Queue processing method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a queue processing method and device, electronic equipment and a readable storage medium, and relates to the technical field of data processing, and the method comprises the steps: monitoring a queue deficit value and data packet integrity of a data packet queue set in a first scheduling period; determining a target congestion queue from the data packet queues of which the queue deficit value is greater than a preset deficit value and the data packet integrity is false; and reducing the deficit value of the target congestion queue based on the deficit adjustment step length corresponding to the target congestion queue. According to the embodiment of the invention, the target congestion queue with congestion can be determined from the data packet queue set according to the deficit value of the queue and the integrity of the data packet, and the deficit value of the target congestion queue with congestion is reduced in each scheduling period, so that the speed of the data packet queue set meeting the deficit resetting condition can be improved; the transmission rate of the network connection based on the data packet queue set can be improved, and the transmission delay can be reduced.

Description

Technical Field

[0001] The present invention belongs to the field of data processing technology, and in particular relates to a queue processing method, device, electronic device and readable storage medium. Background Art

[0002] With the continuous development of Internet technology, various devices (such as cloud servers, terminal devices, etc.) need to transmit a large number of network packets. In order to solve the problem of link congestion, packet queue-based packet sending technology has emerged. This technology can put all data sent by an interface into multiple packet queues and send packets according to the priority of each packet queue.

[0003] In the related art, a queue deficit value is usually assigned to each data packet queue. Each data packet queue consumes the corresponding queue deficit value when sending a data packet. When the queue deficit value of a data packet queue is too small, the scheduling of the data packet queue is stopped. When the queue deficit values ​​of all data packet queues are low, a deficit reset operation is triggered to reset the queue deficit values ​​of all data packet queues and start the next round of scheduling.

[0004] If the deficit value of a packet queue cannot be consumed in time due to sudden situations such as link congestion, it will be difficult to trigger the deficit reset operation, affecting the data transmission of other packet queues and easily causing the transmission efficiency of the entire packet queue set. Summary of the invention

[0005] The present invention provides a queue processing method, device, electronic device and readable storage medium to solve the technical problem of low transmission efficiency of data packet queue in related technologies.

[0006] In a first aspect, the present invention provides a queue processing method, the method comprising:

[0007] Monitoring the queue deficit value and packet integrity of the packet queue set in the first scheduling cycle;

[0008] Determining a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false;

[0009] Based on the deficit adjustment step corresponding to the target congested queue, the queue deficit value of the target congested queue is reduced to trigger a deficit reset operation for the set of data packet queues in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

[0010] Optionally, determining the target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false comprises:

[0011] Determine a first queue from the set of data packet queues, the first queue having a queue deficit value greater than a preset deficit value;

[0012] When the number of queues of the first queues is less than or equal to a preset number, a target congested queue is determined from the first queues based on the integrity of data packets of the first queues.

[0013] Optionally, determining a target congested queue from the first queue based on the integrity of the data packets of the first queue includes:

[0014] In the case that there is no complete data packet in the first queue, obtaining a first missing data amount of the data packets in the first queue;

[0015] When the first missing data amount is less than or equal to a preset data amount, obtaining a second missing data amount of the data packet after waiting for a preset time period;

[0016] When the second missing data amount is greater than 0, the first queue is determined as the target congested queue.

[0017] Optionally, the method further comprises:

[0018] Determine the complete packet length of the data packet based on the packet header information of the data packet in the first queue;

[0019] A first missing data amount of the data packet is determined according to an actual packet length of the data packet and the complete packet length.

[0020] Optionally, the method further comprises:

[0021] Obtaining deficit reset condition information of the data packet queue set;

[0022] The preset deficit value is determined based on the deficit reset condition information.

[0023] Optionally, reducing the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue includes:

[0024] Determining a target correction coefficient based on the number of continuous scheduling cycles of the target congested queue; wherein the number of continuous scheduling cycles represents the number of scheduling cycles in which the target congested queue continuously serves as the target congested queue;

[0025] The queue deficit value is reduced based on the deficit adjustment step size and the target correction factor.

[0026] Optionally, the method further comprises:

[0027] Obtaining a maximum transmission unit value and a queue weight corresponding to the target congested queue;

[0028] A deficit adjustment step size of the target congested queue is calculated based on the maximum transmission unit value and the queue weight.

[0029] In a second aspect, the present invention provides a queue processing device, the device comprising:

[0030] A monitoring module, used for monitoring the queue deficit value and data packet integrity of the data packet queue set in the first scheduling cycle;

[0031] A queue module, configured to determine a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false;

[0032] A reduction module is used to reduce the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue, so as to trigger a deficit reset operation for the set of data packet queues in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

[0033] Optionally, the queue module includes:

[0034] A first queue submodule, configured to determine a first queue having a queue deficit value greater than a preset deficit value from the set of data packet queues;

[0035] The target congestion queue submodule is used to determine a target congestion queue from the first queue based on the integrity of the data packets of the first queue when the queue quantity of the first queue is less than or equal to a preset quantity.

[0036] Optionally, the target congestion queue submodule includes:

[0037] A first missing data amount unit, configured to obtain a first missing data amount of data packets in the first queue when there is no complete data packet in the first queue;

[0038] A second missing data amount unit, configured to obtain a second missing data amount of the data packet after waiting for a preset time period if the first missing data amount is less than or equal to a preset data amount;

[0039] The target congestion queue determining unit is configured to determine the first queue as the target congestion queue when the second missing data amount is greater than 0.

[0040] Optionally, the device further comprises:

[0041] A complete packet length module, used to determine the complete packet length of the data packet based on the packet header information of the data packet in the first queue;

[0042] The first missing data amount determining module is used to determine the first missing data amount of the data packet according to the actual packet length of the data packet and the complete packet length.

[0043] Optionally, the device further comprises:

[0044] A condition acquisition module, used to acquire deficit reset condition information of the data packet queue set;

[0045] The preset deficit value module is used to determine the preset deficit value based on the deficit reset condition information.

[0046] Optionally, the reduction module includes:

[0047] A target correction coefficient submodule, configured to determine a target correction coefficient based on the number of continuous scheduling cycles of the target congestion queue; wherein the number of continuous scheduling cycles represents the number of scheduling cycles during which the target congestion queue continuously serves as the target congestion queue;

[0048] The queue deficit value reduction submodule is used to reduce the queue deficit value based on the deficit adjustment step size and the target correction coefficient.

[0049] Optionally, the device further comprises:

[0050] A queue weight module, used to obtain a maximum transmission unit value and a queue weight corresponding to the target congested queue;

[0051] The deficit adjustment step size module is used to calculate the deficit adjustment step size of the target congested queue based on the maximum transmission unit value and the queue weight.

[0052] In a third aspect, the present invention provides an electronic device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned queue processing method when executing the program.

[0053] In a fourth aspect, the present invention provides a readable storage medium, which, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the above-mentioned queue processing method.

[0054] In an embodiment of the present invention, the method includes: monitoring the queue deficit value and packet integrity of the packet queue of the packet queue set in the first scheduling cycle; determining the target congested queue from the packet queues whose queue deficit value is greater than the preset deficit value and whose packet integrity is false; reducing the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue to trigger the deficit reset operation for the packet queue set in the second scheduling cycle; wherein the second scheduling cycle is later than the first scheduling cycle. The target congested queue that is congested can be determined from the packet queue set according to the queue deficit value and packet integrity, and the queue deficit value of the target congested queue that is congested can be reduced in each scheduling cycle, which can increase the speed at which the packet queue set meets the deficit reset condition, thereby reducing the time when the packet queue set is congested, and is helpful to increase the transmission rate of the network connection based on the packet queue set and reduce the transmission delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 is a flow chart of steps of a queue processing method provided by an embodiment of the present invention;

[0057] Figure 2 is a schematic diagram of a scheduling process provided by an embodiment of the present invention;

[0058] Figure 3 is a flowchart of another queue processing method provided by an embodiment of the present invention;

[0059] Figure 4 It is a schematic diagram of an architecture provided by an embodiment of the present invention;

[0060] Figure 5 is a structural diagram of a queue processing device provided by an embodiment of the present invention;

[0061] Figure 6 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Reference Figure 1 , Figure 1 1 is a flow chart of the steps of a queue processing method provided by an embodiment of the present invention. Figure 1 As shown, the method may include:

[0064] Step 101 : monitoring queue deficit values ​​and data packet integrity of data packet queues of a set of data packet queues in a first scheduling period.

[0065] In order to achieve the purpose of improving data transmission efficiency, increasing data transmission speed, reducing data transmission delay, improving service quality (QoS), etc., a data packet queue set can be set for the data transmission subject, and multiple data packet queues in the data packet queue set are used to carry the data packets that the data transmission subject needs to send, and then the data packets are sent through the multiple data packet queues. Among them, the data transmission subject may include but is not limited to an interface, a process, an application, a port, a router, a switch, a terminal device, etc., and the embodiment of the present invention does not specifically limit it.

[0066] The data packet queue set may be composed of multiple data packet queues, and the data packet queues may be used to store data packets of different priorities, different types or different sources. In the embodiment of the present invention, the types of data packet queues may include but are not limited to priority queues (Priority Queue), weighted fair queues (WFQ, Weighted Fair Queue), Earliest Deadline First Queue (Earliest Deadline First Queue), etc., which are not specifically limited in the embodiment of the present invention.

[0067] Network devices or systems usually schedule or process data packet queues intermittently, sending data packets in the data packet queues during each scheduling or processing process. The above mechanism is also used to process the data packet queue set. All data packet queues in the data packet queue set are processed uniformly during each scheduling or processing period. Each scheduling or processing process of the data packet queue set can be called a scheduling cycle.

[0068] The queue deficit value (Deficit) can be used to characterize the data flow limit of the data packet queue. In an embodiment of the present invention, each data packet queue may have a corresponding deficit counter in which the queue deficit value corresponding to the queue is stored. The queue deficit value of the data packet queue can be initialized to the maximum number of bytes allowed during a scheduling period, thereby realizing the data flow limit of the data packet queue. It should be noted that the method of initializing the queue deficit value may vary depending on the scheduling strategy adopted. For example, if the Deficit Round Robin scheduling strategy is adopted, the queue deficit value can be initialized to the maximum transmission unit (Maximum Transmission Unit, MTU) of the transmission interface corresponding to the data packet queue, or other fixed values; if the Weighted Deficit Round Robin scheduling strategy is adopted, the queue deficit value can be initialized to the product of the MTU of the transmission interface corresponding to the data packet queue and the queue weight (Weight), or other dynamic values, which are not specifically limited in the embodiment of the present invention.

[0069] Each data packet queue will initialize its queue deficit value at the beginning of each round of scheduling. The initialization value of the queue deficit value can be flexibly set according to actual business needs (for example, 0 or the quantization value of the data packet queue). In each scheduling cycle, the corresponding quantization value (Quantum) can be added to the queue deficit value of each data packet queue. The quantization values ​​of different data packet queues can be the same or different. In each scheduling cycle, if the queue deficit value of the data packet queue is greater than or equal to the set value (for example, 0), the data packets in the data packet queue can be sent, and the queue deficit value of the data packet queue can be reduced by the size of the data packets sent. If a data packet queue does not send any data packets in the current scheduling cycle (for example, the queue is empty or the queue deficit value is insufficient), its queue deficit value will be retained until the next round. Through the above operations, flow control of each data packet queue can be achieved.

[0070] In an embodiment of the present invention, the integrity of a data packet can indicate whether the data packets in the data packet queue are complete. The integrity of the data packet can also be indicated by the data packet status. The data packet status can include two states: a complete data packet is included and a complete data packet is not included. In one embodiment, since some protocols use specific flags or terminators to indicate the boundaries of data packets, for example, HTTP1.1 can indicate the end of a data packet by the termination flag n0, the integrity of the data packet can be determined by detecting the termination flag in the data packet. If a complete data packet exists in the data packet queue, the data packet integrity of the data packet queue can be true. In another embodiment, the data packet can also be verified to obtain a verification result, and whether the data packet is complete can be determined based on the verification result. For example, if the data packet transmission protocol uses CRC verification, the data packet can be verified to obtain a CRC verification result. If the verification result indicates that the verification is successful, it means that the data packet is complete. If the data packet queue does not contain a complete data packet, the data packet integrity of the data packet queue can be false.

[0071] It should be noted that the technicians can flexibly set the method for obtaining the integrity of the data packet according to the actual business needs, data packet transmission protocol, etc., and the embodiment of the present invention does not specifically limit it. Due to different scheduling strategies, in each scheduling cycle, the scheduled data packet queues in the data packet queue set can be all, or one or part. For example, in the polling scheduling strategy, only one data packet queue may be scheduled in each scheduling cycle. Therefore, in the embodiment of the present invention, only the scheduled data packet queues in the scheduling cycle can be detected, and the subsequent steps can be performed, that is, the target congestion queue is determined from the scheduled data packet queues.

[0072] Step 102 : determining a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false.

[0073] In the embodiment of the present invention, whether the data packet queue is used as a target congestion queue is determined based on the queue deficit value and data packet integrity of the data packet queue, that is, the target congestion queue is determined from the combination of the data packet queues.

[0074] In the usual scheduling strategy, if there are no complete data packets that can be sent in the data packet queue due to data congestion, untimely data transmission, etc., the data packet queue will not be processed in this scheduling cycle, and no data packets will be sent from the data packet queue, so the queue deficit value of the data packet queue will not be consumed, which may easily lead to the queue deficit values ​​of other data packet queues in the data packet queue set being exhausted and unable to be reset and replenished for a long time, resulting in the entire data packet queue being unable to send data packets, which may easily cause network delays and network congestion.

[0075] In an embodiment of the present invention, the target congested queue where data congestion occurs can be accurately located from a set of data packet queues by using queue deficit value and data packet integrity. Specifically, in one embodiment, a data packet queue whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false can be determined as a target congested queue. In another embodiment, a data packet queue whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false can also be determined as a target congested queue.

[0076] Step 103, based on the deficit adjustment step corresponding to the target congested queue, reducing the queue deficit value of the target congested queue to trigger a deficit reset operation for the packet queue set in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

[0077] In the embodiment of the present invention, the difference between the deficit adjustment step corresponding to the target congested queue and the queue deficit value of the target congested queue can be calculated to determine the target deficit value for adjusting the queue deficit value of the target congested queue, and then the queue deficit value of the target congested queue is reset to the target deficit value. The target deficit value is less than the above-mentioned queue deficit value. The queue deficit value can also be directly subtracted from the deficit adjustment step to update the queue deficit value, which is not specifically limited in the embodiment of the present invention.

[0078] Specifically, the target deficit value of the target congested queue can be obtained by subtracting the deficit adjustment step corresponding to the target congested queue from the current queue deficit value of the target congested queue. The above-mentioned deficit adjustment step indicates the amplitude that the target congested queue needs to be adjusted in each scheduling cycle. The deficit adjustment step can be preset to a fixed value (e.g., 100), and all data packet queues can share the same deficit adjustment step; the deficit adjustment step can also be set according to the bandwidth limit of the data packet queue, so that different data packet queues can correspond to different deficit adjustment step, which is not specifically limited in the embodiment of the present invention.

[0079] Exemplarily, in the current scheduling cycle, if it is detected that there are no complete data packets in data packet queue A, and the queue deficit value of data packet queue A is high, then data packet queue A can be determined as the target congestion queue. For all target congestion queues determined in the current scheduling cycle, the corresponding queue deficit value and the deficit adjustment step size can be used to make a difference to obtain the target deficit value of each target congestion queue. For example, the queue deficit value of data packet queue A in the current scheduling cycle is 300, and the deficit adjustment step size corresponding to data packet queue A is 100, then its corresponding target deficit value can be calculated as 300-100=200.

[0080] After the target deficit value of the target congestion queue is calculated, the corresponding target deficit value can be applied to the target congestion queue, that is, the queue deficit value of the target congestion queue is adjusted to the target deficit value, thereby realizing the operation of reducing the queue deficit value based on the deficit adjustment step.

[0081] After the current scheduling cycle ends and the queue deficit values ​​of all target congested queues are reduced, the next scheduling cycle can be entered. The above operations are repeated in the next scheduling cycle to determine a new target congested queue and reduce the queue deficit value of the target congested queue until the data packet queue set reaches the deficit reset condition and the deficit reset operation is performed on the data packet queues.

[0082] The deficit reset condition may include that the number of queues in the data packet queue set whose queue deficit value is greater than the preset deficit value may be 0, or may be other integer values ​​less than the number of queues in the data packet queue set. For example, if a data packet queue set includes 10 data packet queues, and the preset deficit value is set to 0, as scheduling continues, only 0 data packet queues with queue deficit values ​​greater than 0 remain in the 10 data packet queues, then the deficit reset condition is met, and the deficit reset operation for the data packet queue set may be triggered.

[0083] In one embodiment, the above-mentioned deficit reset operation may include: setting the queue deficit values ​​of all data packet queues in the data packet queue set to an initial value, and the initial values ​​corresponding to each data packet queue may be the same or different. For example, the initial value may be the maximum transmission unit (Maximum Transmission Unit, MTU) of the transmission interface corresponding to the data packet queue, and the initial value may also be the maximum transmission unit value of the transmission interface corresponding to the data packet queue, or the product of the maximum transmission unit value and the queue weight, which is not specifically limited in the embodiment of the present invention.

[0084] In another embodiment, the above-mentioned deficit resetting operation may include: setting the queue deficit value of all data packet queues in the data packet queue set to the sum of the current queue deficit value and the initial value, that is, retaining the current queue deficit value to the next round of scheduling, thereby further improving the flow control accuracy of each data packet queue. Specifically, the queue deficit value Deficit may be reset using the following formula: Deficit+=Weight*MTU.

[0085] Reference Figure 2 , Figure 2 A schematic diagram of a scheduling process provided by an embodiment of the present invention is shown. Figure 2As shown, it is assumed that the packet queue set has 3 packet queues, and the data traffic ratio allocated to these 3 packet queues is 5:3:2. It can be simplified to understand that in a round of 10 scheduling cycles, packet queue 1 consumes its queue deficit value in 5 scheduling cycles, packet queue 2 consumes its queue deficit value in 3 scheduling cycles, and packet queue 3 consumes its queue deficit value in 2 scheduling cycles. If the data transmission of packet queue 1 is normal, the packet integrity is true, and the data transmission of packet queues 2 and 3 is congested, then in the first 5 scheduling cycles, packet queue 1 is scheduled, and the queue deficit value of packet queue 1 is consumed to be less than or equal to the preset deficit value. In the 6th to 7th scheduling cycles, the packet integrity of packet queues 2 and 3 is false, and there is no complete data packet available for scheduling. Then, the 6th to 7th scheduling cycles are in an idle state, and the queue deficit values ​​of packet queues 2 and 3 can be reduced. If the packet integrity of packet queue 2 becomes true in the 8th scheduling cycle, and the queue deficit value of packet queue 2 is greater than the preset deficit value 1, then packet queue 2 can still send packets. Assuming that it lasts for 1 scheduling cycle, the 8th scheduling cycle is occupied by packet queue 2, and the queue deficit value of packet queue 2 is less than or equal to the preset deficit value after the end of the 8th scheduling cycle. Starting from the 9th scheduling cycle, the packet integrity of packet queue 3 becomes true, and the packets in packet queue 3 can continue to be scheduled. Until the packet integrity of packet queue 3 becomes false in the 11th scheduling cycle, the queue deficit value of packet queue 3 will continue to be reduced in the 12th scheduling cycle, triggering the queue deficit reset of the packet queue set and starting the next round of scheduling.

[0086] Therefore, this solution can not only speed up the speed of resetting the queue deficit value of each data packet queue when the data packet queue is blocked, but also schedule and send its data packets in time when the data packet integrity change of the data packet queue that is congested in the middle is true. Compared with the solution of directly resetting the queue deficit value of all data packet queues when the data packet queue is congested, it can effectively balance the traffic of each data packet queue and avoid the situation that the congested data packet queue is difficult to schedule to a certain extent.

[0087] In summary, an embodiment of the present invention provides a queue processing method, including: monitoring the queue deficit value and packet integrity of a packet queue set in a first scheduling cycle; determining a target congested queue from a packet queue whose queue deficit value is greater than a preset deficit value and whose packet integrity is false; reducing the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue to trigger a deficit reset operation for the packet queue set in a second scheduling cycle; wherein the second scheduling cycle is later than the first scheduling cycle. The target congested queue that is congested can be determined from the packet queue set based on the queue deficit value and packet integrity, and the queue deficit value of the target congested queue that is congested can be reduced in each scheduling cycle, which can increase the speed at which the packet queue set meets the deficit reset condition, thereby reducing the time when the packet queue set is congested, and helping to increase the transmission rate of the network connection based on the packet queue set and reduce the transmission delay.

[0088] Reference Figure 3 , Figure 3 FIG. 1 is a flowchart of another queue processing method provided by an embodiment of the present invention. Figure 3 As shown, the method may include:

[0089] Step 201 : monitoring the queue deficit value and data packet integrity of the data packet queues of the data packet queue set in the first scheduling period.

[0090] This step can refer to step 101, and will not be described in detail in this embodiment of the present invention.

[0091] In addition, the queue deficit value and / or packet integrity of each data packet queue can be provided autonomously by each data packet queue, or can be determined by a queue scheduling policy, or can be determined by an independent process or program outside the queue scheduling policy, which is not specifically limited in the embodiments of the present invention.

[0092] Step 202: Determine a first queue from the set of data packet queues, the first queue having a queue deficit value greater than a preset deficit value.

[0093] If the queue deficit value of a packet queue is less than or equal to the preset deficit value, it means that the packet queue has consumed all the allocated traffic and should be in a waiting state for traffic redistribution. For a packet queue whose queue deficit value is greater than the preset deficit value, it has not consumed all the traffic quota and may be a target congestion queue that is congested.

[0094] In an embodiment of the present invention, in order to improve execution efficiency, a queue deficit value-related judgment may be first performed on each data packet queue in the data packet queue set, and a first queue whose queue deficit value is greater than a preset deficit value is determined from the data packet queue set. The first queue represents a data packet queue whose queue deficit value is greater than the preset deficit value in the data packet queue set.

[0095] Optionally, the preset deficit value may be calculated based on the following steps A1 to A2:

[0096] Step A1, obtaining deficit reset condition information of the data packet queue set.

[0097] In the embodiment of the present invention, different packet queue combinations can adopt different scheduling strategies, and different scheduling strategies (such as Deficit Round Robin scheduling strategy, Weighted Deficit Round Robin scheduling strategy, etc.) have different deficit reset conditions. In order to improve the adaptability of this solution and make the determination of the target congestion queue more accurate, the deficit reset condition information of the packet queue set can be obtained, and the corresponding preset deficit value can be determined according to the deficit reset condition information.

[0098] Among them, the deficit reset condition information may include but is not limited to a deficit reset condition identifier, a scheduling strategy identifier corresponding to the deficit reset condition, a condition definition of the deficit reset condition, etc., and the embodiment of the present invention does not make specific limitations.

[0099] Step A2: determining the preset deficit value based on the deficit reset condition information.

[0100] In the embodiment of the present invention, a corresponding preset deficit value can be determined according to the deficit reset condition information, and the preset deficit value can be the queue deficit lower limit for triggering the deficit reset of the packet queue set. Therefore, the packet queues that do not meet the deficit reset condition can be selected by the preset deficit value to reduce the queue deficit value, which helps to improve the efficiency and accuracy of the deficit reset operation corresponding to the deficit reset condition used to trigger the current packet queue set.

[0101] Step 203: When the number of queues in the first queues is less than or equal to a preset number, determine a target congested queue from the first queues based on the integrity of the data packets in the first queues.

[0102] The packet queues (not the first queue) in the packet queue set that have consumed all the traffic quota will be in a waiting state for traffic allocation and cannot participate in packet transmission. Therefore, the more packet queues that have consumed all the traffic quota, the greater the cost of congestion in the first queue.

[0103] Therefore, after determining the first queue, the number of queues of the first queue included in the data packet queue set can be further determined. If the number of queues of the first queue is less than or equal to the preset number (for example, 1), it means that there are more data packet queues in the data packet queue set that have consumed the traffic quota. At this time, the target congestion queue can be further determined from the first queue. If the number of queues of the first queue is greater than the preset number, it means that there are fewer data packet queues in the data packet queue set that have consumed the traffic quota. Even if congestion occurs in the first queue, it will not cause serious impact. The current scheduling cycle can be skipped and the target congestion queue will not be determined, thereby improving the stability of the operation of the data packet queue set.

[0104] It should be noted that the above step 203 can be performed only after the first queue is determined from the data packet queue set, that is, it is performed only when the first queue exists in the data packet queue set, so the queue number of the first queue is a positive integer greater than 0.

[0105] In an embodiment of the present invention, when the number of queues in the first queue is less than or equal to a preset number, a target congested queue can be determined from the first queue based on the packet integrity of the first queue. This can avoid performing packet integrity judgment in each cycle, which helps to improve the execution efficiency of the scheme. It can also judge whether the potential congestion of the first queue can seriously affect the network based on the number of queues in the first queue, and skip the subsequent determination step of the target congested queue in the scheduling cycle where the impact is not serious, which helps to further improve the execution efficiency of the scheme and the operation stability of the packet queue set.

[0106] Optionally, step 203 may include:

[0107] Sub-step 2031: when there is no complete data packet in the first queue, obtain a first missing data amount of the data packets in the first queue.

[0108] In order to further improve the data transmission efficiency, the data packet integrity check can be performed on the first queue. If there is no complete data packet in the first queue, that is, the first queue has not completed the complete reception of a data packet, the first missing data amount of the data packet in the first queue can be obtained, and whether to further determine the target congested queue is determined according to the first missing data amount. The first missing data amount can represent the amount of data missing from the incomplete data packet in the first queue.

[0109] It should be noted that, when the first queue contains multiple incomplete data packets, the missing data amount of the data packet closest to the complete data packet can be determined as the first missing data amount. When the first queue is empty, this step can be skipped.

[0110] Optionally, the first missing data amount may be calculated based on the following steps B1 to B2:

[0111] Step B1, determining the complete packet length of the data packet based on the packet header information of the data packet in the first queue.

[0112] In an embodiment of the present invention, in order to improve the efficiency of detecting the first amount of missing data of a data packet in the first queue, the complete packet length of the data packet can be determined based on the header information of the data packet in the first queue. Since the first queue can be a queue for different scenarios and different needs, the types of data packets stored therein may also be different. For example, common data packets at the link layer may include but are not limited to Ethernet frames, Wi-Fi data packets, etc., and common data packets at the network layer may include but are not limited to IPv4 data packets, IPv6 data packets, etc. Therefore, the data packet type can be first determined based on the header information of the data packet, and then the storage format of the packet length information can be determined based on the data packet type, and the header information can be parsed according to the storage format to obtain the complete packet length of the data packet.

[0113] Step B2: determining a first missing data amount of the data packet according to the actual packet length and the complete packet length of the data packet.

[0114] In the embodiment of the present invention, the actual packet length of the data packet can be obtained, and the actual packet length indicates the amount of data actually occupied by the data packet in the data packet queue. The actual packet length can be determined by monitoring the data packet flow of the data packet queue, or can be obtained by actually measuring the space occupied by the current data packet, which is not specifically limited in the embodiment of the present invention.

[0115] The first missing data amount of the data packet can be calculated based on the actual packet length and the complete packet length of the data packet. Specifically, the packet length difference between the complete packet length and the actual packet length can be calculated to obtain the first missing data amount. It should be noted that the second missing data amount in the embodiment of the present invention can also be calculated based on the same or similar method as the above calculation of the first missing data amount, and the embodiment of the present invention will not be repeated.

[0116] By determining the complete packet length of the data packet based on the packet header information of the data packet in the first queue and determining the first missing data amount of the data packet according to the actual packet length and the complete packet length of the data packet, the accuracy of calculating the missing data amount can be improved to a certain extent.

[0117] Sub-step 2032: When the first missing data amount is less than or equal to a preset data amount, obtain a second missing data amount of the data packet after waiting for a preset time period.

[0118] In the embodiment of the present invention, a preset data volume can be set, and the preset data volume can be used to measure whether the incomplete data packet in the first queue is about to be transmitted, that is, to determine whether a complete data packet is about to exist in the first queue. If a complete data packet is about to exist in the first queue, waiting can be performed, waiting for a complete data packet to appear in the first queue, thereby triggering a normal data packet sending process of the first queue, so that the queue deficit value of the first queue is naturally reduced.

[0119] Specifically, when the first missing data amount is less than or equal to the preset data amount, after waiting for a preset time, the missing data amount of the data packet corresponding to the first missing data amount can be obtained again to obtain the second missing data amount. The acquisition time of the second missing data amount is later than the acquisition time of the first missing data amount. The preset time can be flexibly set by the technician according to actual business needs, for example, 0.1ms, and the embodiment of the present invention does not specifically limit it.

[0120] Sub-step 2033: when the second missing data amount is greater than 0, determining the first queue as the target congested queue.

[0121] Since the second missing data amount and the first missing data amount belong to the same data packet, if the data packet is transmitted within the above-mentioned preset time length, the second missing data amount obtained is 0, and if it is still not transmitted within the preset time length, the second missing data amount obtained is greater than 0. Therefore, if the second missing data amount is greater than 0, it means that the first queue still does not have a complete data packet after the preset time length, and its data packet integrity is still false. At this time, it can be determined that the first queue cannot be restored to smooth flow in a short time, and there is a high possibility that congestion will continue to occur. The first queue can be determined as the target congested queue. Specifically, when the second missing data amount is greater than 0, the first queue can be determined as the target congested queue.

[0122] By obtaining a first missing data amount of data packets in the first queue when there is no complete data packet in the first queue; and obtaining a second missing data amount of data packets after waiting for a preset time when the first missing data amount is less than or equal to the preset data amount; and determining the first queue as a target congested queue when the second missing data amount is greater than 0, the determined target congested queue has a greater possibility of congestion, thereby further improving the accuracy of determining the target congested queue.

[0123] Step 204: determining a target correction coefficient based on the number of continuous scheduling cycles of the target congested queue; wherein the number of continuous scheduling cycles represents the number of scheduling cycles in which the target congested queue continuously serves as the target congested queue.

[0124] Since the longer the continuous scheduling cycle number of a packet queue as the target congestion queue is, the more serious the impact of the packet queue on the packet queue set is, which may cause the queue deficit values ​​of other packet queues to be unable to be reset for a long time, the target correction coefficient can be determined based on the continuous scheduling cycle number of the target congestion queue, and then the reduction range of the queue deficit value of the target congestion queue in this scheduling cycle can be adjusted according to the target correction coefficient, so as to reduce the deficit value of the packet queue that has been the target congestion queue for a long time by a large margin. Among them, the continuous scheduling cycle number refers to the number of scheduling cycles in which the target congestion queue is continuously used as the target congestion queue.

[0125] Exemplarily, after scheduling starts, if data packet queue A is determined as the target congestion queue in the first scheduling cycle and the second scheduling cycle, is not the target congestion queue in the third scheduling cycle, and is determined as the target congestion queue in the fourth to sixth scheduling cycles; then in the above-mentioned second scheduling cycle, the number of continuous scheduling cycles of the data packet queue A is 2; in the fourth scheduling cycle, the number of continuous scheduling cycles of the data packet queue A is 1; in the sixth scheduling cycle, the number of continuous scheduling cycles of the target congestion queue is 3.

[0126] Specifically, a preset relationship between the number of continuous scheduling cycles and the correction coefficient can be set, and the preset relationship can be expressed as a function, a table, a model, a rule, etc., which is not specifically limited in the embodiment of the present invention. In the above preset relationship, a smaller number of continuous scheduling cycles can correspond to a smaller correction coefficient, and a larger number of continuous scheduling cycles can correspond to a larger correction coefficient.

[0127] Exemplarily, a preset relationship provided by an embodiment of the present invention is shown as follows:

[0128]

[0129] Wherein, A represents the correction coefficient, and C represents the number of continuous scheduling cycles. When the number of continuous scheduling cycles is 1, the corresponding correction coefficient is 1. The larger the number of continuous scheduling cycles, the larger the corresponding correction coefficient. The value range of the correction coefficient can be [1,2].

[0130] Optionally, the deficit adjustment step size may be calculated through the following steps C1 to C2:

[0131] C1, obtaining a maximum transmission unit value and a queue weight corresponding to the target congested queue.

[0132] In an embodiment of the present invention, different queue weights can be set for different data packet queues. Since there are differences in the amount of bandwidth allocated to data packet queues with different weights, that is, there are differences in the queue deficit values, in order to better maintain the consistency of deficit value reduction for different target congested queues, the maximum transmission unit value of the transmission link served by the target congested queue and the queue weight of the target congested queue can be obtained, and the deficit adjustment step corresponding to the target congested queue is calculated based on the maximum transmission unit value and the queue weight. Among them, the queue weight can be automatically allocated by the scheduling policy used by the data packet queue set, or it can be set by a technician, and the embodiment of the present invention does not specifically limit it.

[0133] C2: Calculate the deficit adjustment step of the target congested queue based on the maximum transmission unit value and the queue weight.

[0134] In an embodiment of the present invention, the maximum transmission unit value can be multiplied by the queue weight of the target congested queue to obtain the deficit adjustment step of the target congested queue. By obtaining the maximum transmission unit value and the queue weight corresponding to the target congested queue; and calculating the deficit adjustment step of the target congested queue based on the maximum transmission unit value and the queue weight, the deficit adjustment step can be adapted to the queue weight of the target congested queue and the link environment in which the target congested queue is located, thereby reducing the queue deficit value of the target congested queue according to the deficit adjustment step to improve the consistency of the reduction effect.

[0135] Step 205: reducing the queue deficit value based on the deficit adjustment step size and the target correction coefficient.

[0136] In the embodiment of the present invention, the queue deficit value of the target congested queue can be reduced by the deficit adjustment step and the target correction coefficient. Specifically, the deficit adjustment step can be multiplied by the target correction coefficient to obtain the target adjustment step, and the queue deficit value can be reduced by the target adjustment step.

[0137] For example, if the deficit adjustment step of a target congestion queue is 10, the queue deficit value is 30, and the target correction coefficient is 1.2, then the reduced queue deficit value can be calculated to be 30-10*1.2=18.

[0138] The target correction coefficient is determined based on the number of continuous scheduling cycles of the target congested queue; the number of continuous scheduling cycles indicates the number of scheduling cycles that the target congested queue has been continuously used as the target congested queue; the queue deficit value is reduced based on the deficit adjustment step and the target correction coefficient. It can speed up the reduction of the queue deficit value for packet queues that have been congested for a long time, help improve the speed at which the packet queue set triggers the queue deficit reset in the case of network congestion, and further optimize the transmission delay and transmission efficiency of the packet queue.

[0139] Reference Figure 4 , Figure 4 A schematic diagram of an architecture provided by an embodiment of the present invention is shown. Figure 4 As shown, this solution can be implemented by modifying the scheduling strategy, or by a process or program independent of the scheduling strategy. The data packets to be transmitted by the data packet queue can be a first-in first-out queue (fifo), that is, each data packet queue can be implemented by a first-in first-out queue. The data packet queue can receive data packet data as input, output a data packet integrity signal pkt_whole_vld, and a queue deficit value signal usedw, as shown in Figure 4 As shown, data packet queue 1 outputs fifo1_usedw, data packet queue 2 outputs fifo2_usedw, and data packet queue 3 outputs fifo3_usedw. The scheduling policy module can receive the output of each data packet queue, and when the data packet integrity signal and the queue deficit value signal output by the data packet queue meet the requirements (that is, the data packet integrity signal is true, and the queue deficit value signal indicates that its queue deficit value is greater than the preset deficit value), output the deficit value reduction signal chnl_grant to the data packet queue to reduce the queue deficit value of the corresponding data packet queue.

[0140] In summary, another queue processing method provided by an embodiment of the present invention includes: monitoring the queue deficit value and packet integrity of a packet queue of a packet queue set in a first scheduling cycle; determining a target congested queue from a packet queue whose queue deficit value is greater than a preset deficit value and whose packet integrity is false; reducing the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue to trigger a deficit reset operation for the packet queue set in a second scheduling cycle; wherein the second scheduling cycle is later than the first scheduling cycle. The target congested queue that is congested can be determined from the packet queue set based on the queue deficit value and packet integrity, and the queue deficit value of the target congested queue that is congested can be reduced in each scheduling cycle, which can increase the speed at which the packet queue set meets the deficit reset condition, thereby reducing the time when the packet queue set is congested, and is helpful to increase the transmission rate of the network connection based on the packet queue set and reduce the transmission delay.

[0141] Reference Figure 5 , Figure 5 is a structural diagram of a queue processing device provided by an embodiment of the present invention. The queue processing device 50 may include:

[0142] A monitoring module 51, configured to monitor a queue deficit value and packet integrity of a data packet queue set in a first scheduling period;

[0143] The queue module 52 is used to determine a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false;

[0144] The reduction module 53 is used to reduce the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue to trigger a deficit reset operation for the data packet queue set in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

[0145] Optionally, the queue module includes:

[0146] A first queue submodule, configured to determine a first queue having a queue deficit value greater than a preset deficit value from the set of data packet queues;

[0147] The target congestion queue submodule is used to determine a target congestion queue from the first queue based on the integrity of the data packets of the first queue when the queue quantity of the first queue is less than or equal to a preset quantity.

[0148] Optionally, the target congestion queue submodule includes:

[0149] A first missing data amount unit, configured to obtain a first missing data amount of data packets in the first queue when there is no complete data packet in the first queue;

[0150] A second missing data amount unit, configured to obtain a second missing data amount of the data packet after waiting for a preset time period if the first missing data amount is less than or equal to a preset data amount;

[0151] The target congestion queue determining unit is configured to determine the first queue as the target congestion queue when the second missing data amount is greater than 0.

[0152] Optionally, the device further comprises:

[0153] A complete packet length module, used to determine the complete packet length of the data packet based on the packet header information of the data packet in the first queue;

[0154] The first missing data amount determining module is used to determine the first missing data amount of the data packet according to the actual packet length of the data packet and the complete packet length.

[0155] Optionally, the device further comprises:

[0156] A condition acquisition module, used to acquire deficit reset condition information of the data packet queue set;

[0157] The preset deficit value module is used to determine the preset deficit value based on the deficit reset condition information.

[0158] Optionally, the reduction module includes:

[0159] A target correction coefficient submodule, configured to determine a target correction coefficient based on the number of continuous scheduling cycles of the target congestion queue; wherein the number of continuous scheduling cycles represents the number of scheduling cycles during which the target congestion queue continuously serves as the target congestion queue;

[0160] The queue deficit value reduction submodule is used to reduce the queue deficit value based on the deficit adjustment step size and the target correction coefficient.

[0161] Optionally, the device further comprises:

[0162] A queue weight module, used to obtain a maximum transmission unit value and a queue weight corresponding to the target congested queue;

[0163] The deficit adjustment step size module is used to calculate the deficit adjustment step size of the target congested queue based on the maximum transmission unit value and the queue weight.

[0164] In summary, the queue processing device provided in the embodiment of the present invention can determine the target congested queue where congestion occurs from the data packet queue set based on the queue deficit value and the data packet integrity, and reduce the queue deficit value of the target congested queue where congestion occurs in each scheduling cycle, which can increase the speed at which the data packet queue set meets the deficit reset condition, thereby reducing the time when congestion occurs in the data packet queue set, helping to improve the transmission rate of the network connection based on the data packet queue set and reduce the transmission delay.

[0165] Reference Figure 6 , the present invention also provides a structural diagram of an electronic device, such as Figure 6 As shown, it includes: a processor 1101, a memory 1102, and a computer program 11021 stored in the memory and executable on the processor, and the processor implements the queue processing method of the aforementioned embodiment when executing the program.

[0166] The present invention also provides a readable storage medium, when the instructions in the storage medium are executed by the processor of the electronic device, the electronic device can execute the queue processing method of the aforementioned embodiment.

[0167] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0168] It should be noted that the various information and data obtained in the embodiments of the present invention are all obtained with the authorization of the information / data holder.

[0169] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0170] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0171] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0172] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0173] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the sorting device according to the present invention. The present invention may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0174] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

[0175] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0176] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

[0177] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A queue processing method, characterized in that: The method comprises: Monitoring the queue deficit value and packet integrity of the packet queue set in the first scheduling cycle; Determining a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false; Based on the deficit adjustment step corresponding to the target congested queue, the queue deficit value of the target congested queue is reduced to trigger a deficit reset operation for the set of data packet queues in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

2. The method according to claim 1, characterized in that The step of determining a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false comprises: Determine a first queue from the set of data packet queues, the first queue having a queue deficit value greater than a preset deficit value; When the number of queues of the first queues is less than or equal to a preset number, a target congested queue is determined from the first queues based on the integrity of data packets of the first queues.

3. The method according to claim 2, characterized in that The determining a target congested queue from the first queue based on the integrity of the data packets of the first queue includes: In the case that there is no complete data packet in the first queue, obtaining a first missing data amount of the data packets in the first queue; When the first missing data amount is less than or equal to a preset data amount, obtaining a second missing data amount of the data packet after waiting for a preset time period; When the second missing data amount is greater than 0, the first queue is determined as the target congested queue.

4. The method according to claim 3, characterized in that The method further comprises: Determine the complete packet length of the data packet based on the packet header information of the data packet in the first queue; A first missing data amount of the data packet is determined according to an actual packet length of the data packet and the complete packet length.

5. The method according to claim 2, characterized in that: The method further comprises: Obtaining deficit reset condition information of the data packet queue set; The preset deficit value is determined based on the deficit reset condition information.

6. The method according to claim 1, characterized in that: The reducing the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue includes: Determining a target correction coefficient based on the number of continuous scheduling cycles of the target congested queue; wherein the number of continuous scheduling cycles represents the number of scheduling cycles in which the target congested queue continuously serves as the target congested queue; The queue deficit value is reduced based on the deficit adjustment step size and the target correction factor.

7. The method according to claim 1, characterized in that The method further comprises: Obtaining a maximum transmission unit value and a queue weight corresponding to the target congested queue; A deficit adjustment step size of the target congested queue is calculated based on the maximum transmission unit value and the queue weight.

8. A queue processing device, characterized in that: The device comprises: A monitoring module, used for monitoring the queue deficit value and data packet integrity of the data packet queue set in the first scheduling cycle; A queue module, configured to determine a target congested queue from the data packet queues whose queue deficit value is greater than a preset deficit value and whose data packet integrity is false; A reduction module is used to reduce the queue deficit value of the target congested queue based on the deficit adjustment step corresponding to the target congested queue, so as to trigger a deficit reset operation for the set of data packet queues in a second scheduling period; wherein the second scheduling period is later than the first scheduling period.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program.

10. A readable storage medium, characterized in that: When the instructions in the readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any one of the methods described in claims 1-7.

Images