Patents by Inventor Mukundan Madhavan
Mukundan Madhavan has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11533238Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: GrantFiled: June 11, 2021Date of Patent: December 20, 2022Assignee: Goldman Sachs & Co. LLCInventor: Mukundan Madhavan
-
Publication number: 20210314231Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: ApplicationFiled: June 11, 2021Publication date: October 7, 2021Inventor: Mukundan Madhavan
-
Patent number: 11063832Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: GrantFiled: May 15, 2020Date of Patent: July 13, 2021Assignee: Goldman Sachs & Co. LLCInventor: Mukundan Madhavan
-
Publication number: 20200313971Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: ApplicationFiled: May 15, 2020Publication date: October 1, 2020Inventor: Mukundan Madhavan
-
Publication number: 20200235998Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: ApplicationFiled: March 7, 2019Publication date: July 23, 2020Inventor: Mukundan Madhavan
-
Patent number: 10715393Abstract: An online system receives time series data and analyzes the data for identifying trend changes or level shifts in the time series. The time series data may describe resource utilization of systems, for example, bandwidth of computer networks. The online system uses prediction models, for example, machine learning based prediction models using regression to predict data values for the time series. The online system determines error residue values based on difference between predicted data values and actual data values of the time series. The online system determines level change in the error residue to identify change-points representing trend changes or level-shifts in the original time-series. The online system takes corrective action based on the change-point information and the trend following the occurrence of the change-point, for example, sending alerts or instructions for causing reconfiguration of the systems such as the computer network.Type: GrantFiled: March 7, 2019Date of Patent: July 14, 2020Assignee: Goldman Sachs & Co. LLCInventor: Mukundan Madhavan
-
Patent number: 10057739Abstract: Methods, systems, and computer program products for distributed and localized policy and charging control in cellular networks are provided herein.Type: GrantFiled: October 21, 2015Date of Patent: August 21, 2018Assignee: International Business Machines CorporationInventors: Ravi Kothari, Dilip Krishnaswamy, Mukundan Madhavan, Vijay Purushottam Gabale
-
Publication number: 20170118616Abstract: Methods, systems, and computer program products for distributed and localized policy and charging control in cellular networks are provided herein.Type: ApplicationFiled: October 21, 2015Publication date: April 27, 2017Inventors: Ravi Kothari, Dilip Krishnaswamy, Mukundan Madhavan, Vijay Purushottam Gabale
-
Patent number: 9565116Abstract: Methods and arrangements for managing variable-priority flows in a network. Requests for transmission of digital content are received from client devices, each request being associated with a priority, the requests including at least one lower-priority delayable request and at least one higher-priority request. A usable bandwidth for delivery of the requested digital content through the network is estimated and, within the estimated usable bandwidth, there is controlled at least one flow associated with the at least one lower-priority request in the presence of at least one flow for the at least one higher-priority request. Other variants and embodiments are broadly contemplated herein.Type: GrantFiled: August 29, 2013Date of Patent: February 7, 2017Assignee: GLOBALFOUNDRIES INC.Inventors: Umamaheswari C. Devi, Ravindranath Kokku, Vinay Kumar Kolar, Mukundan Madhavan
-
Patent number: 9471460Abstract: Methods and arrangements for characterizing software base-station workloads. Input system parameters are mapped to work-determining parameters which act to determine computational requirements of a dynamic workload. A synthetic experiment is undertaken to measure the computational requirements determined by the work-determining parameters.Type: GrantFiled: March 29, 2012Date of Patent: October 18, 2016Assignee: International Business Machines CorporationInventors: Malolan Chetlur, Parul Gupta, Mukundan Madhavan
-
Patent number: 9247559Abstract: Methods, systems, and articles of manufacture for dynamically delayed delivery of content in a network are provided herein. A method includes receiving, from a user device coupled to a network, a request for a content from the network; estimating a measure of available bandwidth on the network for delivery of the requested content at each of multiple periods of time; generating multiple schedules for the delivery of the requested content, wherein each of the multiple schedules is based on (i) the measure of available bandwidth on the network at one of the multiple periods of time and (ii) a predefined set of scheduling rules associated with the user device and the network; and delivering the requested content to the user device according to a schedule selected from the multiple schedules.Type: GrantFiled: January 31, 2014Date of Patent: January 26, 2016Assignee: International Business Machines CorporationInventors: Umamaheswari Devi, Ravindranath Kokku, Shivkumar Kalyanaraman, Mukundan Madhavan
-
Publication number: 20150223260Abstract: Methods, systems, and articles of manufacture for dynamically delayed delivery of content in a network are provided herein. A method includes receiving, from a user device coupled to a network, a request for a content from the network; estimating a measure of available bandwidth on the network for delivery of the requested content at each of multiple periods of time; generating multiple schedules for the delivery of the requested content, wherein each of the multiple schedules is based on (i) the measure of available bandwidth on the network at one of the multiple periods of time and (ii) a predefined set of scheduling rules associated with the user device and the network; and delivering the requested content to the user device according to a schedule selected from the multiple schedules.Type: ApplicationFiled: January 31, 2014Publication date: August 6, 2015Applicant: International Business Machines CorporationInventors: Umamaheswari Devi, Ravindranath Kokku, Shivkumar Kalyanaraman, Mukundan Madhavan
-
Patent number: 9026077Abstract: Methods and arrangements for undertaking delayed delivery of digital content. At least one request for transmission of digital content is received from a client device. There is estimated a usable bandwidth for delivery of the digital content during a predetermined time period. There is offered to the client device two or more delivery schedules for delayed digital content delivery during the predetermined time period. Other variants and embodiments are broadly contemplated herein.Type: GrantFiled: March 30, 2013Date of Patent: May 5, 2015Assignee: International Business Machines CorporationInventors: Umamaheswari Devi, Shivkumar Kalyanaraman, Ravindranath Kokku, Mukundan Madhavan
-
Patent number: 9008710Abstract: Methods, apparatuses, and computer program products for controlling communication between whitespace devices are provided.Type: GrantFiled: June 14, 2012Date of Patent: April 14, 2015Assignee: International Business Machines CorporationInventors: Malolan Chetlur, Harish Ganapathy, Shivkumar Kalyanaraman, Mukundan Madhavan
-
Publication number: 20150063106Abstract: Methods and arrangements for managing variable-priority flows in a network. Requests for transmission of digital content are received from client devices, each request being associated with a priority, the requests including at least one lower-priority delayable request and at least one higher-priority request. A usable bandwidth for delivery of the requested digital content through the network is estimated and, within the estimated usable bandwidth, there is controlled at least one flow associated with the at least one lower-priority request in the presence of at least one flow for the at least one higher-priority request. Other variants and embodiments are broadly contemplated herein.Type: ApplicationFiled: August 29, 2013Publication date: March 5, 2015Applicant: International Business Machines CorporationInventors: Umamaheswari C. Devi, Ravindranath Kokku, Vinay Kumar Kolar, Mukundan Madhavan
-
Publication number: 20140295789Abstract: Methods and arrangements for undertaking delayed delivery of digital content. At least one request for transmission of digital content is received from a client device. There is estimated a usable bandwidth for delivery of the digital content during a predetermined time period. There is offered to the client device two or more delivery schedules for delayed digital content delivery during the predetermined time period. Other variants and embodiments are broadly contemplated herein.Type: ApplicationFiled: March 30, 2013Publication date: October 2, 2014Applicant: International Business Machines CorporationInventors: Umamaheswari Devi, Shivkumar Kalyanaraman, Ravindranath Kokku, Mukundan Madhavan
-
Publication number: 20130263135Abstract: Methods and arrangements for characterizing software base-station workloads. Input system parameters are mapped to work-determining parameters which act to determine computational requirements of a dynamic workload. A synthetic experiment is undertaken to measure the computational requirements determined by the work-determining parameters.Type: ApplicationFiled: March 29, 2012Publication date: October 3, 2013Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Malolan Chetlur, Parul Gupta, Mukundan Madhavan
-
Publication number: 20130172033Abstract: Methods, apparatuses, and computer program products for controlling communication between whitespace devices are provided.Type: ApplicationFiled: June 14, 2012Publication date: July 4, 2013Applicant: International Business Machines CorporationInventors: Malolan Chetlur, Harish Ganapathy, Shivkumar Kalyanaraman, Mukundan Madhavan
-
Publication number: 20130172032Abstract: Methods, apparatuses, and computer program products for controlling communication between whitespace devices are provided.Type: ApplicationFiled: December 29, 2011Publication date: July 4, 2013Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Malolan Chetlur, Harish Ganapathy, Shivkumar Kalyanaraman, Mukundan Madhavan
-
Publication number: 20130046703Abstract: A method, an apparatus and an article of manufacture for communicating power consumption information. The method includes comparing power consumption information of at least one user with power consumption information of at least one peer user to generate a ranking of each user in order of power consumption, monitoring the user power consumption derived from non-smart appliance usage and to generate a recommendation pertaining to at least one smart appliance for the at least one user, and publishing the generated ranking and recommendation information to a network for communication with the at least one user.Type: ApplicationFiled: August 19, 2011Publication date: February 21, 2013Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Ashok Ambati, Kunal K. Korgaonkar, Mukundan Madhavan, Ramana V. Polavarapu