Data retransmission method, device, chip, network interface card, equipment, medium and program product

Abstract

The application relates to a data retransmission method, device, chip, network interface card, equipment, medium and program product. The method comprises the following steps: sending target data to a data receiving end; receiving a packet loss information message returned by the data receiving end; the packet loss information message is generated for the lost target data in the case that the data receiving end determines that there is packet loss, and only one packet loss information message is sent for each lost target data; and the lost target data is re-sent to the data receiving end based on the packet loss information message. The method can reduce the reverse load.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a data retransmission method, apparatus, chip, network interface card, device, medium, and program product. Background Technology

[0002] Due to the instability of network physical links and improper congestion control, packet loss may occur during transmission. In order to ensure the stability and reliability of the transmission process, lost packets need to be retransmitted.

[0003] In traditional technologies, the go-back-n scheme requires retransmission of all subsequent packets if a single packet is lost. This large number of retransmissions exacerbates network congestion, leading to even more packet loss and creating a negative feedback loop. Traditional selective repeat, upon detecting a lost packet, generates a bitmap (BMP) packet for every subsequent out-of-order packet received. While BMP packets are small, they still consume limited bandwidth, increasing the load on the reverse side. Summary of the Invention

[0004] Therefore, it is necessary to provide a data retransmission method, apparatus, chip, network interface card, device, medium, and program product that can reduce reverse load in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a data retransmission method applied at a data sending end, the method comprising:

[0006] Send the target data to the data receiving end;

[0007] Receive the packet loss information message returned by the data receiving end; the packet loss information message is generated by the data receiving end for the lost target data when it determines that packet loss exists, and only one packet loss information message is sent for each lost target data.

[0008] Based on the packet loss information message, the lost target data is retransmitted to the data receiving end.

[0009] In one embodiment, after sending the target data to the data receiver, the method further includes:

[0010] Start the timeout timer corresponding to the target data, and the timeout timer duration is the historical maximum RTT;

[0011] If the timeout timer expires and no feedback message is received from the data receiver, the target data is resent to the data receiver. The feedback message is either an acknowledgment message or a packet loss information message.

[0012] In one embodiment, the step of retransmitting the target data to the data receiver and retransmitting the lost target data to the data receiver includes:

[0013] The priority of the target data that needs to be retransmitted is set to the first target priority, and the target data is retransmitted to the data receiving end based on the first target priority.

[0014] In one embodiment, after retransmitting the target data to the data receiver and retransmitting the lost target data to the data receiver, the process includes:

[0015] Reset the timeout timer corresponding to the retransmitted target data.

[0016] In one embodiment, the method further includes:

[0017] Upon receiving the acknowledgment message, cancel the timeout timer for the target message corresponding to the acknowledgment message.

[0018] In one embodiment, before retransmitting the lost target data to the data receiver, the process includes:

[0019] The mode of the packet loss information packet is determined based on the mode bits of the packet loss information packet;

[0020] When the packet loss information message is identified by bit bits, the number of valid bits is read from the packet loss information message, and the packet loss information message is parsed based on the number of valid bits to determine the target data that needs to be retransmitted.

[0021] In one embodiment, before retransmitting the lost target data to the data receiver, the process includes:

[0022] The mode of the packet loss information packet is determined based on the mode bits of the packet loss information packet;

[0023] If the packet loss information message is not in the mode of identifying packet loss by bits, the packet loss information message is parsed to obtain the minimum and maximum identifiers of the successfully received target data; based on the minimum and maximum identifiers of the successfully received target data, the target data that needs to be retransmitted is determined.

[0024] In one embodiment, the method further includes:

[0025] The current RTT is obtained based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message;

[0026] The current RTT smoothing value is obtained based on the current RTT and the previous RTT smoothing value;

[0027] Determine the maximum RTT based on historical RTTs and the current RTT;

[0028] Based on the current RTT smoothing value and the maximum RTT, the historical maximum RTT is obtained.

[0029] Secondly, this application also provides a data retransmission method, applied at a data receiving end, the method comprising:

[0030] Receive the target data sent by the data sender;

[0031] If packet loss is determined based on the target data, and the lost target data is not notified to the data sender, a packet loss information message is generated for the lost target data.

[0032] The packet loss information message is sent to the data sending end, and only one packet loss information message is sent for each lost target data. The packet loss information message is used to instruct the data sending end to resend the lost target data to the data receiving end.

[0033] In one embodiment, the method further includes:

[0034] The receiving end receives retransmitted target data sent by the data sending end. The retransmitted target data is sent to the data receiving end when the data sending end last sent the target data, after a timeout timer is started and the timeout timer expires and no feedback message corresponding to the target data is received. The timeout timer duration is the historical maximum RTT, and the feedback message is an acknowledgment message or a packet loss information message.

[0035] In one embodiment, the historical maximum RTT is determined based on the current RTT smoothing value and the maximum RTT. The current RTT smoothing value is determined based on the current RTT and the previous RTT smoothing value. The maximum RTT is determined based on each historical RTT and the current RTT. The current RTT is determined based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message.

[0036] In one embodiment, sending the packet loss information message to the data sender includes:

[0037] The priority of the packet loss information message is set to the second target priority, and the packet loss information message is sent to the data sending end based on the second target priority.

[0038] In one embodiment, before sending the packet loss information message to the data sender, the method further includes:

[0039] In the case of congestion-induced packet loss, a first mode bit of a packet loss information message corresponding to the congestion-induced packet loss is generated;

[0040] Determine the number of packet losses, and based on the number of packet losses, determine the number of valid bits;

[0041] The value of the bits in the packet loss information message is determined based on the lost data;

[0042] Based on the first mode bit, the number of effective bits, and the value of the bit, a packet loss information message is obtained.

[0043] In one embodiment, before sending the packet loss information message to the data sender, the method further includes:

[0044] In the case of packet loss being a faulty packet loss, a second mode bit of a packet loss information message corresponding to the faulty packet loss is generated;

[0045] Determine the minimum and maximum identifiers of the successfully received target data based on the lost data;

[0046] Based on the second mode bit, the minimum and maximum identifiers of the successfully received target data, a packet loss information message is obtained.

[0047] Thirdly, this application also provides a data retransmission device, applied at a data sending end, the device comprising:

[0048] The first sending module is used to send target data to the data receiving end;

[0049] The first receiving module is used to receive the packet loss information message returned by the data receiving end; the packet loss information message is generated by the data receiving end for the lost target data when it determines that packet loss exists, and only one packet loss information message is sent for each lost target data.

[0050] The retransmission module is used to retransmit the lost target data to the data receiving end based on the packet loss information message.

[0051] Fourthly, this application also provides a data retransmission device, applied at a data receiving end, the device comprising:

[0052] The second receiving module is used to receive target data sent by the data sending end;

[0053] The packet loss information generation module is used to generate a packet loss information message for the lost target data when packet loss is determined based on the target data and the lost target data is not notified to the data sending end.

[0054] The second sending module is used to send the packet loss information message to the data sending end, and only sends the packet loss information message once for each lost target data. The packet loss information message is used to instruct the data sending end to resend the lost target data to the data receiving end.

[0055] Fifthly, this application also provides a chip including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0056] Sixthly, this application also provides a network interface card, including the chip described above and multiple interfaces.

[0057] The chip processes data or communicates with the outside world through the interface.

[0058] Seventhly, this application also provides a computer device including the aforementioned network interface card, which is used for processing data or external communication.

[0059] Eighthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0060] Ninthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0061] The aforementioned data retransmission method, apparatus, chip, network interface card, device, medium, and program products send target data to a data receiving end; receive packet loss information messages returned by the data receiving end; the packet loss information messages are generated by the data receiving end in response to the determination of packet loss, and only one packet loss information message is sent for each lost target data; based on the packet loss information messages, the lost target data is retransmitted to the data receiving end. In this way, only one packet loss information message is sent to notify the data sending end for the same packet loss, reducing the additional bandwidth occupation of packet loss information messages and reducing reverse load. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a schematic diagram of the go-back-n retransmission method in traditional technology;

[0064] Figure 2 This is a schematic diagram of the selective repeat retransmission method in traditional technology;

[0065] Figure 3 This is a diagram illustrating the application environment of a data retransmission method in one embodiment.

[0066] Figure 4 This is a flowchart illustrating a data retransmission method in one embodiment;

[0067] Figure 5 This is a schematic diagram of congestion and packet loss in one embodiment;

[0068] Figure 6 This is a schematic diagram of packet loss failure in one embodiment;

[0069] Figure 7 This is a flowchart illustrating the data retransmission method in another embodiment;

[0070] Figure 8 This is a framework diagram of data retransmission in one embodiment;

[0071] Figure 9 This is a structural block diagram of a data retransmission device in one embodiment;

[0072] Figure 10 This is a structural block diagram of the data retransmission device in another embodiment;

[0073] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0075] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0076] Combination Figure 1 As shown, Figure 1 This is a schematic diagram of the traditional go-back-n retransmission method. In this method, the receiver notifies the sender of the point of packet loss after detecting it, and the sender then reverts to that point to retransmit all packets. The process is as follows:

[0077] a) The sender assigns a number to each message and sends the packet sequence number (PSN) along with the data message.

[0078] b) If the intermediate switch loses packets (as shown in the figure, packets with PSN=3 are dropped by the switch).

[0079] c) The receiving end determines that packets with PSN values ​​of 0, 1, and 2 are consecutive and therefore considered to have not been lost, sending an ACK message to the sending end to notify that the packets have been received correctly. However, if a packet with PSN=2 is received immediately afterward and then a packet with PSN=4 is received, the receiving end detects the PSN is not consecutive and determines that the packet with PSN=3 has been discarded. Therefore, it sends a NAK message to the sending end to notify that the packet with PSN=3 has been discarded. Furthermore, any subsequent packets with PSN greater than 3 are discarded by the receiving end without further response.

[0080] d) Although the sender has already sent out the packets with PSN=4 and PSN=5, it will still retransmit the packets with PSN=3, 4, and 5 once.

[0081] Combination Figure 2 As shown, Figure 2 This diagram illustrates the selective repeat retransmission method in traditional technology. Similar to go-back-n, the receiver checks the PSN to determine if packets have been lost. The difference lies in the information the receiver returns to the sender after detecting packet loss and the sender's subsequent retransmission behavior. The specific process is as follows:

[0082] a) The sender assigns a number to each message and sends the packet sequence number (PSN) along with the data message.

[0083] b) If the intermediate switch loses packets (as shown in the figure, packets with PSN=3 are dropped by the switch).

[0084] c) The receiving end determines that packets with PSNs of 0, 1, and 2 are consecutive and therefore considered to have not been lost, sending an ACK to notify the sending end that the packets have been received correctly. However, if a packet with PSN=2 is received followed immediately by a packet with PSN=4, the receiver finds the PSNs are not consecutive and determines that the packet with PSN=3 has been discarded. In this case, the receiving end does not directly send a NAK message but instead sends a bitmap (BMP). The bitmap consists of multiple bits, each representing a packet; 0 indicates the packet was not received, and 1 indicates it was received. A BMP message is sent for each subsequent valid packet received.

[0085] d) After receiving a BMP message, the sending end needs to parse the BMP bitmap information, retransmit messages with a bit value of 0, and messages with a bit value of 1 indicate that the receiving end has received the message correctly and does not need to be retransmitted.

[0086] Therefore, traditional technologies have at least the following problems:

[0087] 1) Low efficiency: The go-back-n scheme requires retransmission of all subsequent packets if a single packet is lost. A large number of retransmissions exacerbate network congestion, leading to more packet loss and creating a negative feedback loop. Traditional selective repeat generates a BMP packet for every out-of-order packet received after detecting a lost packet. While BMP packets are small, they still consume limited bandwidth, increasing the load in the reverse direction.

[0088] 2) Information Redundancy: Traditional selective repeat senders receive a large number of BMP messages with identical information, and these BMP messages carrying the same information are discarded. This forces the sender to additionally check whether a message with a BMP bit of 0 has been retransmitted before.

[0089] 3) Complex implementation: Traditional selective repeat requires historical records to determine whether a packet has been retransmitted after receiving a large number of BMPs, necessitating additional resources for recording and calculation. This makes the implementation relatively complex.

[0090] 4) Traditional Selective Repeat BMP messages use bitmaps to represent packet loss information. A large number of consecutive packet losses will waste a lot of bitmaps, which may lead to insufficient bitmaps or a waste of a lot of bandwidth.

[0091] To address at least one of the aforementioned technical problems, the data retransmission method provided in this application embodiment can be applied to, for example... Figure 3 In the application environment shown, the data sender 102 communicates with the data receiver 104 via a network.

[0092] Data sender 102 sends target data to data receiver 104. Data receiver 104 receives the target data sent by data sender 102 and determines whether packet loss has occurred based on the target data. If packet loss exists and the lost target data has not been notified to data sender 102, a packet loss information message is generated for the lost target data and sent to data sender 102. Based on the packet loss information message, data sender 102 retransmits the lost target data to data receiver 104. In this way, only one packet loss information message is sent to notify the data sender for the same packet loss, reducing the additional bandwidth consumption of packet loss information messages and reducing reverse load.

[0093] The data transmitter 102 and data receiver 104 can be terminals or servers. Terminals can be, but are not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection equipment, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted displays, etc. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. Servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing cloud computing services.

[0094] In one exemplary embodiment, such as Figure 4 As shown, a data retransmission method is provided, which is applied to... Figure 3 Taking the data sending end as an example, the explanation includes the following steps 402 to 406. Wherein:

[0095] S402: Send the target data to the data receiving end.

[0096] The target data refers to the data that needs to be sent to the data receiving end. In this application, each target data to be sent to the data receiving end is identified to determine the transmission order of the target data. In this way, the data receiving end can determine whether there is packet loss based on the identification of each target data.

[0097] S404: Receive packet loss information message returned by the data receiver; the packet loss information message is generated by the data receiver when it determines that packet loss has occurred, and only one packet loss information message is sent for each lost target data.

[0098] A packet loss information message is generated when the data receiver determines that packet loss has occurred, and the lost target data has not been notified to the data sender. This packet loss information message is used to indicate the lost target data. Optionally, the packet loss information message includes an identifier of the lost target data, or it includes an identifier of the unlost target data, or it includes both the identifier of the lost target data and the identifier of the unlost target data.

[0099] After receiving the target data, the data receiving end first determines whether there is packet loss based on the identifier of the previously received target data and the identifier of the currently received target data. If packet loss exists and the lost target data has not been notified to the data sending end, a packet loss information message is generated and sent to the data receiving end.

[0100] For ease of understanding, assume the data sender sends data 1 through 10 to the data receiver. The data receiver receives data 1, data 2, and data 5. Upon receiving data 5, the receiver can determine that data 3 and data 4 are lost, and since the loss of data 3 and data 4 was not reported to the data sender, a data loss message corresponding to data 3 and data 4 is generated and sent to the data sender. Subsequently, the data receiver receives data 6, data 7, and data 10. Upon receiving data 10, the receiver determines that data 8 and data 9 are lost, generates a packet loss information message corresponding to data 8 and data 9, and sends it to the data sender.

[0101] S406: Based on the packet loss information message, resend the lost target data to the data receiver.

[0102] After receiving a packet loss information message, the data sender parses the message to determine the lost target data and then retransmits the lost target data.

[0103] Optionally, parsing the packet loss information message may include: if the packet loss information message includes an identifier of the lost target data, directly parsing the packet loss information message to obtain the identifier of the lost target data; if the packet loss information message includes an identifier of the unlost target data, directly parsing the packet loss information message to obtain the unlost target data, and obtaining the lost target data based on the unlost target data; if the packet loss information message includes both an identifier of the lost target data and an identifier of the unlost target data, directly parsing to obtain the identifier of the lost target data.

[0104] The above data retransmission method involves sending the target data to the data receiver; receiving a packet loss information message returned by the data receiver; the packet loss information message is generated by the data receiver when it determines that packet loss has occurred, and only one packet loss information message is sent for each lost target data; based on the packet loss information message, the lost target data is retransmitted to the data receiver. In this way, only one packet loss information message is sent to notify the data sender for the same lost packet, reducing the additional bandwidth occupation of the packet loss information message and reducing the reverse load.

[0105] In one optional embodiment, after sending the target data to the data receiver, the method further includes: starting a timeout timer corresponding to the target data, the timeout timer having a duration equal to the historical maximum RTT; and if the timeout timer expires and no feedback message is received from the data receiver, resending the target data to the data receiver, the feedback message being an acknowledgment message or a packet loss information message.

[0106] In the fast selective retransmission scheme, to reduce the flood of lost packet information, the receiving end only sends a lost packet information packet once each time it detects packet loss. To prevent lost packet information packets from being discarded by the switch, a timer needs to be used on the sending end to detect packet loss. The timer's timeout has a significant impact on performance: a timeout set too short can lead to misjudging network packet loss due to slight fluctuations in RTT, resulting in invalid retransmissions; a timeout set too long can cause packet loss to be retransmitted for a long time, during which no packets are sent, greatly reducing bandwidth utilization. A more reasonable approach is to estimate whether the current packet should be discarded based on the historical maximum round-trip time MAX_RTT (i.e., the maximum RTT).

[0107] In some optional embodiments, the method further includes: obtaining the current RTT based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message; obtaining the current RTT smoothing value based on the current RTT and the previous RTT smoothing value; determining the maximum RTT based on each historical RTT and the current RTT; and obtaining the historical maximum RTT based on the current RTT smoothing value and the maximum RTT.

[0108] RTT can be obtained by adding a timestamp TS1 to the data packet when the data is sent, recording the current time TS2 when the acknowledgment packet ACK is received, and then subtracting TS1 from TS2 to get the current RTT. The current RTT smoothing value SRTT = α * previous RTT smoothing value SRTT + (1-α) * current RTT, where α takes the value between (0, 1), and the historical maximum RTT = max(maximum RTT, current RTT smoothing value).

[0109] The timeout timer is used to prevent the loss of packet information messages. If no feedback message is received from the data receiver after the timeout timer expires, it is determined that the target data has been lost, and the data sender needs to resend the target data to the data receiver.

[0110] In one optional embodiment, after retransmitting the target data to the data receiver and retransmitting the lost target data, the process includes: resetting (resetting) the timeout timer corresponding to the retransmitted target data. Resetting the timeout timer means resetting the timeout timer to restart its countdown.

[0111] In one optional embodiment, the method further includes: upon receiving an acknowledgment message, canceling the timeout timer of the target message corresponding to the acknowledgment message.

[0112] The feedback messages include acknowledgment messages or packet loss information messages. That is, after each target data is sent, the corresponding timeout timer starts counting. If an acknowledgment message for the target data is received, the target data is sent successfully and the timeout timer is canceled. If a packet loss information message for the target data is received, the timeout timer is reset. If no feedback message is received and the timeout timer expires, the target data corresponding to the timeout timer is resent and the timeout timer is reset.

[0113] In one optional embodiment, retransmitting the target data to the data receiver and retransmitting the lost target data to the data receiver includes: setting the priority of the target data that needs to be retransmitted to a first target priority, and retransmitting the target data to the data receiver based on the first target priority.

[0114] When the data sending end receives a packet loss information message, it immediately initiates retransmission, raises the priority of the target data to the highest level, and skips the message scheduling queue to send the target data directly.

[0115] In one optional embodiment, before retransmitting the lost target data to the data receiver, the process includes: determining the mode of the packet loss information message based on the mode bits of the packet loss information message; if the mode of the packet loss information message is to identify packet loss by bit bits, reading the number of valid bits from the packet loss information message, and parsing the packet loss information message based on the number of valid bits to determine the target data that needs to be retransmitted.

[0116] In one optional embodiment, before retransmitting the lost target data to the data receiver, the process includes: determining the mode of the packet loss information message based on the mode bits of the packet loss information message; if the mode of the packet loss information message is not to identify packet loss by bits, parsing the packet loss information message to obtain the minimum and maximum identifiers of the successfully received target data; and determining the target data that needs to be retransmitted based on the minimum and maximum identifiers of the successfully received target data.

[0117] Packet loss can be categorized into congestion loss and fault loss. Congestion loss typically occurs when the switch's queue length exceeds its configured loss threshold. However, this doesn't mean all packets will be dropped after exceeding the threshold. To maintain performance, the switch will randomly drop packets. Figure 5 As shown; however, packet loss due to faults (such as a loose optical module) is uncontrollable and will drop a large number of packets over a period of time, such as... Figure 6 As shown.

[0118] For congestion packet loss, there is no problem using traditional bitmap messages for recording and sending. However, for fault packet loss, since the duration of the fault is uncontrollable, many messages may have been dropped in the middle. If traditional bitmap messages are still used in this case, on the one hand, the bitmap may be too large or even insufficient. On the other hand, the data sending end also needs to check bit by bit to know how many messages have been dropped, which is very inefficient.

[0119] To avoid the aforementioned problems, this application, in addition to the traditional 128 bits of the packet loss information message, also uses two fields, the mode bit and the bmp_len effective bit, to further explain the packet loss information message:

[0120] When the mode bit is 0: the information represented is the same as a traditional bitmap. Each bit in each bitmap represents a message, with a bit of 0 indicating a lost packet and a bit of 1 indicating a received packet. The valid bit bit bmp_len indicates the number of valid bitmaps, avoiding wasting time checking all bits. This mode cannot be properly displayed when the number of consecutively dropped messages exceeds the maximum length of the bitmap.

[0121] When the mode bit is 1: bitmap (0~31) represents the minimum PSN of the received packets, and bitmap (32~63) represents the maximum PSN of the received packets. In this case, regardless of how many packets are lost consecutively, the loss information can be easily presented, and when parsing the packet loss information, it is easy to identify which packets were lost and which were not, without needing to poll each packet individually, saving a significant amount of time.

[0122] Table 1: Packet Loss Information Message Format

[0123]

[0124] In one exemplary embodiment, such as Figure 7 As shown, a data retransmission method is provided, which is applied to... Figure 1 Taking the data receiving end as an example, the explanation includes the following steps 702 to 706. Wherein:

[0125] S702: Receives target data sent by the data sender.

[0126] The data receiver can receive the target data sent by the data sender. The target data may be lost during transmission as it passes through the switch.

[0127] S704: If packet loss is determined based on the target data, and the lost target data is not notified to the data sender, a packet loss information message is generated for the lost target data.

[0128] The data sending end identifies each target data to be sent to the data receiving end to determine the sending order of the target data. In this way, the data receiving end can determine whether there is packet loss based on the identification of each target data. For example, if the data receiving end should receive data in the order of data 1, data 2, data 3, but the data receiving end actually receives data 1 and data 3, then when data 3 is received, it is determined that data 2 is lost.

[0129] A packet loss information message is generated when the data receiver determines that packet loss has occurred, and the lost target data has not been notified to the data sender. This packet loss information message is used to indicate the lost target data. Optionally, the packet loss information message includes an identifier of the lost target data, or it includes an identifier of the unlost target data, or it includes both the identifier of the lost target data and the identifier of the unlost target data.

[0130] After receiving the target data, the data receiving end first determines whether there is packet loss based on the identifier of the previously received target data and the identifier of the currently received target data. If packet loss exists and the lost target data has not been notified to the data sending end, a packet loss information message is generated and sent to the data receiving end.

[0131] S706: Send a packet loss information message to the data sender, and send the packet loss information message only once for each lost target data. The packet loss information message is used to instruct the data sender to resend the lost target data to the data receiver.

[0132] After generating a packet loss information message, the data receiving end sends the packet loss information message to the data sending end, and only sends a packet loss information message once for each lost target data.

[0133] After receiving a packet loss information message, the data sender parses the message to determine the lost target data and then retransmits the lost target data.

[0134] In the above embodiments, by sending only one packet loss information message to the data sender for the same packet loss, the additional bandwidth usage of the packet loss information message is reduced, and the reverse load is lowered.

[0135] In one optional embodiment, the method further includes: receiving retransmitted target data sent by the data sender, wherein the retransmitted target data is retransmitted to the data receiver when the data sender started a timeout timer when the target data was last sent, and when the timeout timer expires and no feedback message corresponding to the target data is received, the timeout timer duration is the historical maximum RTT, and the feedback message is an acknowledgment message or a packet loss information message.

[0136] In one optional embodiment, the historical maximum RTT is determined based on the current RTT smoothing value and the maximum RTT. The current RTT smoothing value is determined based on the current RTT and the previous RTT smoothing value. The maximum RTT is determined based on each historical RTT and the current RTT. The current RTT is determined based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message.

[0137] The limitation on the historical maximum RTT can be found above and will not be repeated here. In this application, in order to avoid the loss of packet loss information messages, a timeout timer is also introduced. Each time the data sender sends the target data, the corresponding timeout timer is triggered. If no feedback message is received from the data receiver after the timeout timer expires, it is determined that the target data has been lost, and the data sender needs to resend the target data to the data receiver.

[0138] The feedback messages include acknowledgment messages or packet loss information messages. That is, after each target data is sent, the corresponding timeout timer starts counting. If an acknowledgment message for the target data is received, the target data is sent successfully and the timeout timer is canceled. If a packet loss information message for the target data is received, the timeout timer is reset. If no feedback message is received and the timeout timer expires, the target data corresponding to the timeout timer is resent and the timeout timer is reset.

[0139] In one optional embodiment, sending a packet loss information message to the data sender includes: setting the priority of the packet loss information message to a second target priority, and sending the packet loss information message to the data sender based on the second target priority.

[0140] In this embodiment, after detecting packet loss, the data receiving end prioritizes the packet loss information message to the highest level, skips the message scheduling queue, and directly sends the packet loss information message to the data sending end.

[0141] In one optional embodiment, before sending the packet loss information message to the data sender, the method further includes: generating a first mode bit for the packet loss information message corresponding to the congestion packet loss if the packet loss situation is congestion packet loss; determining the number of lost packets and determining the number of valid bits based on the number of lost packets; determining the value of the bits in the packet loss information message based on the lost data; and obtaining the packet loss information message based on the first mode bit, the number of valid bits, and the value of the bits.

[0142] In one optional embodiment, before sending the packet loss information message to the data sender, the method further includes: generating a second mode bit for the packet loss information message corresponding to the faulty packet loss when the packet loss is faulty packet loss; determining the minimum and maximum identifiers of the successfully received target data based on the lost data; and obtaining the packet loss information message based on the second mode bit, the minimum and maximum identifiers of the successfully received target data.

[0143] The specific limitations regarding the format of packet loss information messages can be found above and will not be repeated here.

[0144] To enable those skilled in the art to better understand this application, in conjunction with Figure 8 As shown, Figure 8 This is a framework diagram of data retransmission in one embodiment. In this embodiment, the data retransmission method includes:

[0145] 1) The first transmit doorbell interface module (TDI, referred to as the first TDI module) of the data sending end schedules one of the QPs (queue pare) from the first management module of the host queue (HQM, referred to as the first HQM module for distinction) to send the target data.

[0146] 2) The first sending module (TXP) of the data sending end sorts the target data, marks each target data with a PSN tag, and updates the timeout timer. The timeout timer duration is the historical maximum RTT.

[0147] 3) After receiving the target data, the second receiving module (RXP) of the data receiving end determines whether there is packet loss by identifying the PSN: if there is no packet loss, it notifies the second management module of the data receiving end that the QP has an acknowledgment message ACK that needs to be sent; otherwise, it skips the second management module and the second transmission doorbell interface module and directly notifies the second sending module (TXP) to send a packet loss information message.

[0148] 4) When the first receiving module (RXP) of the data sending end receives a normal message (a request message or an ACK message sent by the data receiving end), it notifies the first management module to process the response. If a packet loss information message is received, it directly notifies the first sending module (TXP) to retransmit, without queuing in the first management module or waiting for scheduling by the first transmission doorbell interface module. Regardless of whether an ACK message or a packet loss information message is received, the timer needs to be reset to avoid timeout.

[0149] 5) If no ACK or packet loss message is received after the timeout timer expires, the corresponding target data will be retransmitted quickly.

[0150] In the above data retransmission method, the data receiving end sends the packet loss information back to the data sending end through only one packet loss information message for the target data that was dropped by the switch, thereby reducing the additional bandwidth occupation of a large number of packet loss information messages.

[0151] Lost packet information messages are sent directly without queuing at the data receiving end, and are also sent directly without queuing when the data sending end retransmits the target data, reducing retransmission queuing latency and improving retransmission efficiency.

[0152] The lost information carried in the packet loss message will not be repeated. After receiving the packet loss message, the data sender does not need to consume additional resources to calculate redundant information, nor does it need to consume additional memory to save historical retransmission information, thus saving computing and storage resources.

[0153] When the data sender retransmits, it only needs to extract the information from the packet loss message to directly retransmit, without the need for complex calculation and storage modules. The timer module can reuse the original timeout mechanism, only adding an interface to dynamically configure the timeout time based on the RTT result; the RTT detection module reuses the RTT detection of the CC (congestion control) congestion algorithm without adding complex implementation.

[0154] Different patterns of packet loss information messages can easily handle different packet loss patterns (congestion packet loss and fault packet loss).

[0155] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0156] Based on the same inventive concept, this application also provides a data retransmission apparatus for implementing the data retransmission method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more data retransmission apparatus embodiments provided below can be found in the limitations of the data retransmission method described above, and will not be repeated here.

[0157] In one exemplary embodiment, such as Figure 9 As shown, a data retransmission device is provided, comprising: a first transmitting module 901, a first receiving module 902, and a retransmission module 903, wherein:

[0158] The first transmitting module 901 is used to transmit target data to the data receiving end;

[0159] The first receiving module 902 is used to receive the packet loss information message returned by the data receiving end; the packet loss information message is generated by the data receiving end for the lost target data when it determines that packet loss exists, and only one packet loss information message is sent for each lost target data.

[0160] The retransmission module 903 is used to resend the lost target data to the data receiver based on the packet loss information message.

[0161] In one optional embodiment, the above device further includes: a timer module, used to start a timeout timer corresponding to the target data, the timeout timer having a timing duration equal to the historical maximum RTT; and, if the timeout timer expires and no feedback message is received from the data receiver, to resend the target data to the data receiver, the feedback message being an acknowledgment message or a packet loss information message.

[0162] In one optional embodiment, the retransmission module 903 is specifically used to set the priority of the target data that needs to be retransmitted to a first target priority, and retransmit the target data to the data receiving end based on the first target priority.

[0163] In one optional embodiment, the timer module is specifically used to reset the timeout timer corresponding to the retransmitted target data.

[0164] In one optional embodiment, the timer module is specifically used to cancel the timeout timer of the target message corresponding to the acknowledgment message upon receiving the acknowledgment message.

[0165] In one optional embodiment, the above apparatus further includes: a retransmission data detection module, configured to determine the mode of the packet loss information message based on the mode bits of the packet loss information message; when the mode of the packet loss information message is packet loss identified by bit bits, read the number of valid bits from the packet loss information message, and parse the packet loss information message based on the number of valid bits to determine the target data that needs to be retransmitted.

[0166] In one optional embodiment, the above apparatus further includes: a retransmission data detection module, configured to determine the mode of the packet loss information message based on the mode bits of the packet loss information message; if the mode of the packet loss information message is not to identify packet loss by bit bits, to parse the packet loss information message to obtain the minimum and maximum identifiers of the successfully received target data; and to determine the target data that needs to be retransmitted based on the minimum and maximum identifiers of the successfully received target data.

[0167] In one optional embodiment, the above apparatus further includes: an RTT update module, configured to obtain the current RTT based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message; obtain the current RTT smoothing value based on the current RTT and the previous RTT smoothing value; determine the maximum RTT based on each historical RTT and the current RTT; and obtain the historical maximum RTT based on the current RTT smoothing value and the maximum RTT.

[0168] In one exemplary embodiment, such as Figure 10 As shown, a data retransmission device is provided, comprising: a second receiving module 1001, a lost information message generation module 1002, and a second sending module 1003, wherein:

[0169] The second receiving module 1001 is used to receive target data sent by the data sending end;

[0170] The packet loss information generation module 1002 is used to generate a packet loss information message for the lost target data when packet loss is determined based on the target data and the lost target data is not notified to the data sender.

[0171] The second sending module 1003 is used to send a packet loss information message to the data sending end, and only sends a packet loss information message once for each lost target data. The packet loss information message is used to instruct the data sending end to resend the lost target data to the data receiving end.

[0172] In one optional embodiment, the second receiving module 1001 is further configured to receive retransmitted target data sent by the data sending end. The retransmitted target data is sent to the data receiving end when the data sending end last sent the target data, after a timeout timer is started and the timeout timer expires and no feedback message corresponding to the target data is received. The timeout timer duration is the historical maximum RTT, and the feedback message is an acknowledgment message or a packet loss information message.

[0173] In one optional embodiment, the historical maximum RTT is determined based on the current RTT smoothing value and the maximum RTT. The current RTT smoothing value is determined based on the current RTT and the previous RTT smoothing value. The maximum RTT is determined based on each historical RTT and the current RTT. The current RTT is determined based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message.

[0174] In one optional embodiment, the second sending module 1003 is specifically used to set the priority of the packet loss information message to a second target priority, and send the packet loss information message to the data sending end based on the second target priority.

[0175] In one optional embodiment, the aforementioned lost information message generation module 1002 is further configured to, when the packet loss situation is congestion packet loss, generate a first mode bit of a lost packet information message corresponding to congestion packet loss; determine the number of lost packets and determine the number of valid bits based on the number of lost packets; determine the value of the bits of the lost packet information message based on the lost data; and obtain the lost packet information message based on the first mode bit, the number of valid bits, and the value of the bits.

[0176] In one optional embodiment, the aforementioned lost information message generation module 1002 is further configured to generate a second mode bit of a lost packet information message corresponding to the faulty lost packet when the packet loss is a faulty lost packet; determine the minimum and maximum identifiers of the successfully received target data based on the lost data; and obtain a lost packet information message based on the second mode bit, the minimum and maximum identifiers of the successfully received target data.

[0177] Each module in the aforementioned data retransmission device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0178] In one exemplary embodiment, a chip is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described in any of the above embodiments.

[0179] In one exemplary embodiment, a network interface card is provided, including a chip as described in any of the above embodiments and multiple interfaces, wherein the chip processes data or communicates externally through the interfaces.

[0180] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus. The communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The communication interface can also process data or perform external communication via a network interface card. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a data retransmission method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0181] Those skilled in the art will understand that Figure 11The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0182] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0183] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0184] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0185] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0186] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0187] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0188] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A data retransmission method, characterized in that, Applied to a data sending end, the method includes: sending target data to a data receiving end and starting a timeout timer corresponding to the target data; Receive the packet loss information message returned by the data receiving end; the packet loss information message is generated by the data receiving end for the lost target data when it determines that packet loss exists, and only one packet loss information message is sent for each lost target data. Based on the packet loss information message, the lost target data is resent to the data receiving end, and the timeout timer corresponding to the resent target data is reset; The mode of the packet loss information packet is determined based on the mode bits of the packet loss information packet; When the packet loss information message pattern is characterized by identifying packet loss through bits, the number of valid bits is read from the packet loss information message, and the packet loss information message is parsed based on the number of valid bits to determine the target data that needs to be retransmitted. In the case of congestion packet loss, the packet loss information message pattern is characterized by identifying packet loss through bits. If the packet loss information message does not use bit-based identification of packet loss, the packet loss information message is parsed to obtain the minimum and maximum identifiers of the successfully received target data; based on the minimum and maximum identifiers of the successfully received target data, the target data that needs to be retransmitted is determined, wherein, in the case of fault-based packet loss, the packet loss information message does not use bit-based identification of packet loss.

2. The method according to claim 1, characterized in that, The timeout timer's duration is the historical maximum RTT; After sending the target data to the data receiving end, the process also includes: If the timeout timer expires and no feedback message is received from the data receiving end, the target data is resent to the data receiving end. The feedback message is either an acknowledgment message or a packet loss information message.

3. The method according to claim 1 or 2, characterized in that, The step of retransmitting the target data to the data receiving end, and retransmitting the lost target data to the data receiving end, includes: The priority of the target data that needs to be retransmitted is set to the first target priority, and the target data is retransmitted to the data receiving end based on the first target priority.

4. The method according to claim 2, characterized in that, After retransmitting the target data to the data receiving end, the process includes: Reset the timeout timer corresponding to the retransmitted target data.

5. The method according to claim 2, characterized in that, The method further includes: Upon receiving the acknowledgment message, cancel the timeout timer for the target message corresponding to the acknowledgment message.

6. The method according to claim 2, characterized in that, The method further includes: The current RTT is obtained based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message; The current RTT smoothing value is obtained based on the current RTT and the previous RTT smoothing value; Determine the maximum RTT based on historical RTTs and the current RTT; Based on the current RTT smoothing value and the maximum RTT, the historical maximum RTT is obtained.

7. A data retransmission method, characterized in that, Applied to a data receiving end, the method includes: The data sender receives target data sent by the data sender. After sending the target data, the data sender also starts a timeout timer corresponding to the target data. If packet loss is determined based on the target data, and the lost target data is not notified to the data sender, a packet loss information message is generated for the lost target data. The packet loss information message is sent to the data sending end, and only one packet loss information message is sent for each lost target data. The packet loss information message is used to instruct the data sending end to resend the lost target data to the data receiving end, and to reset the timeout timer corresponding to the resent target data. In the case of congestion-induced packet loss, a first mode bit of a packet loss information message corresponding to the congestion-induced packet loss is generated; Determine the number of packet losses, and based on the number of packet losses, determine the number of valid bits; The value of the bits in the packet loss information message is determined based on the lost data; Based on the first mode bit, the number of effective bits, and the value of the bit, a packet loss information message is obtained; In the case of packet loss being a faulty packet loss, a second mode bit of a packet loss information message corresponding to the faulty packet loss is generated; Determine the minimum and maximum identifiers of the successfully received target data based on the lost data; Based on the second mode bit, the minimum and maximum identifiers of the successfully received target data, a packet loss information message is obtained.

8. The method according to claim 7, characterized in that, The method further includes: The receiving end receives retransmitted target data sent by the data sending end. The retransmitted target data is sent to the data receiving end when the timeout timer at the data sending end expires and no feedback message corresponding to the target data is received. The timeout timer duration is the historical maximum RTT, and the feedback message is an acknowledgment message or a packet loss information message.

9. The method according to claim 8, characterized in that, The historical maximum RTT is determined based on the current RTT smoothing value and the maximum RTT. The current RTT smoothing value is determined based on the current RTT and the previous RTT smoothing value. The maximum RTT is determined based on each historical RTT and the current RTT. The current RTT is determined based on the reception time of the latest acknowledgment message and the transmission time of the target data corresponding to the latest acknowledgment message.

10. The method according to claim 7, characterized in that, Sending the packet loss information message to the data sending end includes: The priority of the packet loss information message is set to the second target priority, and the packet loss information message is sent to the data sending end based on the second target priority.

11. A data retransmission device, characterized in that, The device, applied to a data transmitting end, includes: The first sending module is used to send target data to the data receiving end and start a timeout timer corresponding to the target data; The first receiving module is used to receive the packet loss information message returned by the data receiving end; the packet loss information message is generated by the data receiving end for the lost target data when it determines that packet loss exists, and only one packet loss information message is sent for each lost target data. The retransmission module is used to retransmit the lost target data to the data receiving end based on the packet loss information message, and reset the timeout timer corresponding to the retransmitted target data; The retransmission data detection module is used to determine the mode of the packet loss information packet based on the mode bits of the packet loss information packet; when the mode of the packet loss information packet is packet loss identified by bit bits, it reads the number of valid bits from the packet loss information packet and parses the packet loss information packet based on the number of valid bits to determine the target data that needs to be retransmitted, wherein in the case of congestion packet loss, the mode of the packet loss information packet is packet loss identified by bit bits; when the mode of the packet loss information packet is not packet loss identified by bit bits, it parses the packet loss information packet to obtain the minimum and maximum identifiers of the successfully received target data; based on the minimum and maximum identifiers of the successfully received target data, it determines the target data that needs to be retransmitted, wherein in the case of fault packet loss, the mode of the packet loss information packet is not packet loss identified by bit bits.

12. A data retransmission device, characterized in that, The device, applied to a data receiving end, includes: The second receiving module is used to receive target data sent by the data sending end. After sending the target data, the data sending end also starts a timeout timer corresponding to the target data. The packet loss information generation module is used to generate a packet loss information message for the lost target data when packet loss is determined based on the target data and the lost target data is not notified to the data sending end. The second sending module is used to send the packet loss information message to the data sending end, and only send the packet loss information message once for each lost target data. The packet loss information message is used to instruct the data sending end to resend the lost target data to the data receiving end, and to reset the timeout timer corresponding to the resent target data. The packet loss information generation module is further configured to: generate a first mode bit for a packet loss information message corresponding to the congestion packet loss when the packet loss is congestion packet loss; determine the number of lost packets and determine the number of valid bits based on the number of lost packets; determine the value of the bits in the packet loss information message based on the lost data; obtain a packet loss information message based on the first mode bit, the number of valid bits, and the value of the bits; generate a second mode bit for a packet loss information message corresponding to the fault packet loss when the packet loss is fault packet loss; determine the minimum and maximum identifiers of the successfully received target data based on the lost data; and obtain a packet loss information message based on the second mode bit, the minimum and maximum identifiers of the successfully received target data.

13. A chip comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

14. A network interface card, characterized in that, Includes the chip as described in claim 13 and multiple interfaces. The chip processes data or communicates with the outside world through the interface.

15. A computer device, characterized in that, Includes the network interface card as described in claim 14, the network interface card being used for processing data or external communication.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

17. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

Images