Near-source end traffic coordination and control method and system, and computer program product

Abstract

Embodiments of the present disclosure provide a near-source end traffic coordination and control method and system, and a computer program product. An access device sends a plurality of source‑end traffics to an aggregation device, wherein the plurality of source‑end traffics are divided into a plurality of traffic groups by the access device; the access device receives a feedback packet from the aggregation device; and the access device coordinates and controls each source‑end traffic in the traffic groups on the basis of the feedback packet.

Description

Near-source flow coordination control methods, systems, and computer program products

[0001] Cross-reference of related applications

[0002] This application is based on and claims priority to Chinese patent application CN202510186363.7, filed on February 19, 2025, entitled “Near-source flow coordination control method, system and computer program product”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to the field of communications, and more specifically, to a near-source traffic coordination control method, system, and computer program product. Background Technology

[0004] In networks, network device egress congestion caused by the convergence of service flows is a frequent occurrence. Congestion can lead to packet loss, retransmission, and other problems, thus degrading transmission performance. To address this issue, various congestion control algorithms have been proposed at the transport layer. These algorithms utilize network congestion signals, such as changes in Explicit Congestion Notification (ECN) and Round Trip Time (RTT), as well as packet loss, to adjust the sending window and rate at the sending end. Currently, in Linux systems, the default congestion control algorithm for the Transmission Control Protocol (TCP) transport layer is Cubic. This algorithm adjusts the sending window based on packet loss signals; however, it suffers from slow convergence, with convergence times potentially reaching 200 seconds when a new service flow enters the network. Another popular congestion control algorithm is Bottleneck Bandwidth and Round-trip Propagation Time (BBR) and its improved versions, such as BBRv2 and BBRv2+. These algorithms adjust the sending window by probing the bandwidth of the bottleneck link and the network's round-trip time (RTT) in an attempt to solve the buffer bloat problem in long, fat networks. However, the BBR algorithm also takes 30 to 100 seconds to converge. Furthermore, in a wide area network environment, when the delays from each sender to the bottleneck link are unequal, the BBR algorithm cannot guarantee fairness.

[0005] In recent years, the industry has proposed various improved algorithms driven by network-side devices, such as the Robust Congestion Control (RoCC) algorithm. This type of algorithm calculates or probes the fair rate based on information such as the real-time queue depth, queue depth changes, and real-time transmission rate of the bottleneck link port, and sends this rate to each source end. Each source end adjusts its traffic transmission rate based on the rate information fed back by the network device, thereby improving the convergence rate and the utilization of the bottleneck link. However, in a wide area network environment, because the congestion feedback latency is related to the basic RTT (e.g., the basic RTT for 3000KM is 30ms), this algorithm still suffers from unfair inter-flow rate allocation and insufficient link utilization. Summary of the Invention

[0006] This disclosure provides a near-source flow coordination control method, system, and computer program product.

[0007] According to one embodiment of this disclosure, a near-source traffic coordination and control method is provided, comprising: an access device sending multiple originating traffic flows to an aggregation device, wherein the multiple originating traffic flows are divided into multiple traffic groups by the access device; the access device receiving a feedback message from the aggregation device; and the access device coordinating and controlling each of the originating traffic flows within the traffic groups based on the feedback message.

[0008] According to another embodiment of this disclosure, a near-source traffic coordination and control system is provided, including: a transmitting device, an access device, and a aggregation device. The access device is used to send multiple transmitting traffic streams from the transmitting device to the aggregation device, and coordinate and control each transmitting traffic stream within a traffic group based on feedback messages from the aggregation device, wherein the multiple transmitting traffic streams are divided into multiple traffic groups by the access device.

[0009] According to yet another embodiment of this disclosure, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps in any of the above method embodiments when executed.

[0010] According to yet another embodiment of this disclosure, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0011] According to yet another embodiment of this disclosure, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments. Attached Figure Description

[0012] Figure 1 is a schematic diagram of an example of inter-flow velocity distribution in related technologies;

[0013] Figure 2 is a schematic diagram of another scenario example of inter-flow velocity distribution in related technologies;

[0014] Figure 3 is a hardware structure block diagram of a mobile terminal for a near-source traffic coordination control method according to an embodiment of the present disclosure.

[0015] Figure 4 is a flowchart of a near-source flow coordination control method according to an embodiment of the present disclosure;

[0016] Figure 5 is a schematic diagram of the architecture principle of service traffic transmission according to an embodiment of this disclosure;

[0017] Figure 6 is a schematic diagram of another architectural principle of service flow allocation according to an embodiment of this disclosure;

[0018] Figure 7 is a flowchart of a near-source flow coordination control method according to an embodiment of the present disclosure;

[0019] Figure 8 is a schematic diagram of the architecture of near-source traffic coordination control for the initiation of new traffic transmission according to an embodiment of this disclosure;

[0020] Figure 9 is a schematic diagram of the architecture principle of near-source traffic coordination control for new traffic transmission according to an embodiment of this disclosure;

[0021] Figure 10 is a schematic diagram of the architecture of near-source flow coordination control with decreasing feedback rate in related technologies;

[0022] Figure 11 is a schematic diagram of the architecture of near-source traffic coordination control for traffic cessation in an embodiment of this disclosure;

[0023] Figure 12 is a sub-TLV extension example diagram of an embodiment of this disclosure. Detailed Implementation

[0024] The embodiments of this disclosure will be described in detail below with reference to the accompanying drawings and examples.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of the embodiments of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0026] Figure 1 is a schematic diagram illustrating an example of inter-flow rate allocation in related technologies. As shown in Figure 1, assuming the aggregation switch has already allocated rates r to each originating end host2, host3, and host4, a new service flow enters host1. This flow could be a long flow or a short flow. Under the existing mechanism, if the service flow starts sending quickly, when the new service flow arrives at the aggregation switch, it may cause the aggregation switch's buffer to become congested, leading to buffer bloat and rate reduction. If the traffic on host1 is a short flow, the rate reduction operation of the switch will cause a loss in throughput. If the service flow starts at a very slow rate, although it will not cause increased congestion on the aggregation switch, if the traffic on host1 is a long flow with a long duration, it will remain in a low-rate state for a long time before receiving rate feedback from the switch, increasing the completion time of host1's traffic and exacerbating the unfairness of inter-flow rates.

[0027] Figure 2 is a schematic diagram illustrating another scenario of inter-flow rate allocation in related technologies. As shown in Figure 2, after the aggregation switch accumulates buffer data, it reduces its rate to empty the buffer. To allow the aggregation switch to empty its buffer, the sum of the sending rates will be lower than the aggregation switch's egress bandwidth (e.g., the sum of the rates of host1 to host4 is 4 / 5*Bw). The rate will only increase when the switch's buffer is lower than the desired queue depth. Since the accumulated buffer data of devices in a wide area network is large, the buffer emptying time often lasts for several milliseconds. During this stage, if some sending ends stop sending data, the rates of other sending ends will not increase, reducing the overall utilization of the bottleneck switch. The fact that host1 and host3 stop sending data during the buffer emptying stage will result in a 50% throughput loss.

[0028] In related technologies, in a wide area network (WAN) environment, due to the long base time-to-transmission (RTT), switches and originating ends cannot perceive real-time changes in a timely manner, which presents some problems. However, if aggregation switches that are close to each other, such as those belonging to the same point of delivery (POD), can be grouped together, sharing traffic within the group and performing near-end traffic control based on the traffic status within the group, it is possible to smooth the traffic sent to the aggregation switch, improve fairness between flows, and improve the bandwidth utilization of the aggregation link.

[0029] The method embodiments provided in this application can be executed in a mobile terminal, computer terminal, or similar computing device. Taking a mobile terminal as an example, FIG3 is a hardware structure block diagram of a mobile terminal for a near-source flow coordination control method according to an embodiment of this disclosure. As shown in FIG3, the mobile terminal may include one or more (only one is shown in FIG3) processors 302 (processor 302 may include, but is not limited to, processing devices such as microprocessors MCU or programmable logic devices FPGA) and a memory 304 for storing data. The mobile terminal may also include a transmission device 306 for communication functions and an input / output device 308. Those skilled in the art will understand that the structure shown in FIG3 is only illustrative and does not limit the structure of the mobile terminal. For example, the mobile terminal may also include more or fewer components than shown in FIG3, or have a different configuration than shown in FIG3.

[0030] The memory 304 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the near-source flow coordination control method in this embodiment. The processor 302 executes various functional applications and data processing by running the computer program stored in the memory 304, thereby implementing the above-described method. The memory 304 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 304 may further include memory remotely located relative to the processor 302, and these remote memories can be connected to the mobile terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0031] The transmission device 306 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the mobile terminal's communication provider. In one example, the transmission device 306 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 306 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0032] This embodiment provides a near-source flow coordination control method. Figure 4 is a flowchart of the near-source flow coordination control method according to this embodiment. As shown in Figure 4, the process includes the following steps:

[0033] In step S402, the access device sends multiple originating traffic streams to the aggregation device, wherein the multiple originating traffic streams are divided into multiple traffic groups by the access device.

[0034] In one exemplary embodiment, multiple originating traffic flows in each traffic group have the same destination node, and the physical distance between the access devices corresponding to the multiple originating traffic flows in each traffic group satisfies a first threshold.

[0035] In this embodiment of the disclosure, multiple originating traffic flows in each traffic group have the same destination node, i.e., the same destination IP address. The physical distance between the access devices corresponding to the multiple originating traffic flows in each traffic group meets a first threshold. The specific type and value of the first threshold can be determined according to the actual situation. For example, it can be less than 100us latency, and no specific limitation is made here.

[0036] Step S404: The access device receives a feedback message from the aggregation device.

[0037] In one exemplary embodiment, the feedback message is obtained by the aggregation device based on the congestion control algorithm; the protocol type of the feedback message includes: Congestion Notification Protocol message; Internet Control Protocol message.

[0038] In this embodiment of the disclosure, the aggregation device calculates the fair rate between flows through congestion control, and then constructs a feedback message based on the fair rate. The form of the feedback message includes, but is not limited to, a Congestion Notification Protocol message or an Internet Control Protocol message.

[0039] In one exemplary embodiment, the content of the feedback message includes: the rate calculated or detected by the aggregation device; the queue depth of the aggregation device; and the port transmission rate of the aggregation device.

[0040] In this embodiment of the disclosure, the feedback message includes information needed to calculate the fair rate, such as switch queue depth, port transmission rate, etc.

[0041] In step S406, the access device coordinates and controls the traffic of each originating end within the traffic group based on the feedback message.

[0042] In an exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group based on the feedback message, including: the access device obtaining a feedback rate value and a corresponding originating traffic identifier based on the feedback message, wherein the feedback rate value includes the rate value of each originating traffic and / or the sum of feedback rates; and the access device coordinating and controlling the originating traffic within each traffic group based on the feedback rate value and the corresponding originating traffic identifier.

[0043] In one exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group, including: the access device coordinates and controls the originating traffic within each traffic group based on a priority flow control protocol or a credit flow control protocol.

[0044] In this embodiment of the disclosure, the access device coordinates and controls each originating traffic within a traffic group through a near-end flow control protocol. Typical near-end flow control protocols may include Priority Flow Control (PFC), Credit Based Flow Control (CBFC), etc., without limitation.

[0045] In an exemplary embodiment, the access device coordinates and controls the originating traffic in each traffic group based on the feedback rate value and the corresponding originating traffic identifier. This includes: the access device coordinates and controls the rate value of each originating traffic in each traffic group to be the quotient of the sum of the feedback rates divided by the total number of originating traffic.

[0046] In one exemplary embodiment, the method further includes: the access device storing and maintaining flow information of the originating traffic in each traffic group, wherein the flow information includes: the destination IP address corresponding to the originating traffic in each traffic group; the number of originating traffic in each traffic group; the originating traffic identifier of the originating traffic in each traffic group; the sum of the rates of the originating traffic in each traffic group; and the rate limit value of the originating traffic in each traffic group.

[0047] In this embodiment of the disclosure, the access device may store and maintain flow information of originating traffic in each traffic group by maintaining entries in the access switch (i.e., the access device) to record information of the flow with shared rate in the current group.

[0048] In an exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group, including: in response to the access device adding new originating traffic and the new originating traffic belonging to the current traffic group maintained by the access device, the access device updates the flow information corresponding to the new originating traffic and synchronizes the updated flow information to other access devices in the same traffic group; the access device coordinates and controls the rate value of the originating traffic in each traffic group to be the quotient of the sum of the feedback rates divided by the total number of new originating traffic.

[0049] In an exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group, including: in response to the access device reducing the originating traffic, the access device updates the flow information corresponding to the reduced originating traffic and synchronizes the updated flow information to other access devices in the same traffic group; the access device coordinates and controls the rate value of the originating traffic in each traffic group to be the quotient of the sum of the feedback rates divided by the total number of new originating traffic.

[0050] In one exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group, including: in response to the existence of originating traffic with a limited flow rate in the access device, the access device adjusts the originating traffic with the limited flow rate to a corresponding rate limit value, and coordinates and controls the rate value of each originating traffic with an unlimited flow rate to be the quotient of the sum of the unlimited flow rates divided by the total number of originating traffic with unlimited flow rates, wherein the sum of the unlimited flow rates is the difference between the sum of the feedback rates and the limited flow rates.

[0051] In an exemplary embodiment, the access device coordinates and controls each originating traffic within a traffic group, including: in response to at least two access devices having newly added, reduced, or rate-limited originating traffic belonging to the same traffic group, the access device synchronously adds, reduces, or rate-limits originating traffic identifiers for the newly added, reduced, or rate-limited originating traffic among the access devices in the same traffic group; and the access devices in the same traffic group update the corresponding originating traffic identifiers.

[0052] This disclosure provides a near-source traffic coordination and control method. Multiple originating traffic streams are sent from an access device to an aggregation device, where the access device divides these streams into multiple traffic groups. The access device receives feedback messages from the aggregation device. Based on these feedback messages, the access device coordinates and controls each originating traffic stream within its traffic group. This method solves the problems of unfair inter-stream rate allocation and insufficient link utilization in related technologies, achieving fair allocation of inter-stream rates and improved link utilization in service flow transmission.

[0053] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solutions of the embodiments of this disclosure, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the embodiments of this disclosure.

[0054] This disclosure also provides a near-source flow coordination and control system, including: a transmitting device, an access device, and a aggregation device. The access device is used to send multiple transmitting flows from the transmitting device to the aggregation device, and coordinate and control each transmitting flow within a flow group based on feedback messages from the aggregation device, wherein the multiple transmitting flows are divided into multiple flow groups by the access device.

[0055] In this disclosure, the type of access device is not limited, typically such as an access switch. The type of aggregation device is also not limited, typically such as an aggregation switch.

[0056] In one exemplary embodiment, multiple originating traffic flows in each traffic group have the same destination node, and the physical distance between the access devices corresponding to the multiple originating traffic flows in each traffic group satisfies a first threshold.

[0057] In this embodiment of the disclosure, multiple originating traffic flows in each traffic group have the same destination node, i.e., the same destination IP address. The physical distance between the access devices corresponding to the multiple originating traffic flows in each traffic group meets a first threshold. The specific type and value of the first threshold can be determined according to the actual situation. For example, it can be less than 100us latency, and no specific limitation is made here.

[0058] In an exemplary embodiment, the access device is further configured to store and maintain flow information for each group of originating traffic, wherein the flow information includes: the destination IP address corresponding to the originating traffic in each traffic group; the number of originating traffic in each traffic group; the originating traffic identifier of each originating traffic in each traffic group; the sum of the rates of the originating traffic in each traffic group; and the rate limit value of the originating traffic in each traffic group.

[0059] In this embodiment of the disclosure, the access device may store and maintain flow information of originating traffic in each traffic group by maintaining entries in the access switch (i.e., the access device) to record information of the flow with shared rate in the current group.

[0060] In one exemplary embodiment, the near-source flow coordination control system further includes: an information maintenance device, which stores and maintains flow information of the originating flow in each flow group.

[0061] In this embodiment of the disclosure, in addition to being stored and maintained in the access device, the stream information can also be stored and maintained in another device, such as an information maintenance device, wherein the specific type of information maintenance device is not limited.

[0062] This embodiment also provides a near-source flow coordination control device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0063] In this embodiment, the near-source flow coordination control device may further include different modules, and the naming and functional division of these modules may be selected in different ways according to the actual situation, without specific limitations.

[0064] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0065] This disclosure also provides a computer-readable storage medium storing a computer program configured to perform the steps in any of the above method embodiments when executed.

[0066] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0067] This disclosure also provides an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0068] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0069] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0070] In one exemplary embodiment, the computer program product described above includes a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the methods described in various embodiments of this application.

[0071] Specific examples in this embodiment can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0072] Obviously, those skilled in the art should understand that the modules or steps of the embodiments of this disclosure described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the embodiments of this disclosure are not limited to any particular combination of hardware and software.

[0073] To enable those skilled in the art to better understand the technical solutions of the embodiments of this disclosure, the following description is provided in conjunction with different embodiments.

[0074] Example 1

[0075] This disclosure proposes a near-source traffic coordination and control method to improve network transmission performance. This method treats traffic destined for the same bottleneck link (e.g., originating access devices are physically close together, with latency less than 100µs, without strict limitations) as a group, and coordinates and shares the inter-flow transmission rate within the group. This improves upon the unfairness and insufficient link utilization issues inherent in traditional network-side rate-driven algorithms in wide area networks.

[0076] The originating traffic enters the network through the access device (this embodiment does not limit the type of access device, such as the access switch, which will be used as an example in the following description). The coordination and control between traffic is completed through the access switch, and a signaling mechanism is added between the access switches to transmit flow control coordination information.

[0077] Figure 5 is a schematic diagram of the architecture principle of service traffic transmission in an embodiment of this disclosure. As shown in Figure 5, each access switch may connect to multiple originating devices, and the originating traffic is transmitted through the same aggregation device (this embodiment of the disclosure does not limit the type of aggregation device, for example, an aggregation switch is used as an example). The sum of the rates of traffic feedback from the aggregation device to host1 to host3 is assumed to be R.

[0078] In one embodiment, the access device (i.e., the access switch) is configured in a traditional operating mode, where no traffic group information coordination or rate sharing occurs between access switches.

[0079] In this embodiment of the disclosure, it is assumed that the number of traffic belonging to the same bottleneck link within the current switch group is N, and the sum of the feedback rates of all flows is R. The overall approach to traffic coordination and control is as follows:

[0080] If a new traffic flow destined for the same bottleneck link is added to the group, the rate of each flow will be set to R / (N+1) through the inter-switch coordination mechanism.

[0081] If one traffic item in the group stops sending, then through the coordination mechanism between switches, the rate of each traffic item is set to R / (N-1).

[0082] If there is active rate limiting of originating traffic in the group (e.g., the originating port rate is less than the feedback rate), and the number of rate-limited traffic is n, and the sum of the rate-limited traffic rates is r, then through the coordination mechanism between switches, the rate of other non-rate-limited traffic is set to (Rr) / (Nn).

[0083] In this embodiment of the disclosure, the traffic rate limiting mechanism between the access device and the originating end is relatively flexible. For example, the originating end rate limiting can be performed based on the existing rate feedback messages, or the near-end flow control backpressure can be performed based on the rate limiting configuration. This disclosure does not restrict the interaction method between the access switch and the originating end.

[0084] It's important to note that if the newly added traffic is the first traffic entry in the group, or if the stopped / rate-limited traffic is the last traffic entry in the group (i.e., there is only one traffic entry), the access switch will not perform near-end rate adjustment as described above. Another scenario is if the access switch has already determined that multiple traffic entries belong to the same group, but has not yet received the first rate information from the network device (at which point the total group rate is 0). In this case, the access switch only performs information exchange and maintenance such as the number of traffic entries, but does not perform near-end flow control feedback. Only after receiving the rate information from the network device will it implement a near-end control mechanism based on traffic coordination and sharing within the group.

[0085] In the embodiments of this disclosure, other related intra-group rate coordination mechanisms, such as sharing throughput among traffic flows in a certain proportion (e.g., traffic-based purchase quotas), or adjusting rates among traffic flows proportionally, are all optional mechanisms and are not limited in this disclosure. Since the access switch and the originating device are physically close (typically within a single POD), the coordination time can be controlled at the microsecond level, thereby greatly improving efficiency. Therefore, this disclosure does not involve coordination mechanisms between originating traffic flows that are physically far apart (e.g., in different data centers).

[0086] In this embodiment of the disclosure, the access switch needs to maintain entries to record information about flows with shared rates within the current group, where flows share the same bottleneck link. For example, the destination IP address is used to determine whether these flows belong to the same group, and the number N of flows within the group is maintained. Table 1 is an example table of flow information maintenance within a traffic group according to this embodiment of the disclosure. As shown in Table 1, in one embodiment, other necessary information can also be maintained in the access switch, such as the total rate of the group, the local traffic rate limit, etc.

[0087] Table 1. Example of maintaining flow information within a flow group.

[0088] In this embodiment of the disclosure, if there is traffic belonging to the same group among the access switches, signaling will be used to exchange and synchronize traffic coordination signaling information within the group, including the following processing methods:

[0089] 1) When traffic joins an access switch, it first determines whether it belongs to a certain group through signaling protocol. If so, the switches exchange signaling to synchronize the current entry information of that group. The count of flows belonging to that destination IP is incremented by 1, and this information is sent to other access switches via signaling.

[0090] 2) When traffic stops at an access switch, the number of flows belonging to that destination IP is decremented by 1, and this information is sent to other access switches via signaling.

[0091] 3) When the access switch has traffic rate limiting / cancels the rate limiting, the rate limiting information of the flow is sent to other access switches.

[0092] After synchronizing flow information within the group, each switch updates the rate of each flow. For cases 1) and 2), each switch directly divides the total rate by the number of flows after synchronization to obtain the rate limit value for each flow, and controls the flow rate of each flow through near-end flow control. For case 3), each switch obtains the flow rate of each flow by (total rate - rate limit rate) / number of unrate-limited flows, and performs near-end flow control on rate-limited flows according to the rate limit value.

[0093] In this embodiment, a special case arises when traffic belonging to the same group simultaneously (or at very close intervals) joins, stops, or is rate-limited on two access switches. The coordination mechanism between the access switches needs to support this situation. One alternative is that each switch does not directly maintain the number of flows, but instead maintains the flow ID information (e.g., for ease of storage, the five-tuple information such as source IP, source port, and destination port is converted into a single number), as shown in Figure 6. The number of flow IDs within the group then represents the number of flows. When traffic joins or leaves, the added or stopped flow ID information is synchronized to other node access switches via signaling. The access switches then add or delete the flow ID information in their local flow ID list.

[0094] In this embodiment, flow information can be stored and maintained not only in the access device but also in another device, such as an information maintenance device. The specific type of information maintenance device is not limited. Figure 6 is a schematic diagram of another architecture principle for service flow allocation in this embodiment. As shown in Figure 6, flow information belonging to the same group is maintained by a separate device (i.e., an information maintenance device). When each access switch needs to update information, it queries the table entries maintained in that device. The content of the table entries is the same as the information in Table 1. As shown in Figure 6, compared to each access switch maintaining table entries, this method allows the access switch to interact with the device via signaling when new flow is added or flow is stopped. This embodiment does not limit the method of maintaining table entry information.

[0095] As shown in Figure 6, the information carried by the signaling between the access switch and the information maintenance equipment includes, but is not limited to: group information, typically destination address information; number of flows; rate limit value fed back to each flow by the aggregation device (used to calculate the total rate); flow ID information; and rate limit information at the flow origin.

[0096] In this embodiment of the disclosure, access switches can synchronize group information in various ways, such as Network Configuration Protocol (Netconf), Link Layer Discovery Protocol (LLDP), Resource Reservation Protocol-Traffic Engineering Extension (RSVP-TE), or Border Gateway Protocol-Link State (BGP-LS). This embodiment of the disclosure does not limit the types of protocols used or the specific encapsulation format of the extended information.

[0097] In this embodiment, information is fed back to the source via rate feedback messages. The protocol type of the rate feedback messages is not limited; typical examples may be based on Congestion Notification Protocol (CNP) messages, Internet Control Message Protocol (ICMP) messages, etc. The information fed back by the switch may include:

[0098] a. The rate calculated or detected by the switch congestion control algorithm;

[0099] b. Or information needed to calculate fair rate, such as switch queue depth, port transmission rate, etc.

[0100] Figure 7 is a flowchart of the near-source flow coordination control method according to an embodiment of this disclosure. As shown in Figure 7, near-source flow control coordination is completed through the access switch. The overall process of coordination between the access switch and the rate feedback congestion control algorithm is as follows:

[0101] 1) The congestion control algorithm calculates the rate of traffic at each originating end and sends the calculated rate information back to each originating end through feedback messages.

[0102] 2) When the feedback message passes through the access switch, the access switch extracts the information carried in the feedback message, including flow ID information, feedback rate value, etc.

[0103] 3) The access switch updates the total rate limit for the traffic in this group by multiplying the rate limit value by the number of flows in the group;

[0104] 4) The access switch controls the transmission rate of each flow according to the new per-flow rate limit. Typical near-end flow control protocols may be Priority Flow Control (PFC), Credit Based Flow Control (CBFC), etc., without imposing restrictions.

[0105] 5) When new traffic is added to / stopped / rate-limited at the originating access switch, the access switch updates the entries and rates of each traffic according to the above signaling interaction, thereby performing rapid control of near-end traffic.

[0106] In this embodiment of the disclosure, as shown in FIG7, the process of coordination control performed by the near-end access switch when a rate feedback message is received is as follows: if the number of non-rate-limited traffic originating at the origin has reached 0 after traffic stops or rate limiting within the group, then it is not necessary to update the rate of each flow.

[0107] Example 2

[0108] In this embodiment, the near-end flow control coordination feedback mechanism based on the present disclosure is described when new traffic begins to be sent.

[0109] Figure 8 is a schematic diagram of the near-source traffic coordination control architecture of a new traffic transmission embodiment of this disclosure. As shown in Figure 8, there are three originating ends, host1, host2, and host3, each sending one service flow, with the destination node being sink1. The RTT of the links between the three hosts and the access switches (switch1, switch2, switch3) is 1µs. The out-of-band interconnection between the access switches is used to transmit rate coordination messages, and the maximum RTT between the access switches is 3µs (within 300 meters). The traffic sent by the three originating ends is sent to sink1 node (IP address 192.168.1.1) via the wide area network. sink1 node is connected to the aggregation switch aggrSwitch. The RTT from switches1 to switch3 to aggrSwitch is 30ms (3000KM). A rate feedback congestion control algorithm (RoCC algorithm is used as an example) is deployed at the egress of aggrSwitch. The egress port rate is assumed to be 40Gbps, and the port bandwidth of each originating end is also 40Gbps.

[0110] As shown in Figure 8, assuming that host1 and host2 have sent traffic flow1 and flow2, and relevant information has been exchanged between switch1 and switch2 through the signaling protocol, the RoCC algorithm feeds back a fair rate of 12Gbps to the sending end.

[0111] Figure 9 is a schematic diagram of the near-end flow coordination control architecture for newly added traffic transmission according to an embodiment of this disclosure. As shown in Figure 9, when host3 sends traffic flow3 to sink1, when access switch switch3 receives the traffic from host3, it discovers through signaling interaction with switch1 or switch2 that flow3 belongs to the same group as flow1 and flow2. Then switch1 obtains the current group's traffic information through the signaling protocol and performs traffic coordination. After flow3 joins, the rate limit value for each flow becomes: (12+12) / 3 = 8Gbps. Then, the three access switches limit the transmission rate of the three flows to 8Gbps through the near-end flow control mechanism. The traffic of flow3 can be smoothly added to the network, and fairness between flows is guaranteed as much as possible.

[0112] In the above implementation process, if the near-end flow control coordination mechanism described in this disclosure is not adopted, and flow3 is a short flow (assuming a size of 5MB) that starts at line speed, then flow3 will arrive suddenly within 1ms of the aggregation switch, causing an increase in the aggregation switch's buffer. If the aggregation switch determines that the fair rate is reduced to 4Gbps at this time, then both host2 and host3 will be reduced to 4Gbps. However, since flow3 is a short flow, by the time the rate feedback information reaches the sending end, flow3 has already stopped sending traffic. This situation will lead to a decrease in the utilization of the bottleneck link of the aggregation switch. Similarly, if the near-end flow control coordination mechanism described in this disclosure is not adopted, assuming flow3 is a long flow that starts at a slow speed, then during the startup phase (before receiving the rate feedback from the switch), flow3 will always maintain a low rate, thus failing to guarantee fairness with flow1 and flow2.

[0113] Using the method described in this embodiment, both short and long flows benefit during the startup phase, ensuring fairness while increasing throughput. In subsequent phases, if the new traffic is a long flow, the original RoCC feedback rate adjustment method continues to achieve improved transmission performance.

[0114] Example 3

[0115] In this embodiment, when traffic flow is stopped, the near-end flow control coordination mechanism described in this disclosure is used for flow coordination control. The network topology assumption in this embodiment can be the same as that used in Embodiment 2.

[0116] In this embodiment of the disclosure, the RoCC congestion control algorithm is deployed at the bottleneck link egress. During RoCC operation in a wide area network, there is a rate-reduction phase that causes the sum of the originating rates to be less than the aggregation link bandwidth, thereby allowing the aggregation switch to empty its buffer. This process may last for several milliseconds. The rate calculated and fed back by the aggregation switch, typically based on the RoCC algorithm, shows fluctuations in the feedback rate that last for several milliseconds (e.g., approximately 25 milliseconds) at its lowest point.

[0117] Figure 10 is a schematic diagram of the near-source traffic coordination control architecture for reduced feedback rate in related technologies. As shown in Figure 10, assuming the current switch feedback rate drops to 8Gbps (the total rate of the three originating ends is 24Gbps), since the buffer has not been emptied, the aggregation switch will not increase the feedback rate regardless of the originating end rate (until the buffer is lower than the expected value). At this time, if the traffic transmission of originating end host1 ends, in the existing scheme, host2 and host3 will continue to transmit at 8Gbps, which will cause the throughput of the aggregation link to drop by 8Gbps (i.e., the rate of flow1) for a period of time.

[0118] Figure 11 is a schematic diagram of the near-source traffic coordination control architecture for traffic cessation according to an embodiment of this disclosure. As shown in Figure 11, using the rate coordination mechanism of this embodiment, when flow1 stops transmitting traffic, it notifies other access switches through the signaling protocol between switches. Switch2 and Switch3 then update their flow count to 2 and update the traffic rate limit values ​​for host2 and host3 to 12Gbps. This approach avoids throughput loss when the feedback rate is low.

[0119] Example 4

[0120] In this embodiment, an extended encapsulation method for signaling protocol messages between access switches is introduced.

[0121] As described in this embodiment, the information that needs to be coordinated between switches includes group information (typically destination IP address), number of flows, network-side feedback rate, flow ID, flow rate limit, etc. Assuming that group entry information is maintained in a separate device (typically a POD management unit), the access switch can interact with the management control unit via the Netconf protocol to transmit the above information. When traffic changes occur on the access device, the management control unit can be notified through the notification mechanism provided by the Netconf protocol, and the management control unit will then update the corresponding entries. The extended method for announcing the above information based on the Netconf notification mechanism is as follows:

[0122] In this embodiment of the disclosure, it is assumed that the group information is maintained locally by each access switch, and the access switches interact and transmit the above information through signaling protocols. Figure 12 is an example diagram of the sub-TLV extension of this embodiment of the disclosure. As shown in Figure 12, it is assumed that the extended sub-TLV is defined based on the ISIS protocol. The sub-TLV includes the traffic transmission rate control field "send-rate-control-flow-num", the traffic identifier "flow-id", the rate limit "rate-limit", etc.

[0123] This disclosure provides a mechanism for near-end flow control and coordination, addressing some issues arising from the long feedback cycle of network-side device-driven algorithms in wide area networks. By grouping traffic originating from similar physical locations and belonging to the same destination node, near-end flow control coordination is performed within each group. Access switches exchange signaling protocol channels to synchronize the status of traffic within the group, including the number of flows. When traffic receives a network-side rate feedback message, or when traffic joins / stops within the group, inter-flow rate coordination and sharing are performed, thereby smoothing traffic flow and improving utilization.

[0124] Current mainstream congestion control algorithms still suffer from slow convergence speed, unfairness, and low bottleneck link utilization in wide-area environments. In this context, congestion control algorithms and variants based on network-side device rate probing and feedback have been proposed, offering performance improvements. However, due to the inherently long base RTT (Round-Trip Time), the feedback distance between the originating device and the bottleneck link device is significant. The originating device coordination control mechanism proposed in this disclosure leverages the proximity of the originating devices' physical locations (e.g., within the same POD) to quickly share traffic shares. This mechanism smooths traffic flow, improves bandwidth utilization, and enhances inter-flow fairness at various stages, including traffic initiation and cessation.

[0125] The embodiments disclosed herein can be applied to routers and switches, as well as the corresponding configuration units of Remote Direct Memory Access (RDMA) servers, etc.

[0126] The above description is merely a preferred embodiment of this disclosure and is not intended to limit the scope of this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A near-source flow coordination control method, comprising: The access device sends multiple originating traffic streams to the aggregation device, wherein the multiple originating traffic streams are divided into multiple traffic groups by the access device; The access device receives a feedback message from the aggregation device; Based on the feedback message, the access device coordinates and controls the traffic of each originating end within the traffic group.

2. The method according to claim 1, wherein, The multiple originating traffic in each traffic group has the same destination node, and the physical distance between the multiple originating traffic in each traffic group and the access device satisfies a first threshold.

3. The method according to claim 1, wherein, The feedback message is obtained by the aggregation device based on a congestion control algorithm; the protocol type of the feedback message includes: Congestion Notification Protocol (CIP) messages; Internet Control Protocol (ICP) messages.

4. The method according to claim 3, wherein, The content of the feedback message includes: The rate at which the aggregation device calculates or detects data; the queue depth of the aggregation device; and the port transmission rate of the aggregation device.

5. The method according to claim 1, wherein, Based on the feedback message, the access device coordinates and controls the traffic of each originating end within the traffic group, including: The access device obtains a feedback rate value and a corresponding originating traffic identifier based on the feedback message, wherein the feedback rate value includes the rate value of each originating traffic and / or the sum of the feedback rates; The access device coordinates and controls the originating traffic within each traffic group based on the feedback rate value and the corresponding originating traffic identifier.

6. The method according to claim 5, wherein, The access device coordinates and controls the traffic of each originating end within the traffic group, including: The access device coordinates and controls the originating traffic within each traffic group based on either a priority flow control protocol or a credit flow control protocol.

7. The method according to claim 5, wherein, The access device, based on the feedback rate value and the corresponding originating traffic identifier, coordinates and controls the originating traffic within each traffic group, including: Based on the feedback rate value and the corresponding originating traffic identifier, the access device coordinates and controls the rate value of each originating traffic in each traffic group to be the quotient of the sum of the feedback rates divided by the total number of originating traffic.

8. The method according to claim 1, wherein, Also includes: The access device stores and maintains flow information of the originating traffic in each of the traffic groups, wherein the flow information includes: The destination IP address corresponding to the originating traffic in each traffic group; the number of originating traffic in each traffic group; the originating traffic identifier of the originating traffic in each traffic group; the sum of the rates of the originating traffic in each traffic group; and the rate limit value of the originating traffic in each traffic group.

9. The method according to claim 5, wherein, The access device coordinates and controls the traffic of each originating end within the traffic group, including: In response to the access device adding the originating traffic and the added originating traffic belonging to the current traffic group maintained by the access device, the access device updates the flow information corresponding to the added originating traffic and synchronizes the updated flow information to other access devices in the same traffic group; The access devices coordinate and control the rate value of the originating traffic in each of the traffic groups to be the quotient of the sum of the feedback rates divided by the total number of new originating traffic.

10. The method according to claim 5, wherein, The access device coordinates and controls the traffic of each originating end within the traffic group, including: In response to the access device reducing the originating traffic, the access device updates the flow information corresponding to the reduced originating traffic and synchronizes the updated flow information to other access devices in the same traffic group; The access devices coordinate and control the rate value of the originating traffic in each of the traffic groups to be the quotient of the sum of the feedback rates divided by the total number of new originating traffic.

11. The method according to claim 5, wherein, The access device coordinates and controls the traffic of each originating end within the traffic group, including: In response to the existence of a rate-limited originating traffic in the access device, the access device adjusts the rate-limited originating traffic to the corresponding rate limit value, and coordinates the rate value of each unrestricted originating traffic to be the quotient of the sum of unrestricted rates divided by the total number of unrestricted originating traffic, wherein the sum of unrestricted rates is the difference between the sum of feedback rates and the rate limit.

12. The method according to claim 5, wherein, The access device coordinates and controls the traffic of each originating end within the traffic group, including: In response to the existence of new, reduced, or rate-limited originating traffic belonging to the same traffic group in at least two access devices, the originating traffic identifier of the new, reduced, or rate-limited originating traffic is synchronously added, reduced, or rate-limited between the access devices in the same traffic group; The access device in the same traffic group updates the corresponding originating traffic identifier.

13. A near-source flow coordination control system, comprising: Generating equipment, access equipment, and aggregation equipment, The access device is configured to send multiple originating traffic flows from the originating device to the aggregation device, and coordinate and control each originating traffic flow within a traffic group based on feedback messages from the aggregation device, wherein the multiple originating traffic flows are divided into multiple traffic groups by the access device.

14. The system according to claim 13, wherein, The multiple originating traffic in each traffic group has the same destination node, and the physical distance between the multiple originating traffic in each traffic group and the access device satisfies a first threshold.

15. The system according to claim 13, wherein, The access device is also used to store and maintain flow information for each group of originating traffic, wherein the flow information includes: The destination IP address corresponding to the originating traffic in each traffic group; the number of originating traffic in each traffic group; the originating traffic identifier of each originating traffic in each traffic group; the sum of the rates of the originating traffic in each traffic group; and the rate limit value of the originating traffic in each traffic group.

16. The system according to claim 13, wherein, It also includes: an information maintenance device, which is used to store and maintain flow information of the originating flow in each of the flow groups.

17. A computer-readable storage medium storing a computer program, wherein, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 12.

18. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the steps of the method of any one of claims 1 to 12.

19. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 12.

Images