72 results about "Explicit Congestion Notification" patented technology

An Explicit Congestion Notification (ECN) method is disclosed for wireless applications to avoid network congestion in a TCP / IP packet-switched network. Such method comprises transmitting, at a source node, data packets to a destination node, via at least an intermediate node; determining, at the intermediate node, if an incipient congestion is encountered, setting a Congestion Experienced (CE) flag in each data packet to notify congestion; sending, at the destination node, an ECN-Echo acknowledgment packet back to the source node to inform congestion; reducing, at the source node, a congestion window and a transmission rate to avoid congestion; if the packet loss is due to congestion, re-transmitting, at the source node, only a lost packet to the destination node; alternatively, if the packet loss is due to transmission error, re-transmitting, the lost packet to the destination node, while increasing a round-trip timeout but maintaining the same congestion window.

A system provides congestion control and includes multiple queues that temporarily store data and a drop engine. The system associates a value with each of the queues, where each of the values relates to an amount of memory associated with the queue. The drop engine compares the value associated with a particular one of the queues to one or more programmable thresholds and selectively performs explicit congestion notification or packet dropping on data in the particular queue based on a result of the comparison.

Embodiments of the invention provide a method for quantized congestion notification in a virtual networking system comprising multiple virtual networks (VNs). Each VN comprises at least one virtual machine (VM) configured to generate one or more packet flows. Each packet of each packet flow is tagged with a congestion notification (CN) tag. Each packet flow is mapped to a corresponding virtual tunnel end point (TEP) that distributes each packet of the packet flow. A congestion notification message (CNM) is generated for each congestion point (CP) associated with each packet flow. Each CP is mapped to a corresponding TEP that distributes each CNM for the CP, wherein the corresponding VTEP forwards the CNM to a VM contributing to packet congestion at the CP.

A method and system are provided for automatically rerouting data from an overbalanced logical circuit in a data network. Status information is received for a logical circuit communicating data in the data network. The status information may include a forward explicit congestion notification (“FECN”) and a backward explicit congestion notification (“BECN”) indicating an overbalanced condition in the logical circuit. An overbalanced condition is indicative of network congestion in the logical circuit. A logical failover circuit including an alternate communication path for communicating the data in the overbalanced logical circuit is then identified. The data in the overbalanced logical circuit is rerouted to the logical failover circuit without manual intervention. After rerouting the data to the logical failover circuit, a determination is made as to whether the overbalanced condition in the logical circuit has been cleared. If it is determined that the overbalanced condition in the logical circuit has been cleared, then the data from the logical failover circuit is rerouted to the logical circuit in the data network without manual intervention.

Methods and systems for processing network data are disclosed herein and may include receiving from a switching device, a congestion indicator that indicates congestion. In response to the congestion indicator, latency of reaction by a source end point, may be reduced by preventing introduction of queued up new frames to affected flow or CoS before the local stack adjusts its rate to congestion conditions and / or by rate limiting the processing of unprocessed network frames in hardware. The unprocessed network frames may include unprocessed network frames of a particular type. In response to the received congestion indicator, by a destination end point, congestion indicator flags may be set in processed network frames of the particular type, faster than an expected reaction of a local stack. The congestion indicator flags may be explicit congestion notification (ECN)-Echo flags.

One embodiment of the present invention provides a congestion notification. During operation, the system receives a congestion notification message with a destination layer-2 address set as the receiving mechanism's layer-2 address. The system modifies a destination layer-2 address, a source layer-2 address, and optionally a VLAN identifier of the congestion notification message, and forwards the modified congestion notification message.

Systems and methods described herein are directed to solutions for NoC interconnects that provide congestion avoidance and end-to-end uniform and weighted-fair allocation of resource bandwidths among various contenders in a mesh or torus interconnect. The example implementations are fully distributed and involve using explicit congestion notification messages or local congestion identification for congestion detection. Based on the congestion level detected, the injection rates of traffic at various agents are regulated that avoids congestion and also provides end-to-end QoS. Alternative example implementations may also utilize end-to-end credit based flow control between communicating agents for resource and bandwidth allocation of the destination between the contending sources. The resource allocation is performed so that both the weighted and strict bandwidth allocation QoS policies are satisfied.

The embodiment of the invention discloses a speed adjusting method, system and coding scheme selection method and a device thereof. The speed adjusting method includes: a downlink data packet sent by transmitting equipment is received, wherein the explicit congestion notification ECN mark the down link data packet is an explicit congestion notification capacity transmission ECT mark or a non-ECT mark; when congestion happens, the ECN mark of the downlink data packet is set to be a congestion mark, and the downlink data packet carrying the congestion mark is sent to receiving equipment, so as to be convenient for the receiving equipment to control downlink speed according to the congestion mark. The embodiment of the invention realizes that ECN can be utilized to control congestion even if transmitting equipment or a router on the route between transmitting equipment and receiving equipment does not support ECN, so that ECN scheme is stronger; and coding scheme can be selected according to network load condition, thus improving network capacity.

An apparatus and a smethod providing fair bandwidth sharing by adjusting video image quality in a data packet network comprises a network load detection means for detecting a network load and a video encoding control circuit. The network load has an uncongested state, a loaded state, and a congested state. The video encoding control circuit adjusts a video quality to a target video quality, by increasing the video quality when the network load is in the uncongested state and decreasing the video quality when the network load is in the congested state. The video quality is determined as a peak mean squared error between an uncompressed image and a corresponding decoded image. The network load is detected by using a forward explicit congestion notification bit.

The invention discloses an explicit congestion control method based on bandwidth estimation in a high-bandwidth delay network, which comprises the following steps of: calculating the current available bandwidth of a network by a TCP (Transmission Control Protocol) sender through observing an ACK (Acknowledgement Character) receiving interval, loading a router by a packet header ECN (Explicit Congestion Notification) 2bits information feedback network, carrying out network state judgment by the sender through explicit feedback information and estimated available bandwidth values, and adjusting a congestion mechanism according to a judgment result. The congestion is judged by combining end-to-end feedback and router explicit feedback, i.e. finer congestion information is given by using the bandwidth estimation values based on the explicit feedback of coarseness, thus an algorithm ensures high congestion feedback precision while reducing the explicit feedback overhead and improving the protocol popularization, and improves the efficiency and the convergence rate of the protocol.

The invention provides a buffer queue management method used in a wireless local area network. In the wireless local area network, when related sites have large quantity and large data amount, network congestion phenomenon can happen. In order to detect the network congestion before the congestion happens or at the beginning period that the congestion happens, the in-queue detection of a data packet is realized by the buffer queue management method of exchanging nodes (such as AP), and a sending end is notified with the congestion by packet loss or setting an explicit congestion notification (ECN) of an IP packet head, so as to relieve the network congestion. The buffer queue management method provided by the invention uses the queue length and flux speed as congestion measurement, so that the invention can evaluate the network congestion degree more exactly, and the invention calculates different packet loss probability for each queue according to the priority difference of buffer queues to realize the random packet loss according to different priorities, so as to achieve the goal of distinguishing packet loss rate.

The invention provides a TCP (transmission control protocol) congestion control method based on throughout change rate and an ECN (Explicit Congestion Notification) mechanism. The method uses an ECN mechanism to prevent congestion, and comprises the following steps of when no ECN marks exist, if a packet is lost, judging the reason of packet loss according to the change rate of throughout, further adjusting a transmitting velocity instead of halving blindly, thereby avoiding severe performance problems caused by wireless random packet loss; and meanwhile, when no packet is lost, adjusting the increasing of windows according to the change rate of the throughout, further preventing congestion. By combining the ECN technology and the control method based on the throughout change rate, ECN can perform controlling in mild congestion of a network, thereby avoiding TCP Incast phenomenon in a wired data center; the control method based on the throughout change rate can be used for adjusting the sending windows according to the throughout change rate of the current network, when the packet is lost, the reason of the packet loss can be judged, so that the halving of the sending windows caused by wireless random packet loss can be avoided.

The present invention relates to a differential flow control method and device for a cloud computing data center network, which belongs to the field of cloud computing. The method comprises the following steps that 1) a switch uses an explicit congestion notification (ECN) mechanism to mark packets exceeding the set threshold queue length; 2) a sender server counts the proportion of marked ACK acknowledgment packets, updates a parameter [alpha], and determines whether congestion occurs, wherein the parameter [alpha] indicates the perception of the degree of congestion on the network; 2.1) when the network is not congested, the TCP flow does not carry out congestion window adjustment; 2.2 ) the sender server performs a flow classification decision, establishes a flow classification database, and records an update parameter [beta], wherein the parameter [beta] indicates the degree of reuse of a receiving end link of a large flow; and 3) the sender server executes different congestion back-off strategies for TCP flows of different transmission requirements distinguished by a flow classification decision mechanism. The advantages are that 1, TCP Incast congestion can be effectively avoided; 2, the time to complete the transmission of small flows can be shortened; and 3, the rapid and fair convergence of congestion control can be ensured.

The invention provides a TCP (transmission control protocol) congestion control method based on network effective bandwidth and an ECN (Explicit Congestion Notification) mechanism. The method comprises the following steps of step 1, performing ECN marking on a switch; step 2, setting ECN-echo by a receiver; step 3, dynamically adjusting a sending window by a transmitter, when a transmitting end receives an ACK (acknowledgement character), firstly calculating effective bandwidth BWE and minimum round-trip time delay RTTmin of a current network by a BWE calculation module; and then looking over whether the ACK sets an ECN-echo bit by an ECN control module, if so, setting ecn_flag showing that the sending velocity is needed to be adjusted according to the ECN, and calculating the ratio alpha of the received ACK with the ECN-echo marks to the total number of data packets sent by the previous sending window. By adopting the ECN technology and the control method based on the network effective bandwidth, the ECN can perform controlling in mild congestion of the network, thereby avoiding TCP Incast phenomenon in a wired data center; the control method based on BWE can be used for adjusting the sending window according to the effective bandwidth of the current network when the packet is lost, so that the halving of the sending windows caused by wireless random packet loss can be avoided.

The invention discloses a method for controlling congestion by fusing three kinds of information in a wired / wireless hybrid network. The three kinds of information include packet loss information, RTT (Radiation Tracking Transducer) information and explicit congestion notification information. The method comprises the following steps of: marking a CE (Conformite Europeenne) bit in IP (International Protocol) packet according to the RTS (Ready-To-Send) repeat times of an MAC (Media Access Control) layer by a wireless node; marking the CE bit in the IP packet by a wired router through an array length; processing a received data packet, confirming the carrying capability through an ACK (Acknowledgement) packet and informing a sending end of the network congestion state by a receiving end; and then, comprehensively considering the packet loss information, the RTT information and the explicit congestion notification information for congestion prediction and adjusting a congestion control mechanism according to a prediction result by the sending end. The method provided by the invention judges the congestion by fusing the three kinds of information, effectively distinguishes wireless packet loss and congestion packet loss, improves the utilization and the throughput of the wired and wireless hybrid network with a wireless link as a bottleneck and improves the whole performance of the network.

Systems and methods described herein are directed to solutions for NoC interconnects that provide congestion avoidance and end-to-end uniform and weighted-fair allocation of resource bandwidths among various contenders in a mesh or torus interconnect. The example implementations are fully distributed and involve using explicit congestion notification messages or local congestion identification for congestion detection. Based on the congestion level detected, the injection rates of traffic at various agents are regulated that avoids congestion and also provides end-to-end QoS. Alternative example implementations may also utilize end-to-end credit based flow control between communicating agents for resource and bandwidth allocation of the destination between the contending sources. The resource allocation is performed so that both the weighted and strict bandwidth allocation QoS policies are satisfied.

The invention discloses a marking method based on explicit congestion notification (ECN) in a multi-queue data center environment, and mainly solves the problem that a traditional ECN cannot sense micro burst flow and cannot adapt to a dynamic network. The method mainly comprises three parts, namely ECN threshold lower bound calculation, threshold base value setting and threshold dynamic adjustment. According to the method, firstly, steady-state analysis is carried out on the generation and transmission process of the micro burst flow in a multi-queue data center environment, the lower bound of an ECN threshold value is solved for each switching queue, and it is guaranteed that a switch queue buffer area can effectively absorb the micro burst flow. Then, based on an ideal Generic Packet Schedule (GPS) model, an ECN threshold lower bound is adjusted, a threshold base value is set for each queue, it is guaranteed that the enqueue rate and the dequeue rate of exchange queues are balanced,finally, an optimization model for queue buffer maximum utilization is established for switch memory resources on the basis of the target of improving the network throughput, and an ECN threshold dynamic adjustment algorithm is provided.

Satellite communications are carried out using the Data Over Cable Interface Specification (DOCSIS). Satellite modems are notified of upstream channel congestion by inserting a congestion notification message in a medium access protocol (MAP) message for the upstream channel. Specifically, the congestion notification message is inserted in an unused field of the MAP message, such as the explicit congestion notification (ECN) field. The MAP message can also carry other characteristics of the upstream channel, including priority threshold, multicast access burst availability, and available bandwidth.

The invention provides a policy and charging control (PCC) method and a system based on the network load. The method comprises the following steps: when a policy and charging rule function (PCRF) makes a policy decision, according to whether a wireless base station accessed currently to a user equipment (UE) supports an explicit congestion notification (ECN) and the UE and the correspondent node of the UE support the ECN, the Maximum BitRate (MBR) in the Quality of Service (QoS) information of the policy is set to be greater than a Guarantee BitRate (GBR), or the MBR in the QoS information of the policy is set to be equal to the GBR; the PCRF sends the policy to a policy implementation function, and the policy implementation function implements the policy. The invention also provides a PCC system based on the network load.

The embodiment of the invention provides a method, device and system for multi-path TCP congestion control, and relates to the field of communication. The method, device and system are used for carrying out congestion control before congestion occurs and improve efficiency of data transmission. The method comprises the steps that a data packet sent by a network node of a first access network in at least one multi-path TCP substream constructed in the first access network by a sending end is received by a receiving end; if it is detected that the data packet supports explicit congestion, a transmission ECT zone bit of an ECN is identified by a first identifier; a CE zone bit is identified by a second identifier, and congestion control is carried out on the receiving end; an ECN feedback ECE zoon bit of the data packet is identified by a third identifier; a CWR zoon bit is identified by a fourth identifier. The method, device and system are suitable for congestion control of networks.

The invention provides a message forwarding control method and a device. The method comprises: when detecting that the length of a target queue of a target outgoing port exceeds a first preset threshold, explicit congestion notification ECN processing is carried out on the message of the target queue; after the ECN processing, if the length of the target queue of the target outgoing port exceeds afirst preset threshold, explicit congestion notification ECN processing is carried out. When the ECN and the PFC are simultaneously enabled, the embodiment of the invention ensures that the ECN is triggered first and the utilization rate of the bandwidth is ensured while the flow rate is controlled. At the same time, packet loss can be reduced by PFC.

The invention provides a method for achieving explicit congestion notification, which is applied to a provider edge (PE) device which receives message containing an internet protocol (IP) head and transmitted by a user edge device, and the IP head contains marks supporting explicit congestion notification (ECN). When the device is confirmed to be in a congestion state, the received message is packaged to be multiple protocol label switching (MPLS) message, and an ECN label with a network congestion mark is arranged at the bottom of a label stack of the MPLS message. The invention further provides a method for achieving explicit congestion notification, which is applied to a provider (P) device which receives the MPLS message transmitted by the PE device, and when the device is in a congestion state and the ECN label at the bottom of the label stack of the MPLS message is provided with a network non-congestion label, the ECN label at the bottom of the label stack of the received MPLS message is set to be a network congestion mark and is conducted with public network label exchange processing. The invention further provides the PE device and the P device. The method and the devices can effectively relieve network device congestion in the MPLS network.

The invention discloses a selective explicit congestion notification marking method (SECN for short) for a data center; the SECN implementation is divided into two parts: 1, a sending end writes estimated flow rate into the head of a data packet; 2, a switch judges the state of the flow speed rate and the like by reading the rate value of the head of the data packet, and the switch marks the flowwith high sending speed rate. The method, namely the SECN, in the invention is a first ECN marking method for distinguishing the flow state, and the method is low-load. According to the method, the link throughput can be obviously improved, the flow finishing time can be shortened, and the throughput of the multi-bottleneck flow can be improved.

The invention discloses a transmission control method and system for managing a reverse queue, and the method comprises the steps: configuring a switch reverse active queue management framework at a switch node; In the data transmission process, the switch identifies the control packets, and stores the control packets into an input queue of a switch reverse active queue management framework in sequence according to a first-in first-out strategy; The switch reverse active queue management framework determines a data transmission state according to an output queue, a threshold value and a congestion degree parameter value of the switch reverse active queue management framework; The method also comprises the following steps: determining different dequeuing rules of a control packet accordingto a data transmission state, setting an explicit congestion notification mark for the control packet at a switch when notifying a sending end to perform congestion control, and receiving the controlpacket with the explicit congestion notification mark by the sending end so as to control the increment of a sending window. The method can improve the congestion during data transmission, improve theprocessing efficiency and improve the network throughput.

A system and method are provided for implementing performance improvements in a multicast protocol for networks that support incipient congestion indications via packet marking in instances of packet loss in the network during time-correlated blockages by providing indications in the physical (PHY) layer. A receiver rate calculation is adjusted so that a loss due to blockage is ignored completely and only packets marked using an Explicit Congestion Notification (ECN) packet marking protocol are treated as losses. Receiver rates are modified based on ECN principles to ignore losses. A NORM receiver rate equation may remain substantially unchanged while a sender is enabled to keep sending at a higher data rate, even in instances of blockage in support of higher system throughputs without defining a completely new receiver rate equation. Time-correlated blockages are not treated as losses.

Techniques for sending control information in the header of multiple packets are described. The techniques may allow more control information to be sent using a small number of overhead bits per packet. In one design, a first node (e.g., a network entity) may determine control information to send to a second node (e.g., a UE or another network entity). The first node may send the control information in the header of multiple packets toward the second node. In one design, the control information may include congestion information indicative of traffic congestion at the first node. The congestion information may be sent using Explicit Congestion Notification (ECN) bits in the header of IP packets. The first node may send the control information with or without coding and for all packets or a specific data flow. The first node may also send a synchronization sequence prior to the control information.

The invention provides a method for achieving an explicit congestion notification. The method is applied to provider edge: provider edge (PE: PE) device for receiving original message which contains an internet protocol (IP) header and is forwarded by a customer edge (CE), the IP header contains an identification which supports the explicit congestion notification (ECN); and if a congestion state at present is determined by the PE device, the received original message is encapsulated to be generic routing encapsulation (GRE) message, and a network congestion identification is set on an outer layer IP message header of the GRE message. The invention further provides further a method for achieving an explicit congestion notification, which is applied to provider: provider (P: P) device which receives the GRE message forwarded by a PE; and if the congestion state at present is determined by the P device and the network congestion identification is not set on the outer layer IP message header of the GRE message, the network congestion identification is set on the outer layer IP message header of a received GRE message. PE and P devices for achieving the explicit congestion notification are further provided. By means of the methods and the devices for achieving the explicit congestion notification, network device congestion is effectively relieved in GRE tunnel network.

Codec rate adaptation for wireless circuit-switched voice communications routed through a packet-based network. A Mobile Station (MS) preferably with Circuit Switched services participating in a voice communication with a distant terminal transmits an initial CMR identifying an initial maximum speech coding rate selected as a function of a downlink radio quality. At each intermediate network node, one or more operational parameters are determined by using an Explicit Congestion Notification (ECN) protocol. When a CMR is received at each network node, if the operational parameters are not within a predetermined range suitable for the transmission of speech packets through the network using the initial maximum speech coding rate the maximum speech coding rate identified in the CMR is reduced as a function of the operational parameters and then forwarded toward the distant terminal.