Resource adjustment method and apparatus, network device, and storage medium

Abstract

The present application belongs to the field of communications, and discloses a resource adjustment method and apparatus, a network device, and a storage medium. The method comprises: acquiring the current traffic load rate of a target network entity, the current traffic load rate indicating load conditions when the target network entity forwards current network traffic, and the target network entity being a plurality of interface boards, a plurality of switching network boards, or any one interface board among the plurality of interface boards; on the basis of the current traffic load rate, determining a resource adjustment policy corresponding to the current network traffic, the resource adjustment policy indicating adjustment conditions of traffic forwarding resources in a network device when the current network traffic is forwarded, and the traffic forwarding resources comprising a serial link of a switching network board and an interface board; and on the basis of the resource adjustment policy, adjusting the traffic forwarding resources in the network device. According to the present application, traffic forwarding resources are dynamically adjusted, so that the overall power consumption of the network device changes along with changes in network traffic, and the power demand and heat dissipation requirements of the network device under a low-traffic load are reduced.

Description

Resource adjustment method, device, network equipment and storage medium

[0001] This application claims priority to Chinese patent application number 202311161639.3, filed on September 6, 2023, entitled “Resource Adjustment Method, Apparatus, Network Equipment and Storage Medium,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present application relates to the field of communications, and in particular to a resource adjustment method, apparatus, network equipment, and storage medium. Background Art

[0003] Network devices are used to forward network traffic within a network. Typically, network devices include multiple interface boards and multiple switching fabric boards (SFMs). Each interface board is connected to the SFMs via multiple serial links, enabling network traffic forwarding between different interface boards. Therefore, the more SFMs a network device has, the more network traffic it can forward. Similarly, the more serial links a given interface board has, the more network traffic it can forward.

[0004] In theory, the more network traffic a network device forwards, the higher its traffic load ratio, and the greater its power consumption. Conversely, the less network traffic a network device forwards, the lower its traffic load ratio, and the less power it consumes. The traffic load ratio indicates the load on the network device when forwarding network traffic.

[0005] However, under extremely low traffic load conditions, network devices may still have significant heat dissipation and power requirements. Therefore, reducing the power consumption of network devices has become an urgent issue that needs to be addressed.

[0006] Summary of the Invention

[0007] This application provides a resource adjustment method, apparatus, network device, and storage medium that can dynamically adjust traffic forwarding resources in a network device based on the current traffic load rate of a target network entity, thereby adjusting the overall power consumption of the network device in response to changes in network traffic, thereby reducing the power and heat dissipation requirements of the network device under low traffic loads. The technical solution is as follows.

[0008] In a first aspect, a resource adjustment method is provided, which is applied to a resource manager in a network device, wherein the network device also includes multiple interface boards and multiple switching network boards, and each interface board is connected to at least one switching network board via at least one serial link, and the switching network board is used to forward the network traffic of one interface board to another interface board; the method includes: obtaining the current traffic load rate of the target network entity, the current traffic load rate indicates the load condition of the target network entity when forwarding the current network traffic, and the target network entity is the multiple interface boards, the multiple switching network boards, or any one of the multiple interface boards; based on the current traffic load rate, determining the resource adjustment policy corresponding to the current network traffic, the resource adjustment policy indicates the adjustment condition of the traffic forwarding resources in the network device when forwarding the current network traffic, and the traffic forwarding resources include the serial links of the switching network board and the interface board; based on the resource adjustment policy, adjusting the traffic forwarding resources in the network device.

[0009] As can be seen, the resource manager determines the resource adjustment policy corresponding to the current network traffic based on the current traffic load rate of the target network entity, and then adjusts the traffic forwarding resources in the network device used to forward the current network traffic based on the resource adjustment policy. In this way, the traffic forwarding resources in the network device are dynamically adjusted according to the current traffic load rate of the target network entity, so that the overall power consumption of the network device can change with the changes in the current network traffic. While meeting the current network traffic forwarding requirements, it avoids running a large number of traffic forwarding resources under low traffic load conditions, reducing the power and heat dissipation requirements of the network device under low traffic load conditions.

[0010] Optionally, the target network entity includes the multiple interface boards or the multiple switching network boards, and the resource adjustment strategy includes a first adjustment strategy and a second adjustment strategy, the first adjustment strategy indicates reducing the number of enabled switching network boards in the network device, and the second adjustment strategy indicates increasing the number of enabled switching network boards in the network device; determining the resource adjustment strategy corresponding to the current network traffic based on the current traffic load rate includes: obtaining a first historical traffic load rate of the target network entity; if the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment strategy is the first adjustment strategy; if the current traffic load rate is greater than the first historical traffic load rate, determining that the resource adjustment strategy is the second adjustment strategy.

[0011] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the first adjustment strategy, determining a first single board load rate interval corresponding to the current traffic load rate from multiple single board load rate intervals, and the multiple single board load rate intervals respectively correspond to a number of network boards; according to the first number of network boards corresponding to the first single board load rate interval, at least one switching network board in the network device that is in an open state is shut down, so that the number of switching network boards that have not been shut down in the network device is equal to the first number of network boards.

[0012] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the second adjustment strategy, determining the second single board load rate interval corresponding to the current traffic load rate from multiple single board load rate intervals, and the multiple single board load rate intervals respectively correspond to one network board number; according to the second network board number corresponding to the second single board load rate interval, at least one switching network board in the network device that is in a closed state is turned on, so that the number of switching network boards in the network device that are in an turned-on state is equal to the second network board number.

[0013] Optionally, the multiple single-board load rate intervals are obtained based on the overall ability of the multiple interface boards to process network traffic, or the multiple single-board load rate intervals are obtained based on the overall ability of the multiple switching network boards to process network traffic; wherein the division granularity of the multiple single-board load rate intervals is greater than the reference granularity.

[0014] It can be seen that when the target network entity includes multiple interface boards or multiple switching boards, the resource manager can dynamically adjust the number of enabled switching boards in the network device based on the changing trend of the current traffic load rate of the target network entity. While meeting the current network traffic forwarding requirements, the overall power consumption of the network device is reduced from the dimension of the switching boards.

[0015] Optionally, the interface board includes at least one interface chip, the switching network board includes at least one switching network chip, and the switching network chip is used to forward the network traffic of one interface chip to another interface chip; the target network entity includes the multiple interface boards or the multiple switching network boards, and the resource adjustment strategy includes a third adjustment strategy and a fourth adjustment strategy, the third adjustment strategy indicates to reduce the number of enabled switching network chips in the network device, and the fourth adjustment strategy indicates to increase the number of enabled switching network chips in the network device; the resource adjustment strategy corresponding to the current network traffic is determined based on the current traffic load rate, including: obtaining the first historical traffic load rate of the target network entity; if the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment strategy is the third adjustment strategy; if the current traffic load rate is greater than the first historical traffic load rate, determining that the resource adjustment strategy is the fourth adjustment strategy.

[0016] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the third adjustment strategy, determining a first chip load rate interval corresponding to the current traffic load rate from multiple chip load rate intervals, and the multiple chip load rate intervals respectively correspond to a chip quantity; according to the first chip quantity corresponding to the first chip load rate interval, at least one switching network chip in the network device that is in an turned-on state is shut down, so that the number of switching network chips in the network device that are in an turned-on state is equal to the first chip quantity.

[0017] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the fourth adjustment strategy, determining the second chip load rate interval corresponding to the current traffic load rate from multiple chip load rate intervals, and the multiple chip load rate intervals respectively correspond to a chip quantity; according to the second chip quantity corresponding to the second chip load rate interval, at least one switching network chip in the closed state in the network device is turned on, so that the number of switching network chips in the turned-on state in the network device is equal to the second chip quantity.

[0018] Optionally, the multiple chip load rate intervals are obtained based on the ability of the interface chips included in the multiple interface boards to process network traffic, or the multiple chip load rate intervals are obtained based on the ability of the switching network chips included in the multiple switching network boards to process network traffic; wherein the division granularity of the chip load rate intervals is smaller than the reference granularity.

[0019] It can be seen that when the target network entity includes multiple interface boards or multiple switching network boards, the resource manager dynamically adjusts the number of enabled switching network chips in the network device based on the changing trend of the current traffic load rate of the target network entity. While meeting the current network traffic forwarding requirements, the overall power consumption of the network device is reduced from the dimension of the switching network chip.

[0020] It's important to note that adjusting the number of enabled switch fabric boards is related to adjusting the number of enabled switch fabric chips. This means dynamically adjusting the number of enabled switch fabric chips can be used to dynamically adjust the number of enabled switch fabric boards in a network device. Enabling a switch fabric board enables at least one switch fabric chip on that switch fabric board, while disabling a switch fabric board disables all switch fabric chips on that switch fabric board. Compared to adjusting resources at the switch fabric board level, adjusting resources at the switch fabric chip level provides finer granularity.

[0021] Optionally, the method also includes: obtaining service level agreement (SLA) information corresponding to the current network traffic; if the SLA information meets the SLA requirements corresponding to the current network traffic, executing the step of adjusting the traffic forwarding resources in the network device based on the resource adjustment policy.

[0022] In other words, whether to adjust the number of enabled SFUs or switching chips in a network device depends on whether the current network traffic meets the SLA requirements. If the SLA information for the current network traffic meets the corresponding SLA requirements, the number of enabled SFUs or switching chips in the network device is adjusted according to the resource adjustment policy determined above. This ensures normal forwarding of service traffic while adjusting the network device's traffic forwarding resources.

[0023] Optionally, the method further includes: if the SLA information does not meet the SLA requirement corresponding to the current network traffic, opening at least one switching network board or switching network chip in the network device that is in a closed state according to the target unit opening amount.

[0024] That is, when the SLA information of the current network traffic does not meet the corresponding SLA requirements, the number of enabled switching boards or switching chips in the network device is dynamically adjusted according to the target unit enable quantity, so that the network device can meet the SLA requirements when forwarding the current network traffic.

[0025] Optionally, the target network entity includes a target interface board, and the target interface board is any one of the multiple interface boards. The resource adjustment strategy includes a fifth adjustment strategy and a sixth adjustment strategy, the fifth adjustment strategy indicates reducing the number of serial links of the target interface board that are opened, and the sixth adjustment strategy indicates increasing the number of serial links of the target interface board that are opened; determining the resource adjustment strategy corresponding to the current network traffic based on the current traffic load rate includes: obtaining the second historical traffic load rate of the target network entity; if the current traffic load rate is less than the second historical traffic load rate, determining that the resource adjustment strategy is the fifth adjustment strategy; if the current traffic load rate is greater than the second historical traffic load rate, determining that the resource adjustment strategy is the sixth adjustment strategy.

[0026] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the fifth adjustment strategy, determining a first link load rate interval corresponding to the current traffic load rate from multiple link load rate intervals, and the multiple link load rate intervals respectively correspond to a link number; according to the first link number corresponding to the first link load rate interval, at least one serial link in the target interface board that is in an open state is closed, so that the number of serial links in the target interface board that are in an open state is equal to the first link number.

[0027] Optionally, the traffic forwarding resources in the network device are adjusted based on the resource adjustment strategy, including: if the resource adjustment strategy is the sixth adjustment strategy, determining a second link load rate interval corresponding to the current traffic load rate from multiple link load rate intervals, and the multiple link load rate intervals respectively correspond to a link number; according to the second link number corresponding to the second link load rate interval, at least one serial link in the closed state in the target interface board is opened, so that the number of serial links in the opened state in the target interface board is equal to the second link number.

[0028] Optionally, the multiple link load rate intervals are obtained based on the capability of the target interface board to process network traffic.

[0029] It can be seen that when the target network entity includes a target interface board, the resource manager dynamically adjusts the number of open serial links of the target interface board based on the changing trend of the current traffic load rate of the target interface board, so that the serial links in the open state can meet the resource requirements of the target interface board when forwarding the current network traffic, and at the same time reduce the overall power consumption of the network equipment from the dimension of the interface board.

[0030] In a second aspect, a resource adjustment device is provided, which has the function of implementing the resource adjustment method in the first aspect. The resource adjustment device includes at least one module, which is used to implement the resource adjustment method provided in the first aspect.

[0031] In a third aspect, a network device is provided, comprising a resource manager, a main control board, multiple interface boards, and multiple switching network boards, each interface board being connected to at least one switching network board via at least one serial link, the switching network board being configured to forward network traffic from one interface board to another interface board;

[0032] The resource manager is provided on the main control board or at least one of the plurality of interface boards, and is used to implement the resource adjustment method described in the first aspect.

[0033] In a fourth aspect, a computer-readable storage medium is provided, wherein the storage medium stores a computer program. When the computer program runs on the network device, the network device executes the resource adjustment method described in the first aspect.

[0034] In a fifth aspect, a computer program product comprising instructions is provided. When the instructions are executed on a network device, the network device executes the resource adjustment method described in the first aspect.

[0035] The technical effects obtained in the above-mentioned second, third, fourth and fifth aspects are similar to those obtained by the corresponding technical means in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] FIG1 is a schematic diagram of the structure of a network device provided in an embodiment of the present application;

[0037] FIG2 is a schematic diagram of the structure of another network device provided in an embodiment of the present application;

[0038] FIG3 is a flow chart of a resource adjustment method provided in an embodiment of the present application;

[0039] FIG4 is a schematic diagram of network traffic oscillation at different time scales provided by an embodiment of the present application;

[0040] FIG5 is a schematic structural diagram of a resource adjustment device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0042] Before explaining the resource adjustment method provided in the embodiment of the present application, the application scenario and implementation environment of the embodiment of the present application are first introduced.

[0043] Network devices, such as switches and routers, are used to forward network traffic within a network. Theoretically, when a high volume of network traffic flows through a network device, the overall load on the network device is higher, consuming more power. When a low volume of network traffic flows through the network device, the overall load on the network device is lower, consuming less power. However, in practice, because the main control board, multiple interface boards, and switching fabric boards within the network device are all operating, the power consumption of the network device fluctuates by less than 10% between high and low traffic conditions. Specifically, the power consumption of the network device varies by less than 10% between zero and full traffic conditions. In other words, even when traffic is low, a large number of components within the network device remain powered on, resulting in significant heat dissipation and power requirements.

[0044] Based on this, an embodiment of the present application provides a resource adjustment method, which dynamically adjusts the traffic forwarding resources in the network device by detecting the current traffic load rate of the target network entity in the network device, so that the overall power consumption of the network device changes with the change of network traffic, avoiding the operation of a large number of traffic forwarding resources under low traffic load conditions, and reducing the power demand and heat dissipation requirements of the network device under low traffic load conditions.

[0045] Please refer to Figure 1, which is a schematic diagram of the structure of a network device provided in an embodiment of the present application. The network device includes a main control board 110, multiple interface boards 120, multiple switching network boards 130, and a resource manager 140. Each interface board (also known as a line card or service board) is connected to at least one switching network board via at least one serial link (SerDes). The switching network board is used to forward network traffic from one interface board to another interface board.

[0046] The main control board performs system management, device maintenance, and protocol processing. Multiple interface boards provide various service interfaces, such as POS (packet over SONET / SDH) interfaces (a technology that uses the high-speed transmission channels provided by synchronous optical networking (SONET) or synchronous digital hierarchy (SDH) to directly transmit IP data packets), Gigabit Ethernet (GE) interfaces, and Asynchronous Transfer Mode (ATM) interfaces, and forward network traffic. Multiple switching fabric boards (SFPs) internally forward traffic between the interface boards. The control plane consists of the various control units on the main control board and the control units on the interface boards. The main control board primarily contains three functional units: the system management and control unit, the system clock unit, and the system maintenance unit. The main control board, interface boards, the interface boards, the SFPs, and each other are all connected to the system backplane via a system bus.

[0047] In some embodiments, the interface board includes one or more processors for controlling and managing the interface board, communicating with the central processing unit on the main control board, and forwarding network traffic. The interface board also includes one or more memories for storing forwarding table entries so that the processor can forward network traffic by searching the forwarding table entries. The interface board also includes one or more network interfaces for receiving data streams or information sent by other devices and processing these data streams or information according to the instructions of the processor.

[0048] It should be understood that, as shown in FIG1 , the network device in the embodiment of the present application includes multiple interface boards and adopts a distributed forwarding mechanism. Under this mechanism, the operations on each interface board are basically similar and will not be described in detail for the sake of brevity. In addition, the processor in the interface board can be dedicated hardware or chip, such as a network processor or an application-specific integrated circuit (ASIC) to implement the above functions. This implementation method is what is commonly referred to as the forwarding plane using dedicated hardware or chip processing. Of course, the processor can also be a general-purpose processor, such as a general-purpose central processing unit (CPU) to implement the functions described above, and the embodiment of the present application does not limit this.

[0049] In some embodiments, the resource manager can be set in the main control board, can also be set in any interface board, can also be set separately, and the embodiment of the present application does not limit this. Figure 1 only takes the resource manager as an example of being set in the main control board, and does not constitute a limitation on its setting location.

[0050] Based on this structure, when executing the resource adjustment method provided in the embodiments of the present application, the resource manager obtains the current traffic load rate of the target network entity, determines the resource adjustment policy corresponding to the current network traffic based on the current traffic load rate, and then adjusts the traffic forwarding resources in the network device based on the resource adjustment policy. The target network entity is a plurality of interface boards, a plurality of switching boards, or any one of the plurality of interface boards in the network device, and the traffic forwarding resources include the serial links of the switching boards and interface boards in the network device.

[0051] That is, the resource manager determines the resource adjustment strategy for processing the current network traffic by analyzing the current traffic load rate of multiple interface boards as a whole, the current traffic load rate of multiple switching boards as a whole, or the current traffic load rate of any one interface board, and then dynamically adjusts the number of enabled switching boards in the network device or the number of enabled serial links of at least one interface board in the network device according to the resource adjustment strategy.

[0052] It should be noted that Figure 1 illustrates a network device 100 comprising one main control board (MCB), m interface boards, and n switching fabric boards. In actual applications, the network device 100 may also include multiple MCBs, for example, a master MCB and a backup MCB. Of course, the network device 100 may also include more or fewer interface boards. The greater the data processing capability of the network device, the more interface boards are provided. When multiple interface boards are included, they communicate with each other through at least one switching fabric board, achieving load balancing and redundancy. In a centralized forwarding architecture, the network device may not require a switching fabric board, with multiple interface boards performing the data processing functions of the entire system. In a distributed forwarding architecture, the network device includes multiple interface boards, and data exchange between these multiple interface boards is achieved through the switching fabric boards, providing high-capacity data exchange and processing capabilities. Therefore, the data access and processing capabilities of network devices with distributed architectures are greater than those of network devices with centralized architectures. The specific architecture to be adopted depends on the specific network deployment scenario and is not limited here.

[0053] Based on the network device shown in Figure 1, see Figure 2, when implementing data reception, forwarding and processing, each interface board includes at least one interface chip, and each switching network board includes at least one switching network chip. The switching network chip is used to forward the network traffic of one interface chip to another interface chip.

[0054] An interface chip connects to a switch fabric chip via a serial link, enabling network traffic transmission. Therefore, the number of interface chips and the number of serial links connected to the interface card determine the interface card's ability to process network traffic. Similarly, the number of switch fabric chips and the number of serial links connected to the switch fabric card determine the switch fabric card's ability to forward network traffic.

[0055] Based on this structure, when executing the resource adjustment method provided in the embodiments of this application, the resource manager determines the resource adjustment policy for processing the preceding network traffic by analyzing the current traffic load rate of multiple interface boards as a whole, the current traffic load rate of multiple switching fabric boards as a whole, or the current traffic load rate of any interface board. The resource manager then dynamically adjusts the number of enabled switching fabric chips in the network device according to the resource adjustment policy. This approach provides a finer granularity for adjusting traffic forwarding resources.

[0056] It should be understood that the structure of the network device described in the above example is intended to more clearly illustrate the technical solutions of the embodiments of the present application and does not constitute a limitation on the network device or the technical solutions provided by the embodiments of the present application. A person skilled in the art will appreciate that, as new network devices emerge, the technical solutions provided by the embodiments of the present application will also be applicable to similar technical problems.

[0057] Please refer to FIG. 3 , which is a flowchart of a resource adjustment method provided in an embodiment of the present application. The method is applied to the resource manager 140 shown in FIG. 1 or FIG. 2 , and the method includes the following steps.

[0058] Step 301: Obtain the current traffic load rate of the target network entity, which indicates the load condition of the target network entity when forwarding current network traffic. The target network entity is multiple interface boards, multiple switching network boards, or any one of multiple interface boards.

[0059] In other words, the current traffic load rate is the load rate of the target network entity when it forwards the current network traffic. The load rate is the ratio of the network resources occupied by the target network entity when processing the current network traffic to the target network entity's total network resources. The load rate can reflect the target network entity's ability to carry network traffic.

[0060] As shown in Figure 4, when observing network traffic forwarded by network devices at a microscopic granularity, such as the microsecond (µs) level, the changes are rapid and volatile. However, when observing network traffic at a macroscopic granularity, such as the second or even ten-second level, the changes are relatively stable. Moreover, when processing network traffic, the overall load rate of network devices is also relatively low, and in most cases, the link load rate remains low. Therefore, the current traffic load rate of the target network traffic can be obtained based on the statistical granularity of seconds or ten seconds, and the traffic forwarding resources of the network device can be adjusted accordingly.

[0061] In some embodiments, although the second-level characteristics of network traffic are relatively stable, when network traffic exhibits abnormal jitter, the network device needs to quickly start the corresponding traffic forwarding resources. Therefore, if the second-level characteristics of network traffic are directly used as the basis for adjusting traffic forwarding resources, the delay in the statistics of the traffic load rate of network traffic may cause the duration of the network device wake-up process to not meet the actual processing requirements of the network traffic. Therefore, in order to ensure that the network device can process network traffic in a timely and effective manner when adjusting traffic forwarding resources, the traffic load rate of the target network entity can be obtained based on the millisecond-level statistical granularity, and used as the basis for dynamically adjusting traffic forwarding resources, so as to achieve the effect of dynamically adjusting the power consumption of the network device.

[0062] Based on this, the implementation process of the above step 301 can be: based on the millisecond-level traffic detection results, obtain the current traffic load rate of multiple interface boards as a whole, the current traffic load rate of multiple switching network boards as a whole, or the current traffic load rate of any interface board in the network device.

[0063] It should be noted that the traffic load rate is continuously counted at a millisecond-level statistical granularity, and the resource adjustment period for dynamically adjusting traffic forwarding resources can also be at the millisecond level. When adjusting resources, timely detection and adjustment can be performed; continuous detection and periodic adjustment can also be performed, and this embodiment of the present application does not impose any restrictions on this. For example, the traffic load rate of the target network entity is obtained at a statistical granularity of 1 millisecond, and the statistical results of the traffic load rate are stored. Then, the resource adjustment method provided in the embodiment of the present application is executed according to a resource adjustment period of 10 milliseconds, 100 milliseconds, etc.

[0064] Step 302: Based on the current traffic load rate, determine the resource adjustment strategy corresponding to the current network traffic. The resource adjustment strategy indicates the adjustment of traffic forwarding resources in the network device when forwarding the current network traffic. The traffic forwarding resources include serial links of the switching network board and the interface board.

[0065] It should be understood that network traffic typically flows into or out of a network device from multiple interface boards and is forwarded internally by multiple switching fabric boards within the network device. Therefore, the current traffic load rate of the multiple interface boards as a whole can reflect the load on the multiple switching fabric boards when forwarding network traffic. That is, the current traffic load rate of the multiple interface boards as a whole is positively correlated with the number of enabled switching fabric boards in the network device. The number of enabled switching fabric boards in the network device can be dynamically adjusted based on the current traffic load rates of the multiple interface boards. Similarly, the network traffic forwarding capacity of the switching fabric boards within a network device is positively correlated with the number of enabled switching fabric boards in the network device. Therefore, the current traffic load rates of the multiple switching fabric boards can be used to determine the load on the network device forwarding the current network traffic, thereby dynamically adjusting the number of enabled switching fabric boards in the network device. For a single interface board, its capacity to process network traffic is positively correlated with the number of its enabled serial links. Therefore, the current traffic load rate of a single interface board can be used to determine the load on the interface board processing the current network traffic, thereby dynamically adjusting the number of enabled serial links on that interface board.

[0066] Because the target network entity can be multiple interface boards, multiple switching fabric boards, or any one of multiple interface boards, the implementation method for determining the resource adjustment policy corresponding to the current network traffic based on the current traffic load rate of the target network entity varies depending on the target network entity. These methods are described below.

[0067] In a first implementation, the target network entity includes multiple interface boards or multiple switching network boards. In this case, the resource adjustment policy includes a first adjustment policy and a second adjustment policy, where the first adjustment policy indicates reducing the number of enabled switching network boards in the network device, and the second adjustment policy indicates increasing the number of enabled switching network boards in the network device. Based on this, the implementation process of step 302 can be as follows: obtaining a first historical traffic load rate of the target network entity; if the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment policy is the first adjustment policy; if the current traffic load rate is greater than the first historical traffic load rate, determining that the resource adjustment policy is the second adjustment policy.

[0068] The first historical traffic load rate may be the traffic load rate of the target network entity last counted, or may be the average value of the traffic load rate of the target network entity in the last resource adjustment period, which is not limited in this embodiment of the present application.

[0069] For the target network entity, if the current traffic load rate is less than the first historical traffic load rate, it means that after the last resource adjustment, the switching network boards in the network device that are in an enabled state may have excess resources while meeting the forwarding resource requirements of the current network traffic. Therefore, in order to improve the resource utilization of the network device and reduce the overall power consumption of the network device, the number of enabled switching network boards in the network device can be reduced. Similarly, if the current traffic load rate is greater than the first historical traffic load rate, it means that after the last resource adjustment, the switching network boards in the network device that are in an enabled state may not be able to meet the forwarding resource requirements of the current network traffic. Therefore, it is necessary to increase the number of enabled switching network boards in the network device so that the switching network boards in the network device that are in an enabled state can forward the current network traffic normally.

[0070] That is, the first historical traffic load rate is used as a reference to determine the change trend of the current traffic load rate of the target network entity, thereby judging whether the number of enabled switching fabric units in the network device needs to be increased.

[0071] It should be noted that if the current traffic load rate of the target network entity is equal to the first historical traffic load rate, it means that the switching network boards in the network device that are turned on can meet the forwarding resource requirements of the current network traffic. In this case, the number of turned-on switching network boards in the network device does not need to be adjusted.

[0072] In some embodiments, in order to ensure the normal forwarding of network traffic and avoid large fluctuations in network traffic during the current resource adjustment period, which may cause the switching network board in the turned-on state to be unable to meet the forwarding resource requirements of the actual network traffic, the resource adjustment strategy can also be determined to be the second adjustment strategy when the current traffic load rate is equal to the first historical traffic load rate.

[0073] In this implementation, the resource manager dynamically adjusts the number of enabled switching boards in the network device based on the changing trend of the current traffic load rate of the target network entity. While meeting the current network traffic forwarding requirements, the overall power consumption of the network device is reduced from the perspective of the switching boards.

[0074] In a second implementation, the target network entity includes multiple interface boards or multiple switching network boards, the interface board includes at least one interface chip, and the switching network board includes at least one switching network chip, which is used to forward network traffic from one interface chip to another interface chip. In this case, the resource adjustment strategy includes a third adjustment strategy and a fourth adjustment strategy, the third adjustment strategy indicates reducing the number of enabled switching network chips in the network device, and the fourth adjustment strategy indicates increasing the number of enabled switching network chips in the network device. Based on this, the implementation process of the above step 302 can be: obtaining a first historical traffic load rate of the target network entity; if the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment strategy is the third adjustment strategy; if the current traffic load rate is greater than the first historical traffic load rate, determining that the resource adjustment strategy is the fourth adjustment strategy.

[0075] The first historical traffic load rate may be the traffic load rate of the target network entity last counted, or may be the average value of the traffic load rate of the target network entity in the last resource adjustment period, which is not limited in this embodiment of the present application.

[0076] For the target network entity, if the current traffic load rate is less than the first historical traffic load rate, it means that after the last resource adjustment, the switching network chips in the network device that are in an enabled state may have excess resources while meeting the forwarding resource requirements of the current network traffic. Therefore, in order to improve the resource utilization of the network device and reduce the overall power consumption of the network device, the number of enabled switching network chips in the network device can be reduced. Similarly, if the current traffic load rate is greater than the first historical traffic load rate, it means that after the last resource adjustment, the switching network chips in the network device that are in an enabled state may not be able to meet the forwarding resource requirements of the current network traffic. Therefore, it is necessary to increase the number of enabled switching network chips in the network device so that the switching network chips in the network device that are in an enabled state can forward the current network traffic normally.

[0077] That is, the first historical traffic load rate is used as a reference to determine the change trend of the current traffic load rate of the target network entity, thereby judging whether the number of enabled switching network chips in the network device needs to be increased.

[0078] It should be noted that if the current traffic load rate of the target network entity is equal to the first historical traffic load rate, it means that the switching network chip in the network device that is in the turned-on state can meet the forwarding resource requirements of the current network traffic. In this case, the number of turned-on switching network chips in the network device does not need to be adjusted.

[0079] In some embodiments, in order to ensure the normal forwarding of network traffic and avoid large fluctuations in network traffic during the current resource adjustment period, which may cause the switching network chip in the turned-on state to be unable to meet the forwarding resource requirements of the actual network traffic, the resource adjustment strategy can also be determined to be the fourth adjustment strategy when the current traffic load rate is equal to the first historical traffic load rate.

[0080] It should be noted that adjusting the number of enabled switching network boards is related to adjusting the number of enabled switching network chips. That is, the number of enabled switching network boards in the network device can be dynamically adjusted by dynamically adjusting the number of enabled switching network chips. Turning on a switching network board means turning on at least one switching network chip in the switching network board; and turning off a switching network board means turning off all switching network chips in the switching network board. In other words, the first implementation method mentioned above is to adjust resources in the dimension of the switching network board, while the second implementation method is to adjust resources in the dimension of the switching network chip. Compared with the first implementation method, the second implementation method has a finer granularity in adjusting traffic forwarding resources.

[0081] In this implementation, the resource manager dynamically adjusts the number of enabled switching network chips in the network device based on the changing trend of the current traffic load rate of the target network entity. While meeting the current network traffic forwarding requirements, the overall power consumption of the network device is reduced from the dimension of the switching network chip.

[0082] In a third implementation, the target network entity includes a target interface board, which is any one of multiple interface boards. In this case, the resource adjustment policy includes a fifth adjustment policy and a sixth adjustment policy, where the fifth adjustment policy indicates reducing the number of enabled serial links of the target interface board, and the sixth adjustment policy indicates increasing the number of enabled serial links of the target interface board. Based on this, the implementation process of step 302 may include: obtaining a second historical traffic load rate of the target network entity; if the current traffic load rate is less than the second historical traffic load rate, determining that the resource adjustment policy is the fifth adjustment policy; and if the current traffic load rate is greater than the second historical traffic load rate, determining that the resource adjustment policy is the sixth adjustment policy.

[0083] The second historical traffic load rate may be the traffic load rate of the target interface board counted last time, or may be the average value of the traffic load rate of the target interface board in the last resource adjustment period, which is not limited in this embodiment of the present application.

[0084] For the target interface board, if the current traffic load rate is less than the second historical traffic load rate, it means that after the last resource adjustment, the serial links in the target interface board that are in an open state may have excess resources while meeting the forwarding resource requirements of the current network traffic. Therefore, in order to improve the resource utilization of the target interface board and reduce the overall power consumption of the target interface board, the number of open serial links in the target interface board can be reduced. Similarly, if the current traffic load rate is greater than the second historical traffic load rate, it means that after the last resource adjustment, the serial links in the target interface board that are in an open state may not be able to meet the forwarding resource requirements of the current network traffic. Therefore, it is necessary to increase the number of open serial links in the target interface board so that the serial links in the target interface board that are in an open state can forward the current network traffic normally.

[0085] That is, the second historical traffic load rate is used as a reference to determine the change trend of the current traffic load rate of the target interface board, thereby judging whether the number of enabled serial links in the target interface board needs to be increased.

[0086] It should be noted that if the current traffic load rate of the target interface board is equal to the second historical traffic load rate, it means that the serial links in the target interface board that are in the open state can meet the forwarding resource requirements of the current network traffic. In this case, the number of open serial links of the target interface board does not need to be adjusted.

[0087] In some embodiments, in order to ensure that the target interface board can process network traffic normally and avoid large fluctuations in network traffic during the current resource adjustment period, resulting in the serial link in the open state being unable to meet the forwarding resource requirements of the actual network traffic, the resource adjustment strategy can also be determined as the sixth adjustment strategy when the current traffic load rate is equal to the second historical traffic load rate.

[0088] In this implementation, the resource manager dynamically adjusts the number of open serial links of the target interface board based on the changing trend of the current traffic load rate of the target interface board, so that the serial links in the open state can meet the resource requirements of the target interface board when forwarding the current network traffic, and at the same time reduces the overall power consumption of the network equipment from the dimension of the interface board.

[0089] In some embodiments, according to the third implementation method described above, the number of enabled serial links of each interface board in the network device may be dynamically adjusted to reduce the overall power consumption of the network device from the perspective of the interface board.

[0090] Step 303: Based on the resource adjustment policy, adjust the traffic forwarding resources in the network device.

[0091] As explained in step 302 above, different resource adjustment strategies are determined for different target network entities. Therefore, different resource adjustment strategies lead to different implementations of adjusting traffic forwarding resources. These will be described below.

[0092] Corresponding to the first implementation method illustrated in the above step 302, the resource manager can dynamically adjust the number of enabled switching boards in the network device based on the current traffic load rate of multiple interface boards or multiple switching boards. At this time, the implementation method of step 303 includes the following first and second cases.

[0093] In the first case, if the resource adjustment strategy is the first adjustment strategy, the first single board load rate interval corresponding to the current traffic load rate is determined from multiple single board load rate intervals, and the multiple single board load rate intervals respectively correspond to a number of network boards; according to the first number of network boards corresponding to the first single board load rate interval, at least one switching network board in the network device that is in an open state is shut down, so that the number of switching network boards that have not been shut down in the network device is equal to the number of first network boards.

[0094] It should be noted that, to distinguish it from the shutdown process of the switching network chip, the switching network board that has not been shut down here refers to a switching network board in which at least one switching network chip is in the turned-on state. For a switching network board in a network device, if at least one switching network chip is in the turned-on state, the switching network board is considered to be in the turned-on state. Therefore, when the number of switching network boards that have not been shut down in the network device is equal to the number of first network boards, the switching network chips in the switching network board that has not been shut down may all be in the turned-on state, or may be partially in the turned-on state. This embodiment of the present application does not impose any restrictions on this.

[0095] In a possible implementation, when shutting down the switching network unit, the switching network unit may be shut down by powering off, clock gating, or frequency reduction. The embodiment of the present application does not limit the specific shutting down method.

[0096] Optionally, when the requirements for fast business wake-up are met, the serial link, media access control address (MAC) interface module (also called MAG interface module), switch buffer (SWB) module, etc. of the switching network chip in the switching network board can also be selectively shut down when the switching network board is shut down. The embodiments of the present application do not limit this.

[0097] In the second case, if the resource adjustment strategy is the second adjustment strategy, the second single board load rate interval corresponding to the current traffic load rate is determined from multiple single board load rate intervals, and the multiple single board load rate intervals respectively correspond to a number of network boards; according to the second number of network boards corresponding to the second single board load rate interval, at least one switching network board in the network device that is in a closed state is turned on, so that the number of switching network boards in the network device that are in an open state is equal to the number of second network boards.

[0098] Similarly, in order to distinguish it from the startup processing of the switching network chip, if there is at least one switching network chip in the startup state, the switching network board is considered to be in the startup state. The embodiment of the present application does not limit the number of switching network chips that are turned on in the switching network board in the startup state.

[0099] In a possible implementation, when the switching network unit is turned on, the switching network unit may be turned on by powering on, releasing clock gating, or releasing frequency reduction. The embodiment of the present application does not limit the specific turning-on method.

[0100] Optionally, when the SFU is enabled, the serial link, MAG interface module, SWB module, etc. of the SFU chip in the SFU can be enabled. Specifically, the functional modules disabled during the disable process are correspondingly enabled during the enable process.

[0101] It should be noted that the above-mentioned multiple single-board load rate intervals are obtained based on the overall ability of multiple interface boards to process network traffic, or the multiple single-board load rate intervals are obtained based on the overall ability of multiple switching network boards to process network traffic; wherein the division granularity of the multiple single-board load rate intervals is greater than the reference granularity.

[0102] When the target network entity is a plurality of interface boards, multiple board load rate intervals and the number of network boards corresponding to each board load rate interval are determined based on the overall network traffic processing capability of the plurality of interface boards. When the target network entity is a plurality of switching network boards, multiple board load rate intervals and the number of network boards corresponding to each board load rate interval are determined based on the overall network traffic processing capability of the plurality of switching network boards.

[0103] Combining the above two situations, when the target network entity includes multiple interface boards or multiple switching network boards, based on the current traffic load rate of the target network entity, first determine a single board load rate interval in which the current traffic load rate is located from multiple single board load rate intervals, and then use the number of network boards corresponding to the single board load rate interval as the target adjustment value to adjust the number of enabled switching network boards in the network device.

[0104] As an example, determining multiple single-board load rate intervals and the number of network boards corresponding to each single-board load rate interval can be achieved by dividing the network traffic load ratio into multiple intervals based on the overall network traffic processing capabilities of multiple interface boards or the overall network traffic processing capabilities of multiple switching network boards, thereby determining multiple single-board load rate intervals; and configuring an on / off network board ratio value for each single-board load rate interval to determine the number of network boards corresponding to each single-board load rate interval. Table 1 shows a correspondence between single-board load rate intervals and the number of network boards.

[0105] Table 1

[0106] It should be understood that Table 1 only uses three single board load rate intervals as an example, and does not constitute a limitation on the number of divided single board load rate intervals.

[0107] Referring to Table 1, if the current traffic load rate of the target network entity is 68%, since the current traffic load rate is within the single board load rate range of (56%, 75%), the number of switched-on switching boards in the network device is adjusted according to the on / off board ratio of 80% / 20%, so that the ratio of the number of switched-on switching boards in the network device to the total number of switched-on switching boards is 80%.

[0108] Based on Table 1 above, given a network device consisting of five interface boards and five switching fabric boards, with the five interface boards processing a total traffic rate of 12.8 Tbps and each switching fabric board capable of processing 2.56 Tbps of network traffic, 12.8 Tbps can be divided into three traffic load intervals, thereby determining three single-board load rate intervals, each corresponding to a traffic load interval. A number of network boards to be enabled is then assigned to each single-board load rate interval. Table 2 shows another correspondence between single-board load rate intervals and the number of network boards.

[0109] Table 2

[0110] Referring to Table 2, if the current traffic load rate of the target network entity is 68% and the total capacity of the target network entity is 12.8Tbps, it is determined that the current traffic load of the target network entity is approximately 8.7Mbps. Since the current traffic load is within the single-board load rate range of (7.2Mbps, 9.6Mbps), the number of enabled switching network boards in the network device is adjusted to 4.

[0111] Referring to Table 1 or Table 2 above, if the current traffic load rate of the target network entity is the same as the endpoint value of a certain single-board load rate interval, the number of enabled switching boards in the network device can be dynamically adjusted according to the number of boards corresponding to the single-board load rate interval, or the operation of adjusting the number of enabled switching boards in the network device can be omitted. This embodiment of the present application is not limited to this.

[0112] It should be understood that the above Tables 1 and 2 are only examples of the single-board load rate range and the number of network boards. In actual applications, the correspondence between the single-board load rate range and the number of network boards can also be set in other ways. The embodiments of the present application do not limit this.

[0113] Corresponding to the second implementation method illustrated in the above step 302, the resource manager can dynamically adjust the number of enabled switching network chips in the network device based on the current traffic load rate of multiple interface boards or multiple switching network boards. At this time, the implementation method of step 303 includes the following third and fourth cases.

[0114] In the third case, if the resource adjustment strategy is the third adjustment strategy, the first chip load rate interval corresponding to the current traffic load rate is determined from multiple chip load rate intervals, and the multiple chip load rate intervals respectively correspond to a chip quantity; according to the first chip quantity corresponding to the first chip load rate interval, at least one switching network chip in the network device that is in the turned-on state is shut down, so that the number of switching network chips in the network device that are in the turned-on state is equal to the first chip quantity.

[0115] In one possible implementation, while meeting the requirements for fast business wake-up, the serial link, MAG interface module, SWB module, etc. of the switching network chip can also be selectively shut down when shutting down the switching network chip. This embodiment of the present application does not limit this.

[0116] In the fourth case, if the resource adjustment strategy is the fourth adjustment strategy, the second chip load rate interval corresponding to the current traffic load rate is determined from multiple chip load rate intervals, and the multiple chip load rate intervals respectively correspond to a chip quantity; according to the second chip quantity corresponding to the second chip load rate interval, at least one switching network chip in the closed state in the network device is turned on, so that the number of switching network chips in the turned-on state in the network device is equal to the second chip quantity.

[0117] In one possible implementation, when the switching network chip is enabled, the serial link, MAG interface module, SWB module, etc. of the switching network chip can be enabled. For example, the functional modules that are disabled during the shutdown process will be enabled during the enable process. This embodiment of the present application does not impose any restrictions on this.

[0118] It should be noted that the above-mentioned multiple chip load rate intervals are obtained based on the ability of the interface chips included in multiple interface boards to process network traffic, or the multiple chip load rate intervals are obtained based on the ability of the switching network chips included in multiple switching network boards to process network traffic; wherein the division granularity of the chip load rate intervals is smaller than the reference granularity.

[0119] That is, the granularity of dividing the chip load rate interval is smaller than the granularity of dividing the board load rate interval in the previous article. Therefore, the traffic forwarding resources in the network device can be dynamically adjusted at a finer granularity.

[0120] When the target network entity is a plurality of interface boards, multiple chip load rate intervals and the number of chips corresponding to each chip load rate interval are determined based on the network traffic processing capabilities of the interface chips included in the plurality of interface boards. When the target network entity is a plurality of switching fabric boards, multiple chip load rate intervals and the number of chips corresponding to each chip load rate interval are determined based on the network traffic processing capabilities of the switching fabric chips included in the plurality of switching fabric boards.

[0121] Combining the above two situations, when the target network entity includes multiple interface boards or multiple switching network boards, based on the current traffic load rate of the target network entity, first determine a chip load rate interval in which the current traffic load rate is located from multiple chip load rate intervals, and then use the number of chips corresponding to the chip load rate interval as the target adjustment value to adjust the number of enabled switching network chips in the network device.

[0122] As an example, determining multiple chip load rate intervals and the number of chips corresponding to each chip load rate interval can be accomplished by dividing the network traffic processing capabilities of multiple interface boards or multiple switching network boards into multiple traffic load ratio intervals, thereby determining multiple chip load rate intervals; and configuring an on / off chip ratio value for each chip load rate interval to determine the number of chips corresponding to each chip load rate interval. Table 3 shows a correspondence between chip load rate intervals and chip quantities.

[0123] Table 3

[0124] It should be understood that Table 3 only uses five chip load rate intervals as an example, and does not constitute a limitation on the number of divided chip load rate intervals.

[0125] Referring to Table 3, if the current traffic load rate of the target network entity is 70%, since the current traffic load rate is within the chip load rate range of (60%, 80%), the number of switched-on switching chips in the network device is adjusted according to the on / off chip ratio of 90% / 10%, so that the ratio of the number of switched-on switching chips in the network device to the total number of switched-on switching chips is 90%.

[0126] Based on Table 3 above, if the network device includes 5 interface boards and 5 switching network boards, the overall traffic processing rate of the 5 interface boards is 12.8 Tbps, the capacity of each switching network board to process network traffic is 2.56 Tbps, and one switching network board includes two switching network chips, that is, the network device includes 10 switching network chips, and the processing capacity of each switching network chip is 1.28 Tbps.

[0127] In this case, 12.8 Tbps can be divided into five traffic load intervals, thereby determining five chip load rate intervals, each corresponding to a traffic load interval; and each chip load rate interval is configured with a chip enable quantity. Table 4 shows another correspondence between chip load rate intervals and chip quantities.

[0128] Table 4

[0129] Referring to Table 4, if the current traffic load rate of the target network entity is 70% and the total capacity of the target network entity is 12.8Tbps, it can be determined that the current traffic load of the target network entity is approximately 8.96Mbps. Since the current traffic load is within the chip load rate range of (7.68Mbps, 10.2Mbps), the number of enabled switching network chips in the network device is adjusted to 9.

[0130] Referring to Table 3 or Table 4 above, if the current traffic load rate of the target network entity is the same as the endpoint value of a chip load rate interval, the number of enabled switching network chips in the network device can be dynamically adjusted according to the number of chips corresponding to the chip load rate interval, or the operation of adjusting the number of enabled switching network chips in the network device can be omitted. This embodiment of the present application is not limited to this.

[0131] It should be understood that the above Tables 3 and 4 are only examples of the expression of chip load rate intervals and chip quantities. In actual applications, the correspondence between the chip load rate intervals and chip quantities can also be set in other ways, and the embodiments of the present application do not limit this.

[0132] In some embodiments, when adjusting the number of enabled switching network chips in a network device, at least one switching network chip may be selected from the switching network chips included in the network device for shutdown or enablement according to a certain chip selection strategy.

[0133] The chip selection strategy involves selecting the switching fabric chips to be shut down from SFUs where some of the chips have already been shut down. Similarly, the strategy involves selecting the switching fabric chips to be shut down from SFUs where some of the chips have already been shut down. This strategy involves shutting down or shutting down the switching fabric chips on a particular SFU as centrally as possible.

[0134] In one possible implementation, based on the switch states of the switching fabric chips in multiple switching fabric boards and the number of first chips, a target chip that needs to be shut down is determined from at least one switching fabric chip in an enabled state in the network device, and the target chip is shut down. Similarly, based on the switch states of the switching fabric chips in multiple switching fabric boards and the number of second chips, a target chip that needs to be turned on is determined from at least one switching fabric chip in a disabled state in the network device, and the target chip is turned on.

[0135] As an example, the network device includes five switching network boards, each of which includes three switching network chips, and the enabled states of the switching network chips in the five switching network boards are shown in Table 5 below.

[0136] Table 5

[0137] As can be seen, all the switching fabric chips on SFU 2 in the network device are in the disabled state, while all the switching fabric chips on SFU 4 are in the enabled state. On SFUs 1, 3, and 5, some switching fabric chips are enabled, while others are disabled.

[0138] When the number of the first chips is 3, three switching fabric chips need to be selected from the enabled switching fabric chips in the network device for shutdown. Considering that some switching fabric chips on switching fabric board 1, switching fabric board 3, and switching fabric board 5 are already disabled, three switching fabric chips are selected as target chips from the enabled switching fabric chips 1.1, 1.3, 3.2, and 5.1, and are shut down.

[0139] When the second number of chips is 5, five switching fabric chips need to be selected from the disabled switching fabric chips in the network device for activation. Similarly, considering that some switching fabric chips on switching fabric board 1, switching fabric board 3, and switching fabric board 5 are already enabled, the disabled switching fabric chips 1.2, 3.1, 3.3, 5.2, and 5.3 are selected as target chips and activated.

[0140] In addition, before executing the above step 303, whether to adjust the number of enabled switching network boards or switching network chips in the network device also requires determining whether the current network traffic meets the service level agreement (SLA) requirements.

[0141] In some embodiments, the resource manager obtains SLA information corresponding to the current network traffic; if the SLA information meets the SLA requirements corresponding to the current network traffic, the above step 303 is executed.

[0142] The SLA information for the current network traffic includes at least one SLA parameter, including one or more of packet loss rate, bit error rate, and latency. Correspondingly, the SLA requirement for the current network traffic includes at least one requirement information, including one or more of packet loss rate requirement, bit error rate requirement, and latency requirement. Each SLA parameter corresponds to one piece of requirement information.

[0143] In a possible implementation, when each SLA parameter of the current network traffic satisfies corresponding requirement information, it is determined that the SLA information of the current network traffic meets the SLA requirement of the current network traffic.

[0144] In some embodiments, if the SLA information of the current network traffic does not meet the SLA requirements corresponding to the current network traffic, at least one switching network board or switching network chip in the network device that is in a closed state is opened according to the target unit opening amount.

[0145] In a possible implementation, for multiple SLA parameters of the current network traffic, if there is an SLA parameter that does not meet corresponding requirement information, then the SLA information of the current network traffic does not meet the SLA requirement of the current network traffic.

[0146] The target unit enable quantity can be set to any value, such as 1, 2, or 3. For example, if the target unit enable quantity is 1, then if the SLA information of the current network traffic does not meet the SLA requirement corresponding to the current network traffic, one switching fabric board or one switching fabric chip in the network device that is in a disabled state will be enabled.

[0147] Corresponding to the third implementation method illustrated in the above step 302, the resource manager can dynamically adjust the number of serial links opened on the target interface board based on the current traffic load rate of the target interface board. At this time, the implementation methods of step 303 include the following fifth and sixth cases.

[0148] In the fifth case, if the resource adjustment strategy is the fifth adjustment strategy, the first link load rate interval corresponding to the current traffic load rate is determined from multiple link load rate intervals, and the multiple link load rate intervals respectively correspond to a link quantity; according to the first link quantity corresponding to the first link load rate interval, at least one serial link in the target interface board that is in an open state is closed, so that the number of serial links in the target interface board that are in an open state is equal to the first link quantity.

[0149] It should be noted that since the interface board is involved in forwarding services, a specific interface board cannot be directly shut down. Therefore, when shutting down the serial link of the target interface board, clock control or frequency reduction can be used to shut it down, and this embodiment of the application does not limit this.

[0150] In some embodiments, whether to shut down the serial link of the target interface board can be determined based on the support status of the interface chip in the target interface board. Under the premise of ensuring normal traffic forwarding services and interface chip functions, the serial link of the target interface board can be shut down according to the first link quantity.

[0151] Optionally, on the basis of shutting down the serial link, sub-modules, sub-planes, etc. inside the target interface board may also be shut down, which is not limited in the embodiment of the present application.

[0152] In the sixth case, if the resource adjustment strategy is the sixth adjustment strategy, the second link load rate interval corresponding to the current traffic load rate is determined from multiple link load rate intervals, and the multiple link load rate intervals respectively correspond to a link quantity; according to the second link quantity corresponding to the second link load rate interval, at least one serial link in the closed state in the target interface board is opened, so that the number of serial links in the opened state in the target interface board is equal to the second link quantity.

[0153] In some embodiments, when the serial link is turned on, the sub-modules, sub-planes, etc. that are closed inside the template interface board can also be turned on. The specific turning-on processing method can be to release the clock, cancel the frequency reduction, etc., and may also include table item sending, module wake-up and other operation processes until the target interface board is working normally.

[0154] It should be noted that the multiple link load rate intervals are obtained based on the capability of the target interface board to process network traffic.

[0155] In summary, in the embodiments of the present application, the resource manager determines the resource adjustment policy corresponding to the current network traffic based on the current traffic load rate of the target network entity, and then adjusts the traffic forwarding resources in the network device used to forward the current network traffic based on the resource adjustment policy. In this way, the traffic forwarding resources in the network device are dynamically adjusted according to the current traffic load rate of the target network entity, so that the overall power consumption of the network device can change with the changes in the current network traffic. While meeting the current network traffic forwarding requirements, it avoids running a large number of traffic forwarding resources under low traffic load conditions, thereby reducing the power demand and heat dissipation requirements of the network device under low traffic load conditions.

[0156] Referring to FIG. 5 , FIG. 5 is a schematic diagram illustrating the structure of a resource adjustment device provided in an embodiment of the present application. The resource adjustment device can be implemented by software, hardware, or a combination of both to form part or all of the resource adjuster 140 shown in FIG. 1 or FIG. Resource adjustment device 500 includes a load rate acquisition module 501, a policy determination module 502, and a resource adjustment module 503.

[0157] The load rate acquisition module 501 is used to obtain the current traffic load rate of the target network entity. The current traffic load rate indicates the load condition of the target network entity when forwarding the current network traffic. The target network entity is multiple interface boards, multiple switching network boards, or any one of multiple interface boards. For the detailed implementation process, please refer to the relevant content in the above step 301, which will not be repeated here.

[0158] The policy determination module 502 is used to determine the resource adjustment policy corresponding to the current network traffic based on the current traffic load rate. The resource adjustment policy indicates the adjustment of the traffic forwarding resources in the network device when forwarding the current network traffic. The traffic forwarding resources include the serial links of the switching network board and the interface board. For the detailed implementation process, please refer to the relevant content in the above step 302, which will not be repeated here.

[0159] The resource adjustment module 503 is used to adjust the traffic forwarding resources in the network device based on the resource adjustment policy. For detailed implementation process, please refer to the relevant content in the above step 303, which will not be repeated here.

[0160] Optionally, the target network entity includes multiple interface boards or multiple switching network boards, and the resource adjustment strategy includes a first adjustment strategy and a second adjustment strategy, the first adjustment strategy indicates reducing the number of enabled switching network boards in the network device, and the second adjustment strategy indicates increasing the number of enabled switching network boards in the network device;

[0161] The strategy determination module 502 is specifically configured to:

[0162] Obtaining a first historical traffic load rate of a target network entity;

[0163] If the current traffic load rate is less than the first historical traffic load rate, determining the resource adjustment strategy to be the first adjustment strategy;

[0164] If the current traffic load rate is greater than the first historical traffic load rate, the resource adjustment strategy is determined to be the second adjustment strategy.

[0165] Optionally, the resource adjustment module 503 is specifically configured to:

[0166] If the resource adjustment strategy is the first adjustment strategy, determining a first single board load rate interval corresponding to the current traffic load rate from a plurality of single board load rate intervals, wherein the plurality of single board load rate intervals respectively correspond to one network board number;

[0167] At least one switching network board in the network device that is turned on is shut down according to the first number of network boards corresponding to the first single board load rate interval, so that the number of switching network boards that are not shut down in the network device is equal to the number of first network boards.

[0168] Optionally, the resource adjustment module 503 is specifically configured to:

[0169] If the resource adjustment strategy is the second adjustment strategy, determining a second single board load rate interval corresponding to the current traffic load rate from a plurality of single board load rate intervals, wherein the plurality of single board load rate intervals respectively correspond to one network board number;

[0170] According to the number of second network boards corresponding to the second single board load rate interval, at least one switching network board in the network device that is in a disabled state is enabled, so that the number of enabled switching network boards in the network device is equal to the number of second network boards.

[0171] Optionally, multiple single-board load rate intervals are obtained based on the overall ability of multiple interface boards to process network traffic, or multiple single-board load rate intervals are obtained based on the overall ability of multiple switching network boards to process network traffic; wherein the division granularity of the multiple single-board load rate intervals is greater than the reference granularity.

[0172] Optionally, the interface board includes at least one interface chip, the switching network board includes at least one switching network chip, and the switching network chip is used to forward network traffic of one interface chip to another interface chip; the target network entity includes multiple interface boards or multiple switching network boards, and the resource adjustment policy includes a third adjustment policy and a fourth adjustment policy, the third adjustment policy indicates reducing the number of enabled switching network chips in the network device, and the fourth adjustment policy indicates increasing the number of enabled switching network chips in the network device;

[0173] The strategy determination module 502 is specifically configured to:

[0174] Obtaining a first historical traffic load rate of a target network entity;

[0175] If the current traffic load rate is less than the first historical traffic load rate, determining the resource adjustment strategy to be the third adjustment strategy;

[0176] If the current traffic load rate is greater than the first historical traffic load rate, the resource adjustment strategy is determined to be the fourth adjustment strategy.

[0177] Optionally, the resource adjustment module 503 is specifically configured to:

[0178] If the resource adjustment strategy is the third adjustment strategy, determining a first chip load rate interval corresponding to the current traffic load rate from a plurality of chip load rate intervals, wherein each of the plurality of chip load rate intervals corresponds to a number of chips;

[0179] According to the first chip quantity corresponding to the first chip load rate interval, at least one switching network chip in the network device that is in an enabled state is disabled, so that the number of switching network chips in the network device that are in an enabled state is equal to the first chip quantity.

[0180] Optionally, the resource adjustment module 503 is specifically configured to:

[0181] If the resource adjustment strategy is the fourth adjustment strategy, determining a second chip load rate interval corresponding to the current traffic load rate from a plurality of chip load rate intervals, wherein each of the plurality of chip load rate intervals corresponds to a number of chips;

[0182] According to the second chip quantity corresponding to the second chip load rate interval, at least one switching network chip in the network device that is in a disabled state is enabled, so that the number of enabled switching network chips in the network device is equal to the second chip quantity.

[0183] Optionally, multiple chip load rate intervals are obtained based on the ability of interface chips included in multiple interface boards to process network traffic, or multiple chip load rate intervals are obtained based on the ability of switching network chips included in multiple switching network boards to process network traffic; wherein the division granularity of the chip load rate intervals is smaller than the reference granularity.

[0184] Optionally, the resource adjustment device 500 further includes:

[0185] Information acquisition module, used to obtain the service level agreement (SLA) information corresponding to the current network traffic;

[0186] The resource adjustment module is further configured to execute a step of adjusting the traffic forwarding resources in the network device based on the resource adjustment policy if the SLA information meets the SLA requirements corresponding to the current network traffic.

[0187] Optionally, the resource adjustment module is further configured to enable at least one switching network board or switching network chip in a closed state in the network device according to the target unit enable amount if the SLA information does not meet the SLA requirement corresponding to the current network traffic.

[0188] Optionally, the target network entity includes a target interface board, the target interface board is any one of a plurality of interface boards, the resource adjustment policy includes a fifth adjustment policy and a sixth adjustment policy, the fifth adjustment policy indicates reducing the number of enabled serial links of the target interface board, and the sixth adjustment policy indicates increasing the number of enabled serial links of the target interface board;

[0189] The strategy determination module 502 is specifically configured to:

[0190] Obtaining a second historical traffic load rate of the target network entity;

[0191] If the current traffic load rate is less than the second historical traffic load rate, determining the resource adjustment strategy to be the fifth adjustment strategy;

[0192] If the current traffic load rate is greater than the second historical traffic load rate, the resource adjustment strategy is determined to be the sixth adjustment strategy.

[0193] Optionally, the resource adjustment module 503 is specifically configured to:

[0194] If the resource adjustment strategy is the fifth adjustment strategy, determining a first link load rate interval corresponding to the current traffic load rate from a plurality of link load rate intervals, wherein each of the plurality of link load rate intervals corresponds to a link quantity;

[0195] At least one serial link in the target interface board that is in an open state is closed according to the first link quantity corresponding to the first link load rate interval, so that the number of serial links in the target interface board that are in an open state is equal to the first link quantity.

[0196] Optionally, the resource adjustment module 503 is specifically configured to:

[0197] If the resource adjustment strategy is the sixth adjustment strategy, determining a second link load rate interval corresponding to the current traffic load rate from a plurality of link load rate intervals, wherein each of the plurality of link load rate intervals corresponds to a link quantity;

[0198] At least one closed serial link in the target interface board is opened according to the second link quantity corresponding to the second link load rate interval, so that the number of opened serial links in the target interface board is equal to the second link quantity.

[0199] Optionally, the multiple link load rate intervals are obtained based on the capability of the target interface board to process network traffic.

[0200] In an embodiment of the present application, a resource adjustment device determines a resource adjustment policy corresponding to the current network traffic based on the current traffic load rate of the target network entity, and then adjusts the traffic forwarding resources in the network device used to forward the current network traffic based on the resource adjustment policy. In this way, the traffic forwarding resources in the network device are dynamically adjusted according to the current traffic load rate of the target network entity, so that the overall power consumption of the network device can change with changes in the current network traffic. While meeting the current network traffic forwarding requirements, it avoids running a large number of traffic forwarding resources under low traffic load conditions, thereby reducing the power and heat dissipation requirements of the network device under low traffic load conditions.

[0201] It should be noted that the resource adjustment device provided in the above embodiment only uses the division of the above functional modules as an example to illustrate the dynamic adjustment of traffic forwarding resources in a network device. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the resource adjustment device provided in the above embodiment and the resource adjustment method embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0202] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a network device, the network device executes the steps of the above-mentioned resource adjustment method.

[0203] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program runs on a network device, the network device executes the steps of the above-mentioned resource adjustment method.

[0204] In the above embodiments, the resource adjustment method can be implemented in whole or in part through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a network device, the process or function according to the embodiments of the present application is generated in whole or in part. The network device can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer program can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., a digital versatile disc (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)). It is worth noting that the computer-readable storage medium mentioned in the embodiments of the present application may be a non-volatile storage medium, in other words, a non-transient storage medium.

[0205] It should be understood that the "plurality" mentioned herein refers to two or more. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in order to facilitate a clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, words such as "first" and "second" are used to distinguish between identical or similar items with substantially the same functions and effects. Those skilled in the art will understand that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit them to be different.

[0206] It should be noted that the information (including but not limited to the business type and business information of network traffic), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in the embodiments of this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0207] The above description is an embodiment provided for this application and is not intended to limit this application. Any modifications, equivalent replacements, improvements, etc. made within the principles of this application should be included in the scope of protection of this application.

Claims

1. A resource adjustment method, characterized in that: A resource manager applied to a network device, wherein the network device further comprises a plurality of interface boards and a plurality of switching network boards, and each interface board is connected to at least one switching network board via at least one serial link, and the switching network board is used to forward network traffic of one interface board to another interface board; the method comprises: Acquire a current traffic load rate of a target network entity, where the current traffic load rate indicates a load condition of the target network entity when forwarding current network traffic, and the target network entity is the multiple interface boards, the multiple switching network boards, or any one of the multiple interface boards; Based on the current traffic load rate, determine a resource adjustment strategy corresponding to the current network traffic, the resource adjustment strategy indicating an adjustment of traffic forwarding resources in the network device when forwarding the current network traffic, the traffic forwarding resources including serial links of the switching network board and the interface board; Based on the resource adjustment policy, the traffic forwarding resources in the network device are adjusted.

2. The method according to claim 1, characterized in that The target network entity includes the multiple interface boards or the multiple switching network boards, the resource adjustment strategy includes a first adjustment strategy and a second adjustment strategy, the first adjustment strategy indicates to reduce the number of enabled switching network boards in the network device, and the second adjustment strategy indicates to increase the number of enabled switching network boards in the network device; The determining, based on the current traffic load rate, a resource adjustment strategy corresponding to the current network traffic includes: Obtaining a first historical traffic load rate of the target network entity; If the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment strategy is the first adjustment strategy; If the current traffic load rate is greater than the first historical traffic load rate, the resource adjustment strategy is determined to be the second adjustment strategy.

3. The method according to claim 2, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the first adjustment strategy, determining a first single board load rate interval corresponding to the current traffic load rate from a plurality of single board load rate intervals, wherein the plurality of single board load rate intervals respectively correspond to a number of network boards; According to the first network board quantity corresponding to the first single board load rate interval, at least one switching network board in the network device that is in an enabled state is shut down, so that the number of switching network boards that have not been shut down in the network device is equal to the first network board quantity.

4. The method according to claim 2, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the second adjustment strategy, determining the second single board load rate interval corresponding to the current traffic load rate from a plurality of single board load rate intervals, wherein the plurality of single board load rate intervals respectively correspond to a number of network boards; According to the number of second network boards corresponding to the second single-board load rate interval, at least one switching network board in the closed state in the network device is turned on, so that the number of switching network boards in the opened state in the network device is equal to the number of second network boards.

5. The method according to claim 3 or 4, characterized in that The multiple single-board load rate intervals are obtained based on the overall ability of the multiple interface boards to process network traffic, or the multiple single-board load rate intervals are obtained based on the overall ability of the multiple switching network boards to process network traffic; wherein the division granularity of the multiple single-board load rate intervals is greater than the reference granularity.

6. The method according to claim 1, characterized in that The interface board includes at least one interface chip, the switching network board includes at least one switching network chip, and the switching network chip is used to forward network traffic of one interface chip to another interface chip; the target network entity includes the multiple interface boards or the multiple switching network boards, the resource adjustment strategy includes a third adjustment strategy and a fourth adjustment strategy, the third adjustment strategy indicates to reduce the number of enabled switching network chips in the network device, and the fourth adjustment strategy indicates to increase the number of enabled switching network chips in the network device; The determining, based on the current traffic load rate, a resource adjustment strategy corresponding to the current network traffic includes: Obtaining a first historical traffic load rate of the target network entity; If the current traffic load rate is less than the first historical traffic load rate, determining that the resource adjustment strategy is the third adjustment strategy; If the current traffic load rate is greater than the first historical traffic load rate, the resource adjustment strategy is determined to be the fourth adjustment strategy.

7. The method according to claim 6, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the third adjustment strategy, determining a first chip load rate interval corresponding to the current traffic load rate from a plurality of chip load rate intervals, wherein the plurality of chip load rate intervals respectively correspond to a chip quantity; According to the first chip quantity corresponding to the first chip load rate interval, at least one switching network chip in the network device that is in an enabled state is shut down so that the number of switching network chips in the network device that are in an enabled state is equal to the first chip quantity.

8. The method according to claim 6, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the fourth adjustment strategy, determining a second chip load rate interval corresponding to the current traffic load rate from a plurality of chip load rate intervals, wherein the plurality of chip load rate intervals respectively correspond to a number of chips; According to the second chip quantity corresponding to the second chip load rate interval, at least one switching network chip in the network device that is in a closed state is turned on, so that the number of switching network chips in the network device that are in an opened state is equal to the second chip quantity.

9. The method according to claim 7 or 8, characterized in that The multiple chip load rate intervals are obtained based on the capabilities of the interface chips included in the multiple interface boards to process network traffic, or the multiple chip load rate intervals are obtained based on the capabilities of the switching network chips included in the multiple switching network boards to process network traffic; wherein the division granularity of the chip load rate intervals is smaller than the reference granularity.

10. The method according to any one of claims 1 to 9, characterized in that: The method further comprises: Obtaining service level agreement (SLA) information corresponding to the current network traffic; If the SLA information meets the SLA requirement corresponding to the current network traffic, the step of adjusting the traffic forwarding resources in the network device based on the resource adjustment policy is performed.

11. The method according to claim 10, characterized in that The method further comprises: If the SLA information does not meet the SLA requirement corresponding to the current network traffic, at least one switching network board or switching network chip in the network device that is in a closed state is turned on according to the target unit turn-on amount.

12. The method according to claim 1, characterized in that The target network entity includes a target interface board, the target interface board is any one of the multiple interface boards, the resource adjustment strategy includes a fifth adjustment strategy and a sixth adjustment strategy, the fifth adjustment strategy indicates to reduce the number of enabled serial links of the target interface board, and the sixth adjustment strategy indicates to increase the number of enabled serial links of the target interface board; The determining, based on the current traffic load rate, a resource adjustment strategy corresponding to the current network traffic includes: Acquire a second historical traffic load rate of the target network entity; If the current traffic load rate is less than the second historical traffic load rate, determining that the resource adjustment strategy is the fifth adjustment strategy; If the current traffic load rate is greater than the second historical traffic load rate, the resource adjustment strategy is determined to be the sixth adjustment strategy.

13. The method according to claim 12, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the fifth adjustment strategy, determining a first link load rate interval corresponding to the current traffic load rate from a plurality of link load rate intervals, wherein the plurality of link load rate intervals respectively correspond to a link quantity; According to the first link quantity corresponding to the first link load rate interval, at least one serial link in the target interface board that is in an open state is closed so that the number of serial links in the target interface board that are in an open state is equal to the first link quantity.

14. The method according to claim 12, characterized in that The adjusting the traffic forwarding resources in the network device based on the resource adjustment policy includes: If the resource adjustment strategy is the sixth adjustment strategy, determining a second link load rate interval corresponding to the current traffic load rate from a plurality of link load rate intervals, wherein the plurality of link load rate intervals respectively correspond to a link quantity; According to the second link quantity corresponding to the second link load rate interval, at least one serial link in the target interface board that is in a closed state is opened, so that the number of serial links in the target interface board that are in an opened state is equal to the second link quantity.

15. The method according to claim 13 or 14, characterized in that The multiple link load rate intervals are obtained based on the ability of the target interface board to process network traffic.

16. A resource adjustment device, characterized in that: A resource manager included in a network device, the network device further comprising a plurality of interface boards and a plurality of switching network boards, each interface board being connected to at least one switching network board via at least one serial link, the switching network board being used to forward network traffic of one interface board to another interface board; The resource adjustment device comprises: A load rate acquisition module, used to acquire a current traffic load rate of a target network entity, wherein the current traffic load rate indicates a load condition of the target network entity when forwarding current network traffic, wherein the target network entity is the multiple interface boards, the multiple switching network boards, or any one of the multiple interface boards; A policy determination module, configured to determine a resource adjustment policy corresponding to the current network traffic based on the current traffic load rate, wherein the resource adjustment policy indicates an adjustment of traffic forwarding resources in the network device when forwarding the current network traffic, wherein the traffic forwarding resources include serial links of the switching network board and the interface board; The resource adjustment module is used to adjust the traffic forwarding resources in the network device based on the resource adjustment policy.

17. A network device, characterized in that: The network device includes a resource manager, a main control board, multiple interface boards and multiple switching network boards, each interface board is connected to at least one switching network board through at least one serial link, and the switching network board is used to forward the network traffic of one interface board to another interface board; Wherein, the resource manager is arranged on the main control board or at least one interface board among the multiple interface boards, and is used to execute the method described in any one of claims 1 to 15 above.

18. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program runs on the network device, the network device executes the method according to any one of claims 1 to 15.

Images