-
公开(公告)号:US10700978B2
公开(公告)日:2020-06-30
申请号:US15368800
申请日:2016-12-05
发明人: Kanak B. Agarwal , Wenzhi Cui , Wesley M. Felter , Yu Gu , Eric J. Rozner
IPC分类号: H04L12/741 , H04L12/803 , H04L12/931 , H04L12/933 , H04L29/08
摘要: A determination is made at an application executing in conjunction with a switch that a request packet sent to a virtualized computing entity (VCE) connected to the switch should be offloaded to a second VCE. A destination of the request packet is changed from the VCE to the second VCE, to form a first modified request packet. An evaluation is made whether a response packet from the second VCE, which is responsive to the request packet, is usable in the data communication with a sender of the request packet. When the evaluating is affirmative, the first modified request packet is caused to be transmitted from the switch to the second VCE.
-
公开(公告)号:US20200159301A1
公开(公告)日:2020-05-21
申请号:US16688110
申请日:2019-11-19
发明人: Malcolm S. Allen-Ware , Kanak B. Agarwal , Charles Lefurgy , Guillermo J. Silva , Thomas W. Keller , Karthick Rajamani , Yang Li , Ramakrishnan Rajamony
IPC分类号: G06F1/26 , G06F1/3228 , G06F1/3209 , G06F1/3296 , G05B15/02 , G06F1/3206 , G06F1/3215 , G06F1/3203
摘要: A computer controls power distribution. The computing system determines a power budget for a portion of a topography for a power delivery system. The computing system generates a pool of worker programs for the portion of the topography. The computing system generates a first number of power management tasks to manage power consumption in the portion of the topography based on the power budget. The computing system sends the first number of power management tasks to at least one worker program included in the pool of worker programs.
-
公开(公告)号:US10574569B2
公开(公告)日:2020-02-25
申请号:US16257284
申请日:2019-01-25
发明人: Kanak B. Agarwal , Eric J. Rozner
IPC分类号: H04L12/24 , H04L12/721 , H04L29/06 , H04L12/26
摘要: At an SDN controller executing using a processor and a memory, a tracing packet is configured with a default value in a tag field. The tracing packet is inserted from the SDN controller into the SDN at a switch in the SDN. A returned packet and a port identifier is received at the controller, from the switch, the returned packet including a modified content in a location of the tracing packet that is different from the tag field. The port identifier corresponds to a port of the switch on which the switch received the returned packet from an middlebox. A function of the middlebox is identified by analyzing a modification applied to the modified content by the middlebox. The function of the middlebox and a location of the middlebox in the SDN are saved. The location includes the port identifier and an identifier of the switch.
-
公开(公告)号:US10362148B2
公开(公告)日:2019-07-23
申请号:US14164422
申请日:2014-01-27
发明人: Kanak B. Agarwal
IPC分类号: H04L29/06
摘要: A plurality of SYN packets is sent from a first system, each of the plurality of SYN packets traverses a different path in a first set of paths between the first system and a second system in a multipath data communication network and includes a unique source identifier. From a plurality of SYN-ACK packets, a SYN-ACK packet is selected that is first to arrive at the first system. Each SYN-ACK packet includes a unique source identifier of a selected SYN packet. A path is configured corresponding to the selected SYN packet, as an outbound path from the first system. A path corresponding to the selected SYN-ACK packet is configured as an inbound path to the first system. A session between the first and the second systems is established using the outbound and inbound paths.
-
公开(公告)号:US20190004579A1
公开(公告)日:2019-01-03
申请号:US15635522
申请日:2017-06-28
发明人: Malcolm S. Allen-Ware , Kanak B. Agarwal , Charles Lefurgy , Guillermo J. Silva , Thomas W. Keller , Karthick Rajamani , Yang Li , Ramakrishnan Rajamony
摘要: A computer controls power distribution. The computing system determines a topography for a power delivery system that powers a group of computing devices. The computing system determines a number of worker programs for a pool of worker programs based on the topography. The computing system generates the pool of worker programs. The pool of worker programs includes both the number of worker programs and a number of back-up worker programs. The computing system generates a number of power management tasks to manage power consumption through one or more power elements included in the topography of the power delivery system. The computing system sends one or more power management tasks to a worker program included in the pool of worker programs.
-
公开(公告)号:US10063623B2
公开(公告)日:2018-08-28
申请号:US15597274
申请日:2017-05-17
发明人: Kanak B. Agarwal , Patrick J. Bohrer , Ahmed Gheith , Michael D. Kistler , Ramakrishnan Rajamony
摘要: To manage channel ownership in a publish-subscribe data processing environment, a determination is made at a first server that a subscription request from a client relates to a channel, the channel having a channel identifier. At an exchange component in a messaging middleware, a session queue associated with the client is bound to the channel. A mandatory delivery message is configured to be routed using a publisher identifier by the exchange component, the publisher identifier being related to the channel identifier. The mandatory delivery message is communicated from the first server to the exchange component to indicate an intent of the first server to become a publisher of the channel. The mandatory delivery message causes the channel ownership of the channel to be limited to a single publisher.
-
公开(公告)号:US10045252B2
公开(公告)日:2018-08-07
申请号:US15171044
申请日:2016-06-02
发明人: Kanak B. Agarwal , John Carter , Wesley M. Felter , Yu Gu , Eric J. Rozner , Keqiang He
IPC分类号: H04W28/02 , H04L12/801 , H04L12/807 , H04L12/931 , H04L12/26
摘要: At an application executing in conjunction with a vSwitch a determination is made that a first flow from a first VM is experiencing congestion. The first flow is selected for throttling. a second flow is also selected for throttling, the second flow using a portion of a network path used by the first flow in a data network. At the application, a total CWND adjustment is distributed between the first flow and the second flow. A first CWND value associated with the first flow is adjusted by a first portion of the total CWND window, and a second CWND value associated with the second flow is adjusted by a second portion of the total CWND window.
-
公开(公告)号:US09998513B2
公开(公告)日:2018-06-12
申请号:US14946981
申请日:2015-11-20
发明人: Kanak B. Agarwal , Patrick J. Bohrer , Ketaki Borkar , Trisha E. Hanlon , Ricardo N. Olivieri , John D. Petitto
IPC分类号: H04L29/06 , H04N21/235 , H04N21/234 , H04N21/647 , H04L29/08 , H04L12/835 , H04N21/232 , H04N21/2343 , H04N21/2662 , H04N21/2668 , H04N21/61 , H04N21/845
CPC分类号: H04N21/2662 , H04L47/30 , H04L65/601 , H04L65/607 , H04L65/608 , H04L65/80 , H04L67/02 , H04L67/06 , H04N21/2323 , H04N21/23406 , H04N21/234345 , H04N21/2353 , H04N21/2668 , H04N21/44004 , H04N21/6125 , H04N21/64738 , H04N21/64769 , H04N21/6543 , H04N21/8456
摘要: A method, system and computer program product for selecting a bitrate to stream encoded media. A manifest file is fetched, where the manifest file contains metadata for encoded media segments which includes tags signifying the associated encoded media segments being important to the viewer. A playback occupancy level is then increased before reaching these tagged media segment(s). A high bitrate (high quality) is selected to stream the tagged media segment(s) which are then fetched and played at the selected high bitrate in response to the new playback buffer occupancy level being above a threshold level. In this manner, those media segments that are of higher importance to the viewer will be viewed in a higher quality since content information is used to select the bitrate to receive the encoded media segments. As a result, the user's quality of experience in viewing media is improved.
-
公开(公告)号:US20180159922A1
公开(公告)日:2018-06-07
申请号:US15368776
申请日:2016-12-05
发明人: Kanak B. Agarwal , Wenzhi Cui , Wesley M. Felter , Yu Gu , Eric J. Rozner
CPC分类号: H04L67/1004 , H04L43/0852 , H04L67/325 , H04L69/163
摘要: A detection is made that a first handshake packet has been received from a data processing system at a first system. The first system participates in a load-balanced group managed by a load-balancer. A value is obtained of a metric configured in the first system. from a set of delay functions, a delay function that corresponds to the metric is selected. Using the value of the metric in the selected delay function, a delay period is computed. A transmission of a second handshake packet is delayed for at least the delay period. An intentionally delayed transmission of the second handshake packet is caused after the delay period from the first system to the data processing system.
-
公开(公告)号:US20180157539A1
公开(公告)日:2018-06-07
申请号:US15368763
申请日:2016-12-05
发明人: Kanak B. Agarwal , Wenzhi Cui , Wesley M. Felter , Yu Gu , Eric J. Rozner
IPC分类号: G06F9/50
摘要: A type of a request that is currently being processed at a system is determined. A distribution is selected from a set of processing time distributions, the distribution forming a model that is applicable to the type. A threshold point is computed for the model. A processing time that exceeds a threshold point processing time is regarded as exhibiting tail latency. Tail latency includes a delay in processing of the request due to a reason other than a utilization of a resource of the system exceeding a threshold utilization and a size of a queue in the system exceeding a threshold size. An evaluation is made that the request will experience tail latency during processing at the system and the processing of the request at the system is aborted. The request is offloaded for processing at a peer system in a load-balanced group of systems.
-
-
-
-
-
-
-
-
-