: G06F11/076 , G06F11/079 , G06F11/0709
Abstract: An ensemble of autoencoder models can be trained using different seeds. The trained ensemble of autoencoder models can be run on new time series data to generate a prediction associated with the new time series data. The new time series data can include multiple dimensions per time step. Reconstruction errors can be determined for the prediction. Dimensions having highest reconstruction errors can be selected among the multiple dimensions based on a threshold. The prediction can be segmented based on bursts of the reconstruction errors over time, where temporal segments can be obtained. At least one common pattern including a set of dimensions among the selected dimensions across the temporal segments can be obtained to represent a failure fingerprint.
-
2.发明申请
公开(公告)号：US20230030795A1
公开(公告)日：2023-02-02
申请号：US17391546
申请日：2021-08-02
Applicant: International Business Machines Corporation
Abstract: Methods, systems, and computer program products for an automated resource request mechanism for heterogeneous infrastructure using profiling information are provided herein. A computer-implemented method includes obtaining resource utilization information, pertaining to multiple system resources, from multiple heterogeneous system infrastructure deployments; automatically learning resource interdependencies for the heterogeneous system infrastructure deployments by processing at least a portion of the resource utilization information using a first set of machine learning techniques; automatically determining performance profiles, with respect to the multiple system resources, for the multiple heterogeneous system infrastructure deployments by processing at least a portion of the resource utilization information and at least a portion of the learned resource interdependencies using a second set of machine learning techniques; predicting resource requests for at least one of the heterogeneous system infrastructure deployments using the determined performance profiles; and performing automated actions based on the resource request predictions.
-
公开(公告)号：US20200076571A1
公开(公告)日：2020-03-05
申请号：US16115988
申请日：2018-08-29
Applicant: International Business Machines Corporation
Inventor： Senthilnathan Natarajan , Chander Govindarajan , Manish Sethi , Adarsh Saraf
Abstract: An example operation may include one or more of in one or more peer nodes of a plurality of peer nodes of a blockchain network that stores a blockchain and a state database, periodically generating a state database checkpoint, obtaining a consensus on the state database checkpoint from one or more of the one or more peer nodes, and storing the consensus state database checkpoint.
-
公开(公告)号：US20200074458A1
公开(公告)日：2020-03-05
申请号：US16117018
申请日：2018-08-30
Applicant: International Business Machines Corporation
Inventor： Chander Govindarajan , Pralhad Dinesh Deshpande
Abstract: An example operation may include one or more of connecting, by a consortium server, to a blockchain network configured to store digital assets of senders and recipients of funds, receiving from a sender, by the consortium server, a message containing a payment note value and a recipient data, the message is serially encrypted by a plurality of routers, retrieving, by the consortium server, the payment note value from an account of the sender, generating, by the consortium server, a transaction for a transfer of an unspent transaction output of the sender (UTXOs) to an unspent transaction output of the recipient (UTXOr) based on the message, signing, by the consortium server, the transaction with a private key of the consortium server, and submitting, by the consortium server, the transaction to the blockchain.
-
5.发明授权
公开(公告)号：US12020080B2
公开(公告)日：2024-06-25
申请号：US17391546
申请日：2021-08-02
Applicant: International Business Machines Corporation
CPC classification number: G06F9/5072 , G06F9/4451 , G06F9/5011 , G06F9/5077 , G06N20/00
Abstract: Methods, systems, and computer program products for an automated resource request mechanism for heterogeneous infrastructure using profiling information are provided herein. A computer-implemented method includes obtaining resource utilization information, pertaining to multiple system resources, from multiple heterogeneous system infrastructure deployments; automatically learning resource interdependencies for the heterogeneous system infrastructure deployments by processing at least a portion of the resource utilization information using a first set of machine learning techniques; automatically determining performance profiles, with respect to the multiple system resources, for the multiple heterogeneous system infrastructure deployments by processing at least a portion of the resource utilization information and at least a portion of the learned resource interdependencies using a second set of machine learning techniques; predicting resource requests for at least one of the heterogeneous system infrastructure deployments using the determined performance profiles; and performing automated actions based on the resource request predictions.
-
公开(公告)号：US11656927B1
公开(公告)日：2023-05-23
申请号：US17541453
申请日：2021-12-03
Applicant: International Business Machines Corporation
Inventor： Joshua M Rosenkranz , Pranita Sharad Dewan , Mudhakar Srivatsa , Praveen Jayachandran , Chander Govindarajan , Priyanka Prakash Naik , Kavya Govindarajan
IPC: G06F11/07
CPC classification number: G06F11/076 , G06F11/079 , G06F11/0709
Abstract: An ensemble of autoencoder models can be trained using different seeds. The trained ensemble of autoencoder models can be run on new time series data to generate a prediction associated with the new time series data. The new time series data can include multiple dimensions per time step. Reconstruction errors can be determined for the prediction. Dimensions having highest reconstruction errors can be selected among the multiple dimensions based on a threshold. The prediction can be segmented based on bursts of the reconstruction errors over time, where temporal segments can be obtained. At least one common pattern including a set of dimensions among the selected dimensions across the temporal segments can be obtained to represent a failure fingerprint.
-
公开(公告)号：US20230058090A1
公开(公告)日：2023-02-23
申请号：US17405885
申请日：2021-08-18
Applicant: International Business Machines Corporation
Inventor： Priyanka Prakash Naik , Kavya G , Chander Govindarajan , Sayandeep Sen , Palanivel Andiappan Kodeswaran
Abstract: One embodiment provides a method, including: producing, for each of a plurality of containers, a resource profile for each thread in each of the plurality of containers; identifying, for each of the plurality of containers and from, at least in part, the resource profiles, container dependencies between threads on a single of the plurality of containers; determining service dependencies between threads across different of the plurality of containers; scheduling, based upon the container dependencies and the service dependencies, threads to cores, wherein the scheduling is based upon minimizing thread processing times; and publishing the container dependencies and the service dependencies on a registry of the node clusters.
-
公开(公告)号：US20220329653A1
公开(公告)日：2022-10-13
申请号：US17225865
申请日：2021-04-08
Applicant: International Business Machines Corporation
Inventor： Chander Govindarajan , Bishakh Chandra Ghosh , Nitin Gaur , Venkatraman Ramakrishna , Dushyant K. Behl , Petr Novotny
Abstract: An example operation may include one or more of retrieving decentralized identifiers (DIDs) of a plurality of blockchain peers included within a blockchain network, generating a blockchain declarative descriptor (BDD) which uniquely identifies the blockchain network, where the BDD comprises a machine-readable data file with a first field includes the retrieved DIDs of the blockchain network, a second field including signature data of the plurality of blockchain peers, and a third field including metadata, and transmitting the generated BDD to a blockchain network registry.
-
公开(公告)号：US11095705B2
公开(公告)日：2021-08-17
申请号：US16376454
申请日：2019-04-05
Applicant: International Business Machines Corporation
Inventor： Adarsh Saraf , Prabal Banerjee , Shreya Chakraborty , Chander Govindarajan
IPC: H04L29/08
Abstract: An example operation may include one or more of transmitting a request for web page content to a web server, receiving a hypertext markup language (HTML) boilerplate file of the web page with a content delivery network (CDN) address for CDN content of the web page, retrieving the CDN content from a blockchain based on the CDN address, and displaying, at a client device, the web page based on the HTML boilerplate file and the CDN content retrieved from the blockchain.
-
公开(公告)号：US11997022B2
公开(公告)日：2024-05-28
申请号：US17353219
申请日：2021-06-21
Applicant: International Business Machines Corporation
Inventor： Kavya G , Chander Govindarajan , Mudit Verma
IPC: H04L47/78 , G06N20/00 , H04L47/10 , H04L47/722 , H04L47/80
CPC classification number: H04L47/781 , G06N20/00 , H04L47/29 , H04L47/722 , H04L47/805
Abstract: Methods, systems, and computer program products for service-to-service scheduling in container orchestrators are provided herein. A computer-implemented method includes reserving, by a network orchestrator, network resources requested between a plurality of services, wherein each of the services is implemented as one or more replicas running on a set of nodes of a cluster, managed by the network orchestrator, that use the network resources to serve incoming requests to the plurality services; monitoring utilization of the network resources; and scheduling, by the network orchestrator based on the monitoring, one or more new replicas of the plurality of services and the incoming requests to the plurality of services in a collaborative manner to increase at least one network performance characteristic.
