Cisco Technology, Inc.

Applied Materials, Inc.

Intel Corporation

Pensando Systems Inc.

Leland Stanford Junior University

Ent. Services Development Corporation LP

Rong Pan

Flavio Giovanni Bonomi

Xiaoqing Zhu

Preethi Natarajan

Hiroshi Suzuki

Thomas James Edsall

Sateesh Kumar Addepalli

Lillian Lei Dai

Chiara Piglione

Changhoon Kim

Zhi Li

Joshua B Gahm

Valentina Alaria

Vijaynarayanan Subramanian

Hariprasada Rao Ginjpalli

Parvin Taheri

Norman William Finn

Jeongkeun Lee

Robert G Southworth

Peter Geoffrey Jones

Balaji Satyanarayana Prabhakar

Davide Bergamasco

Rudolph Benedict Klecka

Roberto Penaranda Cebrian

Linda Tin-Wen Cheng

Malek Musleh

Andrea Baldini

Erico Vanini

Pedro Yebenes Segura

Fabio Rodolfo Maino

Mohammadreza Alizadeh Attar

Peter Clarke Newman

Won B Bang

Pere Monclus

Vina Ermagan

Ashok Krishnaji Moghe

Frederick Juergens Baker

Kirby Hane Floyd

Hao Hu

Alexander Loukissas

Mythili Suryanarayana Prabhu

Adrian Q Montgomery

Jennifer Gonzales

Amna Mohammed

Pascal Thubert

Nitin K Ingle

Jun Wang

Shankar Venkataraman

David R Oran

Karthik Janakiraman

Zheng Yuan

Anurag Agrawal

Alon Shlomo Bernstein

Arulkumar P Shanmugasundram

Alexander T Schwarm

Raghuram S Sudhaakar

Chen-An Chen

Chiu Ngok Eric Wong

Shang-Tse Chuang

Ethan M Spiegel

Ming Zhang

Gary S Muntz

Yen-Kun Victor Wang

Douglas Chan

Vikash Banthia

Peter Wai-Man Lee

Sha Ma

Dipankar Bhatt Acharya

George Varghese

Guglielmo Marco Morandin

Krishnan Subramani

Georges Akis

Paul Edward Gee

Francimar Campana

Ashutosh Agrawal

Chien Fang

Ravi Prasad

Sharad Vasantrao Chole

Manuel Hernandez

Tom Edsall

Xinyun Xia

Dong Han

Cary Ching

Felice Bonardi

Robert Edward Somers

Allister Alemania

Aimin Ding

Andrew Michael Robbins

Marc Jean-Philippe Girardot

Padma Akkiraju

Hali J L Forstner

Ishing Lou

Kelly Fong

Paul Le

Junggun Lee

Vijaymarayanan Subramanian

Stanley Arthur Trimble

Roberto Antonio Contreras Masse

Rajesh Dontula

Guo-Xiang Qin

Examples described herein relate to a network interface device that is to adjust a transmission rate of packets based on a number of flows contributing to congestion and/or based on whether latency is increasing or decreasing. In some examples, adjusting the transmission rate of packets based on a number of flows contributing to congestion comprises adjust an additive increase (AI) parameter based on the number of flows contributing to congestion. In some examples, latency is based on a measured roundtrip time and a baseline roundtrip time.

Congestion control

Examples described herein relate to a network agent, when operational, to: receive a packet, determine transmit rate-related information for a sender network device based at least on operational and telemetry information accumulated in the received packet, and transmit the transmit rate-related information to the sender network device. In some examples, the network agent includes a network device coupled to a server, a server, or a network device. In some examples, the operational and telemetry information comprises: telemetry information generated by at least one network device in a path from the sender network device to the network agent.

Receiver-based precision congestion control

Examples described herein relate to a network interface device comprising dataplane circuitry, when operational, is to generate a representation of aggregated network resource consumption information based on network resource consumption at the network interface device or at least one other network device and to transmit at least one packet with a multi-bit representation of the aggregated network resource consumption information to a second network interface device. In some examples, the network resource consumption information comprises one or more of: available transmit bandwidth, transmit bandwidth used by a queue or flow, queue depth, measured queueing time duration, expected queueing time duration, packet latency, or normalized in-flight bytes.

Resource consumption control

Examples described herein relate to a network element comprising an ingress pipeline and at least one queue from which to egress packets. The network element can receive a packet and generate a congestion notification packet at the ingress pipeline to a sender of the packet based on detection of congestion in a target queue that is to store the packet and before the packet is stored in a congested target queue. The network element can generate a congestion notification packet based on a queue depth of the target queue and likelihood the target queue is congested. The likelihood the queue is congested can be based on a probabilistic function including one or more of Proportional-Integral (PI) or Random Early Detection (RED). The network element can determine a pause time for the sender to pause sending particular packets based at least on a time for the target queue to drain to a target level.

Congestion management techniques

This disclosure provides methods and systems for reducing congestion in RoCEv2 networks. The method is configured to operate large-scale in data centers on traffic flowing from a sender node to a receiver node. The method described has three stages: a fast start stage, a transition stage, and a regulation stage. In the fast start stage, the sender sends data to the receiver at a fast initial rate. This may continue until the receiver observes a congestion event. When this happens, the sender reduces the data transfer rate as the method enters the transition stage. From a reduced rate, the method enters the regulation stage, where the rate is increased using a combination of a feedback control loop and an additive increase multiplicative decrease (AIMD) algorithm.

Rate-optimized congestion management

A network device, including ports that receive/send data packets from/to a network, receives data packets of multiple traffic flows, and populates a queue in memory with the data packets. The network device periodically updates a fair rate for the multiple traffic flows to converge a length of the queue to a reference length. Specifically, the network device determines a length of the queue, a change in the length from a previous length, and a deviation of the length from the reference length. The network device detects an increase in the change in length above a threshold that is based on the reference length. If the increase is not above the threshold, the network device derives the fair rate from a previous fair rate using proportional integral control. The network device identifies elephant flows among the multiple traffic flows, and sends the fair rate to a source of each elephant flow.

Layer 3 fair rate congestion control notification

A method in one embodiment includes intercepting a message in an on-board unit (OBU) of a vehicular network environment between a source and a receiver in the vehicular network environment, verifying the message is sent from the source, verifying the message is not altered, evaluating a set of source flow control policies associated with the source, and blocking the message if the set of source flow control policies indicate the message is not permitted. In specific embodiments, the message is not permitted if a level of access assigned to the source in the set of source flow control policies does not match a level of access tagged on the message. In further embodiments, the method includes evaluating a set of receiver flow control policies associated with the receiver, and blocking the message if the set of receiver flow control policies indicates the message is not permitted.

System and method for wireless interface selection and for communication and access control of subsystems, devices, and data in a vehicular environment

Monitoring an object to prevent an occurrence of an issue includes monitoring an object based on a number of parameters, categorizing messages of the object into categories while monitoring for a recurrence of a pattern of messages, detecting a potential issue with the object based on the recurrence of the pattern of messages, and alerting an operations manager agent of the potential issue to prevent an occurrence of the issue.

Monitoring an object to prevent an occurrence of an issue

A network device is configured to transmit acknowledgement packets according to the length of the egress queue. The network device receives data packets from one or more endpoints and buffers the data packets in an egress buffer before transmitting the data packets. The network device also receives acknowledgement packets that are sent in response to data packets previously transmitted by the network device. The network device buffers the acknowledgement packets in an acknowledgement buffer. The network device transmits the acknowledgement packets at an acknowledgment rate that is based on a queue length of the egress buffer.

Adaptive acknowledgement control

In one embodiment, a method includes measuring a rate of traffic received at a leaf node, marking a plurality of packets in the flow as protected at the leaf node to match the rate of traffic with a configured rate of traffic for the flow at the leaf node, and dropping a plurality of non-protected packets at the leaf node when a queue at the leaf node is congested. A minimum bandwidth is provided for the flow based on the configured rate of traffic at the leaf node. The leaf node comprises an ingress node or an egress node connected to a fabric. An apparatus is also disclosed herein.

Fabric-wide bandth management

A first flowlet of a flow from a source network device to a destination network device is assigned to a first path of a plurality of paths between the source device and the destination device. The assignment of the first flowlet to the first path is made by a network connected device. A second flowlet is detected in response to an interruption in transmission of the flow due to congestion along the first path, wherein the interruption is longer in duration than a difference in a transmission time between the source network device and the destination network device along each of the plurality of paths. The second flowlet is assigned to a second path of the plurality of paths by the network connected device. According to some example embodiments, the second path is randomly selected from the plurality of paths.

Load balancing with flowlet granularity

In one embodiment, an HTTP streaming session may be initiated at a client device in a network. The client device may have a buffer and may be configured to request and receive one or more data segments over HTTP from an HTTP server. A first data segment at a first data source rate may be requested and subsequently received. The first data segment may be stored in the buffer. A second data source rate may then be calculated based on a storage level in the buffer, and a second data segment at the second data source rate may be requested.

HTTP streaming client adaptation algorithm based on proportional-integral control

Growing community of inventors

Saratoga, CA, United States of America

Rong Pan