Control method, apparatus, chip, network device, and communication system

Abstract

A control method, device, chip, network device and communication system are disclosed, and relate to the field of communication. The method comprises: a control chip of the network device adjusting opening or closing of a port or a channel corresponding to the port of the network device according to an energy saving strategy and performance data of the network device. The energy saving strategy is used to indicate a condition of opening or closing of the port or the channel corresponding to the port of the network device. The performance data is used to indicate a running feature of the port of the network device in a current period. In this way, the network devices do not need to negotiate with each other and do not need to be controlled by a controller. The network devices only need to adjust the opening or closing of the port or the channel corresponding to the port of the network device based on the performance of the network devices, thereby simplifying a control process of the port or the channel corresponding to the port of the network device, quickly and timely adjusting the network devices, and reducing energy consumption of the network devices.

Description

[0001] This application claims priority to the Chinese Patent Application No. 202310475262.2, filed on April 25, 2023, and entitled "A control method", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of communication, and in particular to a control method, device, chip, network device and communication system. BACKGROUND

[0003] With the gradual increase in the size of data centers, the energy consumption of data centers also increases. Among them, the energy consumption of the network part of the data center accounts for a significant proportion. At present, an additional management channel needs to be built between network devices, signaling is sent through the management channel, and after mutual confirmation, the state of the channel indicated by the control center is adjusted to change the energy consumption of the network device. Therefore, the control process of the network device is relatively complex, and the energy consumption of the network device cannot be adjusted in time. SUMMARY

[0004] The present application provides a control method, device, chip, network device and communication system, thereby simplifying the control process of the port or the channel corresponding to the port of the network device, quickly and timely adjusting the network device, and reducing the energy consumption of the network device.

[0005] In a first aspect, a control method is provided, which is executed by a control chip of a network device. The method comprises adjusting the opening or closing of a port or a channel corresponding to the port of the network device according to an energy saving strategy and performance data of the network device. The energy saving strategy is used to indicate the conditions for opening or closing the port or the channel corresponding to the port of the network device. The performance data is used to indicate the running characteristics of the port of the network device in the current period.

[0006] Compared with the scheme of sending signaling by the network device, mutual confirmation, adjusting the state of the channel indicated by the control center, and changing the energy consumption of the network device, the scheme provided by the present application does not require mutual negotiation between network devices, and does not require control by a controller. The network device only needs to adjust the opening or closing of the port or the channel corresponding to the port of the network device based on its own performance, thereby simplifying the control process of the port or the channel corresponding to the port of the network device, quickly and timely adjusting the network device, and reducing the energy consumption of the network device.

[0007] Among them, the performance data includes at least one of port bandwidth occupancy and queue depth.

[0008] In a possible implementation, the adjusting the state of the port or the channel corresponding to the port according to the energy saving policy and the performance data comprises: adjusting the state of at least one port in the network device or the state of the channel corresponding to the port according to the performance data when the performance data satisfies the closing condition indicated by the energy saving policy. Thus, the state of at least one port in the network device or the state of the channel corresponding to the port is adjusted based on the condition indicated by the energy saving policy, and the accuracy of the adjustment is improved.

[0009] In another possible implementation, the adjusting the state of the channel corresponding to the port comprises: adjusting the state of a serdes corresponding to the channel from an active state to an energy saving state. Thus, the energy consumption of the serdes is reduced when the state of the serdes is adjusted to the energy saving state, and the energy consumption of the network device is reduced.

[0010] In another possible implementation, the port is used for transmitting data, and the adjusting the state of the channel corresponding to the port comprises: instructing a link layer to stop transmitting data, and adjusting the state of the channel corresponding to the port. Thus, it is ensured that the data transmission on the link is completed, and data loss caused in the closing process is prevented.

[0011] In another possible implementation, the network device further comprises a second chip, the port is used for transmitting data, and the adjusting the state of the channel corresponding to the port comprises: instructing the second chip to stop transmitting a training stream, and adjusting the state of the channel corresponding to the port.

[0012] In another possible implementation, the port is used for receiving data, and the adjusting the state of the channel corresponding to the port comprises: adjusting the state of the channel corresponding to the port when no training stream is received.

[0013] Thus, when the training stream is not transmitted from the sending end to the receiving end, the receiving end does not receive the training stream, and the state of the channel corresponding to the port is adjusted as soon as possible when it is determined that the channel training fails, and the energy consumption of the network device is reduced.

[0014] The port is used for transmitting data, or the network device acts as a sending end, and the adjusting the state of the port or the channel corresponding to the port comprises adjusting the state of the port or the channel corresponding to the port used for transmitting data. The port is used for receiving data, or the network device acts as a receiving end, and the adjusting the state of the port or the channel corresponding to the port comprises adjusting the state of the port or the channel corresponding to the port used for receiving data.

[0015] In another possible implementation manner, the state of the port or the channel corresponding to the port is closed, and the method further includes: when the performance data satisfies the opening condition indicated by the energy saving policy, adjusting the port or the channel corresponding to the port to be opened. Thus, when data needs to be transmitted, the port or the channel corresponding to the port is opened to restore data transmission, and the service is prevented from being affected due to the closed port or the channel corresponding to the port.

[0016] In another possible implementation manner, adjusting the port or the channel corresponding to the port to be opened includes: adjusting the state of the serial-parallel converter corresponding to the channel from the energy saving state to the active state. Thus, the serial-parallel converter is in the working state due to the adjustment of the state of the serial-parallel converter to the active state, and data can be transmitted through the channel corresponding to the port.

[0017] For example, when the training code stream is received, the state of the serial-parallel converter corresponding to the channel is adjusted from the energy saving state to the active state.

[0018] In another possible implementation manner, obtaining the energy saving policy of the network device includes: obtaining the energy saving policy of the network device according to the characteristics of the network device, the characteristics of the network device including at least one of the position of the network device in the network, the scale of the network device, the configuration of the network device, and the performance data of the network device. Thus, the network device periodically obtains the energy saving policy, the accuracy of the network device in controlling the port and the channel included in the port according to the energy saving policy is improved, and the energy consumption of the network device is effectively reduced.

[0019] In a second aspect, a control apparatus is provided, which includes various modules for performing the control method in the first aspect or any possible design in the first aspect. For example, the control apparatus includes a communication module and an energy consumption control module. The communication module is configured to obtain an energy saving policy of a network device, the energy saving policy being used to indicate a condition for opening or closing a port or a channel corresponding to the port of the network device. The energy consumption control module is configured to obtain performance data of the network device, the performance data being used to indicate a running feature of the port of the network device in a current period. The energy consumption control module is configured to adjust a state of the port or the channel corresponding to the port according to the energy saving policy and the performance data, the state including opening or closing.

[0020] In a possible implementation manner, when the energy consumption control module adjusts the state of the port or the channel corresponding to the port according to the energy saving policy and the performance data, the energy consumption control module is specifically configured to: when the performance data satisfies a closing condition indicated by the energy saving policy, adjust the state of at least one port in the network device or the state of the channel corresponding to the port according to the performance data.

[0021] In another possible implementation manner, the performance data includes at least one of an interface bandwidth occupancy rate, a queue depth, and a packet delay.

[0022] In a possible implementation, when the energy consumption control module adjusts the state of the channel corresponding to the port, the energy consumption control module is specifically configured to: adjust the state of a serial-parallel converter corresponding to the channel, and migrate from an active state to an energy saving state.

[0023] In a possible implementation, the port is configured to transmit data, and the energy consumption control module is configured to: instruct a second chip to stop transmitting a training code stream, and adjust the state of the channel corresponding to the port.

[0024] In a possible implementation, the port is configured to receive data, and the energy consumption control module is configured to: when no training code stream is received, adjust the state of the channel corresponding to the port.

[0025] In a possible implementation, the port is configured to transmit data, and when the energy consumption control module adjusts the state of the channel corresponding to the port, the energy consumption control module is specifically configured to: instruct a link layer to stop transmitting data, and adjust the state of the channel corresponding to the port.

[0026] In a possible implementation, the state of the port or the channel corresponding to the port is closed, and the energy consumption control module is further configured to: when the performance data satisfies a closing condition indicated by the energy saving strategy, adjust the port or the channel corresponding to the port to be opened.

[0027] In a possible implementation, when the energy consumption control module adjusts the channel corresponding to the port to be opened, the energy consumption control module is specifically configured to: adjust the state of a serial-parallel converter corresponding to the channel, and migrate from an energy saving state to an active state.

[0028] For example, when a training code stream is received, the state of a serial-parallel converter corresponding to the channel is adjusted, and the state is migrated from an energy saving state to an active state.

[0029] In a possible implementation, when the communication module obtains the energy saving strategy of the network device, the communication module is configured to: obtain the energy saving strategy of the network device according to characteristics of the network device, the characteristics of the network device including at least one of a position of the network device in a network, a scale of the network device, a configuration of the network device, and performance data of the network device.

[0030] In a third aspect, a chip is provided, including: a processor and a power supply circuit; the power supply circuit is configured to supply power to the processor; and the processor is configured to perform the operation steps of the method in the first aspect or any possible implementation manner of the first aspect.

[0031] In a fourth aspect, a network device is provided, including at least one processor and a memory, the memory is configured to store a set of computer instructions; when the processor executes the set of computer instructions, the operation steps of the method in the first aspect or any possible implementation manner of the first aspect are performed.

[0032] In a fifth aspect, a communication system is provided, the communication system comprising a first network device and a second network device, the first network device and the second network device performing the operation steps of the method in the first aspect or any possible implementation of the first aspect.

[0033] In a sixth aspect, a computer-readable storage medium is provided, comprising: computer software instructions; when the computer software instructions are run in a network device, causing the network device to perform the operation steps of the method in the first aspect or any possible implementation of the first aspect.

[0034] In a seventh aspect, a computer program product is provided, when the computer program product is run on a computer, causing the computer to perform the operation steps of the method in the first aspect or any possible implementation of the first aspect.

[0035] The technical effects brought by any design in the second aspect to the seventh aspect can be referred to the technical effects brought by the first aspect or different designs in the first aspect, which will not be repeated here.

[0036] On the basis of the implementation manners of the above aspects, the present application can be further combined to provide more implementation manners. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 A schematic diagram of the relationship of a network device, a port, a channel and a serial-parallel converter provided by the present application;

[0038] Figure 2 A schematic diagram of the architecture of a communication system provided by the present application;

[0039] Figure 3 A schematic diagram of the flow of a control method provided by the present application;

[0040] Figure 4 A schematic diagram of the configuration of an energy-saving strategy provided by the present application;

[0041] Figure 5 A schematic diagram of the state transition of a serial-parallel converter provided by the present application;

[0042] Figure 6 A schematic diagram of the state transition of a channel provided by the present application;

[0043] Figure 7 A schematic diagram of the flow of an energy-saving mechanism of a channel provided by the present application;

[0044] Figure 8 A schematic diagram of the flow of forcibly closing a channel provided by the present application;

[0045] Figure 9A flowchart of a non-forced close channel provided for the present application;

[0046] Figure 10 A flowchart of an open channel provided for the present application;

[0047] Figure 11 A schematic diagram of a control process provided for the present application;

[0048] Figure 12 A schematic diagram of a control process provided for the present application;

[0049] Figure 13 A flowchart of a power saving mechanism of a port provided for the present application;

[0050] Figure 14 A schematic diagram of a control device provided for the present application;

[0051] Figure 15 A schematic diagram of a network device provided for the present application. DETAILED DESCRIPTION

[0052] For the convenience of understanding, the main terms involved in the present application are first explained.

[0053] Data center refers to a specific network of devices for delivering, accelerating, presenting, computing, and storing data information on the internet infrastructure, which is a global collaboration.

[0054] Energy consumption refers to an evaluation index of the size of the energy consumption of a product in use.

[0055] Port refers to an interface used for connection between network devices (such as hubs, switches, and routers) in network technology.

[0056] Link refers to a physical line between two nodes without other switching nodes in between. Link can also be referred to as physical link or communication link. A link can include multiple lanes. Lane can also be referred to as physical lane, physical path, or signal lane. For example, a link can include 2, 4, 8, or 16 lanes.

[0057] Serializer Deserializer (Serdes) refers to the conversion of parallel signals into serial signals, or the conversion of serial signals into parallel signals. Serdes can also be referred to as serializer / deserializer. One lane corresponds to one serdes.

[0058] For example, Figure 1A schematic diagram of the relationship among a network device, a port, a channel and a serial-parallel converter is provided in the present application. As shown in Figure 1 Network device A includes port A1. Network device B includes port B1. The link between port A1 and port B1 includes four channels, i.e., channel 1, channel 2, channel 3 and channel 4. When transmitting data from port A1 to port B1 or from port B1 to port A1, the data can be transmitted through the four channels. Network device A further includes serial-parallel converter A2, serial-parallel converter A3, serial-parallel converter A4 and serial-parallel converter A5. Network device B further includes serial-parallel converter B2, serial-parallel converter B3, serial-parallel converter B4 and serial-parallel converter B5. Channel 1 corresponds to serial-parallel converter A2 and serial-parallel converter B2, channel 2 corresponds to serial-parallel converter A3 and serial-parallel converter B3, channel 3 corresponds to serial-parallel converter A4 and serial-parallel converter B4, and channel 4 corresponds to serial-parallel converter A5 and serial-parallel converter B5. For example, serial-parallel converter A2 can convert a parallel signal into a serial signal and transmit the serial signal to port B1 through channel 1, and serial-parallel converter B2 can convert the serial signal into a parallel signal.

[0059] To solve the problem of how to timely adjust the energy consumption of a network device, the present application provides a control method, i.e., adjusting the opening or closing of a port or a channel of a port of a network device according to an energy-saving strategy and performance data of the network device. The energy-saving strategy is used to indicate the condition of the opening or closing of a port or a channel corresponding to the port of the network device. The performance data is used to indicate the running characteristics of the port of the network device in a current period.

[0060] With respect to the scheme of adjusting the state of a channel indicated by a control center to change the energy consumption of a network device after the network device sends signaling and mutual confirmation, the scheme provided by the present application does not require mutual negotiation among network devices and does not require the control of a controller. The network device only needs to adjust the opening or closing of a port or a channel corresponding to the port of the network device based on the performance of the network device, thereby simplifying the control process of a port or a channel of a port of the network device, quickly and timely adjusting the network device and reducing the energy consumption of the network device.

[0061] The scheme provided by the present application can be applied to a communication system such as a cluster network or a supernode interconnection network. For example, a High performance computing (HPC) network or a High Availability (HA) network.

[0062] For example, as shown in Figure 2As shown, the communication system 200 is an entity that provides high performance computing. The communication system 200 includes a plurality of supernodes 210. The supernode 210 includes a plurality of nodes 211. The supernode 210 can be a rack server, or a blade server. The node 211 can be a processor, a server, a desktop computer, a controller of a storage array, a memory, etc. The processor can be a central processing unit (CPU), a graphics processing unit (GPU), a data processing unit (DPU), a neural processing unit (NPU), a neural-network processing unit (NPU), etc. for data processing. For example, the supernode 210 can include a compute node and a storage node.

[0063] The plurality of nodes 211 in the supernode 210 are connected based on a high-speed interconnection technology with high bandwidth and low latency. For example, as shown in FIG. 2, the nodes 211 are connected through a switch 212 based on the high-speed interconnection technology. Figure 2

[0064] Optionally, the plurality of supernodes 210 are connected through a data center network 220. The data center network 220 includes a plurality of core switches 221 and a plurality of aggregation switches 222. The data center network 220 can constitute a scale domain. The plurality of supernodes 210 can constitute a performance domain.

[0065] The data center network topology can include a core layer and an aggregation layer. The core layer is also called a spine layer. The core layer is a high-speed switching backbone layer of the data center network, and is used to connect the data center with devices outside the data center (e.g., external operator devices). The core layer can include core switches and core routers with high bandwidth (e.g., more than gigabit bandwidth). The core layer has at least one of the following characteristics: reliability, efficiency, redundancy, fault tolerance, manageability, adaptability, low latency, etc. The reliability of the data center network can generally be achieved by redundant connection of multiple devices.

[0066] The aggregation layer is an “intermediary” between the access layer and the core layer, and is used to aggregate data output by the access layer to reduce the load of the core layer. The aggregation layer includes aggregation switches.

[0067] ​The network device is configured to receive data transmitted by one device and forward the data to another device. Understandably, the network device comprises a receiving function and a transmitting function. For example, the transmitting and receiving directions of the serdes in the network device are configured according to requirements. If the transmitting and receiving direction of the serdes is the transmitting direction, the network device can transmit data through the channel corresponding to the serdes; if the transmitting and receiving direction of the serdes is the receiving direction, the network device can receive data through the channel corresponding to the serdes.

[0068] In some embodiments, the network device in the communication system is configured with a power saving agent, which is configured to adjust the opening or closing of the ports or channels of the ports of the network device according to the power saving policy and performance data of the network device. The network device can include, but is not limited to, a super node containing a switch, a data center network containing a core switch and a convergence switch, etc.

[0069] Figure 2 The embodiments of the present application do not limit the number of devices included in the communication system, which is only a schematic diagram. Figure 2 The connection mode between the switches and nodes shown in the figure is only a schematic illustration, and in actual applications, the switches and nodes can also be connected in other modes, which are not limited by the present application.

[0070] Optionally, the communication system can also include other devices, such as a controller. The controller is configured to manage the network devices in the communication system and ensure that the network devices work normally.

[0071] In the present application, the controller can send a power saving policy to the network device. For example, after the controller sends an initial power saving policy to the network device, the power saving policy is adjusted according to the performance data periodically reported by the network device, and the updated power saving policy is periodically sent to the network device. In order to improve the accuracy of the power saving policy.

[0072] Optionally, the controller can be an independent server linked with the network device through a dedicated control channel. The controller can also be deployed as a module on the network device. After the network device is managed by the controller, the controller can start the power saving policy configuration and distribution process, that is, send the power saving policy to the network device managed by the controller.

[0073] Next, the embodiments of the control method provided by the present application will be described in detail in combination with the drawings.

[0074] Figure 3 A flowchart of a control method provided by the present application is shown. The method can be executed by a first chip (such as a control chip) of a network device. The control chip runs a power saving agent to implement power saving operations on the network device. The network device can be, for example, Figure 2The communication system shown in the figure includes switches, core switches, aggregation switches, etc. Here, the state of a channel corresponding to a port is taken as an example for illustration. As shown in the figure, Figure 3 The method includes the following steps.

[0075] Step 310: Obtain the energy-saving policy of the network device.

[0076] The energy-saving policy is used to indicate the condition for opening or closing a channel corresponding to a port in the network device. Understandably, when a channel corresponding to a port in the network device is closed, the channel corresponding to the port cannot transmit data, thus reducing the energy consumption of the network device; conversely, when a channel corresponding to a port in the network device is opened, data transmission based on the channel corresponding to the port is restored.

[0077] In some embodiments, the memory of the network device stores a preconfigured energy-saving policy. The network device can obtain the preconfigured energy-saving policy from the memory.

[0078] In other embodiments, the network device receives the energy-saving policy sent by the controller. If the controller is an independent server, the energy-saving policy is received through a link between the network device and the controller. If the controller is deployed in the network device, the controller can be a general-purpose chip with control functions, such as a CPU.

[0079] For example, as shown in the figure, Figure 4 The network device sends configuration information to the controller (step 410). The configuration information includes the number of ports contained in the network device, the number of channels corresponding to the ports, the address, and the device identification. The device identification includes at least one of the device model, the device manufacturer, and the device name.

[0080] The controller generates the energy-saving policy of the network device according to the configuration information and sends the energy-saving policy to the network device (step 420). For example, the controller determines the characteristics of the network device according to the configuration information to generate the energy-saving policy. The characteristics of the network device include the position of the network device in the network, the scale of the network device (such as the number of ports and the number of channels), the configuration of the network device (such as the direction of the channel, the configuration of the port, the routing configuration, and the quality of service), and the importance of the network device in the network, etc. For example, the network device is a device used to build a core layer, the controller generates the energy-saving policy to ensure the reliability and low latency of the network device, avoids affecting the reliability and low latency of the network device due to the reduction of the energy consumption of the network device, and balances the energy consumption, reliability, and low latency of the network device.

[0081] In some embodiments, the energy saving policy of the network device generated by the controller according to the configuration information can be an initial energy saving policy. In order to improve the accuracy of the network device adjusting the state of the channel corresponding to the port according to the energy saving policy and effectively reduce the energy consumption of the network device, the network device can also periodically obtain the energy saving policy. For example, the network device reports the performance data of the network device to the controller (step 430). The controller updates the initial energy saving policy according to the performance data of the network device to generate an updated energy saving policy. The controller sends the updated energy saving policy to the network device (step 440).

[0082] Optionally, the energy saving policy comprises at least one of an energy saving enabling identifier, a port identifier, a port priority, a priority of a channel corresponding to the port, a detection frequency, and a threshold. The energy saving enabling identifier is used to indicate whether to enable energy saving. For example, the energy saving enabling identifier is 1, indicating that energy saving is enabled; the energy saving enabling identifier is 0, indicating that energy saving is not enabled. The port identifier is used to uniquely indicate the port. The detection frequency is used to indicate the number of times of periodically detecting the performance of the network device per unit time. The threshold is used to indicate the limit for starting energy saving based on the performance of the network device. The threshold comprises a minimum threshold and a maximum threshold. The minimum threshold is used to indicate closing the port or the channel corresponding to the port of the network device. The maximum threshold is used to indicate restarting the port or the channel corresponding to the port of the network device.

[0083] Step 320, obtaining the performance data of the network device.

[0084] The performance data is used to indicate the running characteristics of the port of the network device in the current period. For example, the performance data comprises at least one of a data interface bandwidth utilization rate, a queue depth, and a packet delay. The interface bandwidth utilization rate represents the bandwidth utilization condition of the interface bandwidth receiving information or sending information per unit time. For example, the interface bandwidth is 100M, the actual traffic is 90M, and the utilization rate is 90%. When the interface bandwidth utilization rate is large, it indicates that the traffic is large; when the interface bandwidth utilization rate is small, it indicates that the traffic is small. The queue depth is used to indicate the number of packets to be transmitted. The packet delay is used to indicate the time length of the packet from the sending end to the receiving end. The port state comprises an energy saving state and an active state.

[0085] In some embodiments, the network device obtains the performance data of the network device according to the energy saving policy. For example, the network device determines that the energy saving enabling identifier included in the energy saving policy indicates to enable energy saving, and then the network device obtains the performance data of each port in the network device.

[0086] For another example, the network device periodically obtains the performance data of each port in the network device according to the detection frequency included in the energy saving policy.

[0087] For another example, the energy saving policy comprises a port identifier, and the network device obtains the performance data of the port indicated by the port identifier.

[0088] For example, energy-saving strategies include port prioritization, where network devices obtain port performance data based on port priority.

[0089] Step 330: Adjust the status of the channel corresponding to the port based on the energy-saving strategy and performance data.

[0090] When the port's performance data meets the conditions indicated by the energy-saving policy, the status of the corresponding channel in the network device is adjusted based on the port's performance data. The status includes being turned on or off.

[0091] For example, the condition indicates whether the interface bandwidth utilization rate exceeds an interface bandwidth utilization rate threshold. When the port's interface bandwidth utilization rate is less than or equal to a first interface bandwidth utilization rate threshold, it indicates that the port load is low and there is little data to be transmitted, so at least one channel corresponding to the port is closed. When the port's interface bandwidth utilization rate is greater than or equal to a second interface bandwidth utilization rate threshold, it indicates that the port load is high and there is a lot of data to be transmitted, so at least one channel corresponding to the port is opened. The first interface bandwidth utilization rate threshold is less than the second interface bandwidth utilization rate threshold. The first interface bandwidth utilization rate threshold can be the minimum threshold for closing a channel. The second interface bandwidth utilization rate threshold can be the maximum threshold for opening a channel.

[0092] For example, a condition indicates whether the queue depth of a port exceeds a queue depth threshold. Another example is whether a condition indicates whether the packet latency of a port exceeds a latency threshold.

[0093] In some embodiments, adjusting the state of the channel corresponding to the port may refer to adjusting the state of the corresponding SERDES. The state of the SERDES includes an active state and a power-saving state. The active state of the SERDES includes a working state and a non-working state. The working state indicates that the SERDES is converting parallel signals into serial signals, or vice versa; the channel corresponding to the SERDES is transmitting data normally, but the SERDES consumes relatively high power. The non-working state indicates that the SERDES is not transmitting data and is in an idle state. The power-saving state of the SERDES indicates that the SERDES is in a low-power state, and the SERDES disables the function of transmitting parallel and serial signals.

[0094] For example, such as Figure 5 As shown, closing the channel corresponding to the port transitions the SERDES state from active to power-saving mode, and from high-power to low-power mode. Therefore, by adjusting the SERDES state to power-saving mode, its power consumption is reduced, thus lowering the power consumption of the network device. Opening the channel corresponding to the port transitions the SERDES state from power-saving to active mode, resuming data transmission.

[0095] For example, a first port of a first network device is connected with a second port of a second network device through a link. The first network device sends data to the second network device through the first port. The second network device receives the data sent by the first network device through the second port. For the sending side, when the performance data of the first port meets the closing condition indicated by the energy saving policy, a lane in the link between the first port and the second port is closed, that is, the serdes in the network device where the first port is located is adjusted to be in the energy saving state, and no data is transmitted on the lane. For the receiving side, when the performance data of the second port meets the closing condition indicated by the energy saving policy, the serdes in the network device where the second port is located is adjusted to be in the energy saving state, so that the network devices do not need to negotiate with each other and do not need to be controlled by the controller. Only the opening or closing of the lane corresponding to the port of the network device needs to be adjusted based on the performance of the network device, thereby simplifying the control process of the lane corresponding to the port of the network device, quickly and timely adjusting the closing of the lane of the network device, and reducing the energy consumption of the network device.

[0096] It should be noted that the link corresponding to the port includes a plurality of lanes, and the plurality of lanes can form a lane group. When the performance data of the port meets the closing condition indicated by the energy saving policy, at least one lane in the lane group is closed according to a preset rule. For example, at least one lane is randomly selected from the lane group, and the selected at least one lane is closed. For another example, at least one lane is closed according to the order from small to large of the identifiers of the lanes in the lane group. For another example, at least one lane is randomly selected from the lane group according to the performance data of the port, and the selected at least one lane is closed, while ensuring the transmission requirement of the to-be-transmitted data.

[0097] For example, a link includes four lanes, and two lanes are needed to transmit the to-be-transmitted data. Two lanes are randomly selected from the four lanes, and the two lanes are closed.

[0098] In addition, in order to avoid that all the lanes included in the link corresponding to the port are closed, causing the link corresponding to the port to be unable to transmit data, the state of at least one lane in the link corresponding to the port is kept open, ensuring that the link corresponding to the port transmits data, and the port transmits data based on the at least one kept open lane, so as to facilitate judging whether to open the closed lane corresponding to the port according to the performance data of the port.

[0099] When the performance data of the port meets the opening condition indicated by the energy saving policy, the lane in the link corresponding to the port is opened, that is, the serdes corresponding to the lane is adjusted to be in the active state. The number of the opened lanes corresponding to the port is less than or equal to the number of the closed lanes corresponding to the port. At least one lane is randomly opened from the closed lanes corresponding to the port, or at least one lane is opened from the closed lanes corresponding to the port according to the performance data of the port.

[0100] Optionally, for the port receiving data, when the performance data of the port satisfies the opening condition indicated by the energy saving policy, all the closed lanes corresponding to the port are opened. Thus, the port opens these closed lanes to recover the data transmission of these lanes as soon as possible. In some other embodiments, after the at least one lane corresponding to the port is adaptively closed according to the performance data of the port, all the closed lanes corresponding to the port are opened when the timer expires. In some other embodiments, after all the closed lanes corresponding to the port are opened, when the performance data of the port satisfies the closing condition indicated by the energy saving policy, at least one opened lane is closed again, or the energy saving operation is not performed on the opened lanes of the port, so as to receive data based on these opened lanes as soon as possible.

[0101] The above embodiments illustrate that when the performance data of the port satisfies the closing condition indicated by the energy saving policy, the lane corresponding to the port in the network device is closed, that is, the serdes corresponding to the lane is in the energy saving state, and the lane corresponding to the serdes does not transmit data, so as to reduce the power consumption of the network device by forcibly closing the serdes through quickly adjusting the state of the lane.

[0102] In some other embodiments, the non-forced closing of the serdes is used to reduce the power consumption of the network device. When the performance data of the port satisfies the condition indicated by the energy saving policy, the transmission of the training code stream is stopped, so that the port receiving data detects the data transmission exception of the port as soon as possible, and the state of the lane corresponding to the port is adjusted.

[0103] Here, the first network device is taken as the sending end and the second network device is taken as the receiving end as an example for illustration. The first port of the first network device and the second port of the second network device are connected through a first link. The first link includes N lanes.

[0104] When the performance data of the first port satisfies the closing condition indicated by the energy saving policy, it is determined to close the first lane corresponding to the first port, and then the first chip of the first network device instructs the second chip to stop transmitting the training code stream. The first chip is used to perform the energy saving operation according to the control method provided in the present application. The second chip is used to perform the switching function. Further, the first network device closes the first lane corresponding to the first port, that is, adjusts the state of the serdes corresponding to the first lane in the first network device to the energy saving state.

[0105] Since the first network device does not transmit the training code stream to the second network device, the second network device does not receive the training code stream, and then the second network device determines that the second port is abnormal, determines that the performance data of the second port satisfies the closing condition indicated by the energy saving policy, and closes the first lane corresponding to the second port, that is, adjusts the state of the serdes corresponding to the first lane in the second network device to the energy saving state.

[0106] When the performance data of the first port meets the activation conditions indicated by the energy-saving strategy, the first channel corresponding to the first port is activated, and the first network device sends a training stream to the second network device through the first channel corresponding to the first port. When the first channel corresponding to the second port in the second network device is activated, the second port in the second network device receives the training stream based on the first channel, and the data transmission between the first port and the second port through the first channel is resumed.

[0107] In other embodiments, when the port's performance data meets the shutdown conditions indicated by the power-saving strategy, the state of the channel corresponding to the port in the network device is adjusted, instructing the link layer to stop transmitting data. Understandably, the physical layer exerts backpressure on the link layer, preventing the link layer from transmitting data to the physical layer and ensuring that data transmission on the link corresponding to the port is completed, thereby preventing data loss during the shutdown process.

[0108] For example, such as Figure 6 As shown in (a) above, this is a schematic diagram of the channel's state transitions. A channel in the established state indicates normal operation; a channel in the unestablished state indicates an abnormal operation. When the channel is in the unestablished state, it can be trained. Upon successful training, the channel transitions from the unestablished state to the established state. When the channel is in the established state, if training fails, the channel transitions from the established state to the unestablished state. Figure 6 As shown in (b), the physical layer exerts backpressure on the link layer; for example, serdes instructs the link layer to stop transmitting data.

[0109] Optionally, after channel training is complete, the link layer is instructed to cancel backpressure. For example, after adjusting the state of the SERDES corresponding to the first channel in the first network device to an energy-saving state, the link layer is instructed to cancel backpressure.

[0110] The following is combined Figures 7 to 10 A schematic diagram illustrating the overall process of the control method provided in this application. Figure 7 A flowchart illustrating an energy-saving mechanism provided in this application is shown. The network device starts an energy-saving agent and initializes the agent (step 710). The network device scans ports and reads stored configuration information (step 720). The network device reports the configuration information to the controller (step 730). The network device obtains energy-saving policies from the controller (step 740). The network device periodically scans port performance data (step 750). The network device determines whether the port performance data meets the enabling conditions (step 760).

[0111] If the port's performance data does not meet the opening conditions, the network device determines whether the port's performance data meets the closing conditions (step 770).

[0112] When the performance data of the port meets the closing condition, back pressure is implemented to the link layer (step 780). For the sending side, it is determined whether to forcibly close the channel corresponding to the port (step 790). When the channel corresponding to the port is forcibly closed, at least one channel in the channel state corresponding to the port is closed (step 7100). When the channel corresponding to the port is not forcibly closed, the training code stream of at least one channel in the channel state corresponding to the port is closed (step 7110). After waiting for the training to be completed, step 7100 is executed, and the back pressure of the link layer is cancelled (step 7120). For the receiving side, at least one channel in the channel corresponding to the port is closed (step 7130).

[0113] When the performance data of the port meets the opening condition, for the sending side, at least one closed channel in the channel corresponding to the port is opened (step 7140). For the receiving side, all closed channels in the channel corresponding to the port are opened (step 7150), and after waiting for the training to be completed, the channel corresponding to all ports that have not received data after the training is closed (step 7160).

[0114] Figure 8 A flowchart for forcibly closing a channel is provided. 1, the sending side energy saving agent detects that the port load is lower than the energy saving threshold; 2, the sending side energy saving agent notifies the sending side physical layer to implement back pressure; 3, the physical layer back pressures the sending side link layer; 4, the sending side energy saving agent determines that the back pressure takes effect, and after the data transmission is completed; 5, the sending side energy saving agent selects n channels according to the energy saving strategy, and closes the serdes corresponding to the n channels; 6, the serdes corresponding to the n channels are closed; 7, the sending side energy saving agent determines that the closing takes effect; 8, the sending side energy saving agent notifies the sending side physical layer to cancel the back pressure; 9, the physical layer cancels the back pressure to the sending side link layer. 10, the receiving side physical layer detects an exception and starts training; 11, after the training, the receiving side does not receive the training code stream corresponding to the sending side closed channel; 12, the serdes of the channel that does not receive the training code stream is closed; 13, the serdes is closed.

[0115] Figure 9A flowchart of a non-mandatory closed channel provided by the present application is shown. 1, detecting that the port load is lower than the energy saving threshold; 2, the sending side energy saving agent informs the sending side physical layer to implement back pressure; 3, the physical layer back pressures the sending side link layer; 4, the sending side energy saving agent determines that the back pressure takes effect, and after the data transmission is completed; 5, the sending side energy saving agent selects n channels according to the energy saving strategy, and closes the training code stream corresponding to the n channels; 6, closing the training code stream corresponding to the n channels; 7, delaying and waiting for the training to be completed; 8, the training is completed; 9, the sending side energy saving agent informs the sending side physical layer to cancel the back pressure; 10, the physical layer cancels the back pressure to the sending side link layer. 11, the receiving side physical layer detects an exception and starts training; 12, after training, the sending side closed channel corresponds to the receiving side that does not receive training data; 13, closing the serdes of the channel that does not receive training data; 14, the serdes is closed.

[0116] Figure 10 A flowchart of an open channel provided by the present application is shown. 1, the sending side energy saving agent detects that the port load is higher than the energy saving threshold; 2, the sending side energy saving agent selects n channels according to the energy saving strategy, and opens the serdes corresponding to the n channels; 3, opening the serdes corresponding to the n channels; 4, training the n channels; 5, delaying and waiting for the training to be completed; 6, the training is completed; 7, after training, both ends open n channels, but the n channels are in an inoperative state. 8, the receiving side energy saving agent detects that the port load is higher than the energy saving threshold; 9, opening the serdes corresponding to all closed channels; 10, opening the serdes; 11, training all channels; 12, delaying and waiting for the training to be completed; 13, training the channels at both ends. Both ends open the channel and training is successful, and any end at both ends is not opened, and training fails; 14, the training is completed; 15, closing the serdes of the training failed channel; 16, the serdes is closed.

[0117] Regarding Figures 7 to 10 The detailed explanation is described with reference to the above-mentioned embodiments.

[0118] Next, the control method provided by the present application is illustrated by combining Figure 11 and Figure 12 the above-mentioned embodiments.

[0119] Assume network device A includes ports A1 to A4. Network device B includes ports B1 to B4. Port A1 of network device A and port B1 of network device B are connected via link 1. Port A1 is used to send data, and port B1 is used to receive data. Network device A sends data to network device B through port A1, which is transmitted through link 1. Network device B receives data sent by network device A through port B1. Port A2 of network device A and port B2 of network device B are connected via link 2. Port A2 is used to receive data, and port B2 is used to send data. Each link includes two channels.

[0120] Network device A and network device B are configured with energy-saving agents. The energy-saving agents adjust the opening or closing of the corresponding channels on the ports based on the energy-saving policies and performance data obtained from the controller. Figure 11 The closing process of channels (a) to (c) in the figure is the energy-saving process. Figure 11 The opening process of channels (d) to (f) in the figure is the energy enhancement process.

[0121] like Figure 11 As shown in (a), the performance data of port A1 meets the shutdown conditions indicated by the energy-saving policy, so channel 1 corresponding to port A1 is shut down. The performance data of port A4 meets the shutdown conditions indicated by the energy-saving policy, so channel 1 corresponding to port A4 is shut down. The performance data of port B2 meets the shutdown conditions indicated by the energy-saving policy, so channel 2 corresponding to port B2 is shut down.

[0122] like Figure 11 As shown in (b), network device A stops sending the training stream. Network device B does not receive the training stream through channel 1 corresponding to port B1, therefore network device B determines that port B1 is inactive, and the performance data of port B1 meets the shutdown condition indicated by the power-saving policy. After channel 1 corresponding to port A4 and channel 2 corresponding to port B2 are trained, network device B determines that port B4 is inactive, and the performance data of port B4 meets the shutdown condition indicated by the power-saving policy. Network device A determines that port A2 is inactive, meaning that the performance data of port A2 meets the shutdown condition indicated by the power-saving policy.

[0123] like Figure 11 As shown in (c), the performance data of port B1 meets the shutdown conditions indicated by the energy-saving policy, so channel 1 corresponding to port B1 is shut down. The performance data of port B4 meets the shutdown conditions indicated by the energy-saving policy, so channel 1 corresponding to port B4 is shut down. The performance data of port A2 meets the shutdown conditions indicated by the energy-saving policy, so channel 2 corresponding to port A2 is shut down.

[0124] like Figure 11As shown in (d), the performance data of port A1 meets the activation conditions of the energy-saving strategy, thus enabling channel 1 corresponding to port A1. Channel 2 corresponding to port A2 is enabled. Channel 1 corresponding to port B4 is enabled. Channel 1 corresponding to port B1 is enabled.

[0125] like Figure 11 As shown in (e), after port A1 of network device A and port B1 of network device B successfully train through channel 1 of link 1, data transmission resumes. After port A2 of network device A and port B2 of network device B successfully train through channel 2 of link 2, data transmission resumes. Channel 1 corresponding to port A4 of network device A is closed, and training fails.

[0126] like Figure 11 As shown in (f), since no data is received through channel 1 of port B4 after channel 1 of port B4 is opened, channel 1 corresponding to port B4 is closed. Port A4 of network device A and port B2 of network device B are in power-saving state through channel 1 of link 4.

[0127] Figure 12 (a) in the diagram is a schematic diagram of an energy-saving command issuance method provided in this application. The control chip in the network device runs an energy-saving agent application, sending monitoring and commands to the SERDES. When the performance data meets the shutdown conditions indicated by the energy-saving policy, back pressure is applied, shutting down the SERDES corresponding to the channel. Figure 12 (b) in this application is a schematic diagram of an energy-saving instruction issuance method.

[0128] Let network device A be the sender (A end) and network device B be the receiver (B end) for the explanation.

[0129] a) The service stream is transmitted through M channels of serdes;

[0130] b) The energy-saving agent on end A detects that the load on the serdes interface is lower than the preset data traffic threshold;

[0131] c) The physical layer at end A sends a reverse pressure signal to the link layer to stop transmitting data to the physical layer;

[0132] d) The energy-saving agent on end A controls the shutdown of the serdes of n channels on the transmitting side;

[0133] e) Some channels on the B-end receiving side have started retraining the link because they have not received data streams;

[0134] f) The physical layer at end A coordinates with end B to retrain the link;

[0135] g) Re-establish the link between terminals A and B using the (MN) channel;

[0136] h) the physical layer of the A end sends a close back pressure signal to the link layer, and the link layer is ready to transmit data to the physical layer;

[0137] i) the link layer of the B end sends a retransmission request to the link layer of the A end;

[0138] j) the A end starts transmitting data to the B end in response to the retransmission request;

[0139] k) the power saving agent of the A end sends a command to the power saving agent of the B end to control the B end to close the receiving side of the serdes of the un-established N channels.

[0140] The above embodiments are based on the performance data of the ports to adjust the states of the channels corresponding to the ports in the network device, and the energy saving of the channels corresponding to each port is taken as an example for illustration.

[0141] In other embodiments, a plurality of ports included in the network device are taken as a port group, and the energy saving is implemented for the port group as a whole. The difference from the above embodiments is that the performance data of the network device can refer to the performance data of the entire network device. For example, the performance data of the network device includes the sum of the performance data of each port included in the network device. When the performance data of the network device meets the closing condition indicated by the energy saving policy, at least one port of the plurality of ports included in the network device is closed, i.e. all the channels corresponding to the port are closed. For example, the port with the smallest load among the plurality of ports included in the network device is closed. When the performance data of the network device meets the opening condition indicated by the energy saving policy, at least one port of the plurality of ports included in the network device is opened, i.e. at least one closed channel corresponding to the port is opened.

[0142] Optionally, before closing at least one port of the plurality of ports included in the network device, the traffic on the at least one port can be migrated to a port in the network device that does not need to be closed, so as to avoid data transmission failure caused by closing the at least one port.

[0143] It should be noted that the state of at least one port between the sending side network device and the receiving end network device is kept open to ensure the transmission of data between the sending side network device and the receiving end network device, and the sending side network device sends data based on the at least one reserved link, and the receiving end network device receives data based on the at least one reserved link, so as to facilitate the determination of whether to open the closed receiving port according to the performance data of the receiving end network device.

[0144] For specific explanations of adjusting the state of the port, reference can be made to the above description of the state of the channel corresponding to the port of the embodiment, which will not be repeated here.

[0145] For example, as shown in FIG. 1, a flowchart of a control method provided by the present application is shown. Figure 13 For example, as shown in FIG. 1, a flowchart of a control method provided by the present application is shown.

[0146] 1. There are M ports between network device A and network device B. These ports share the traffic between network device A and network device B.

[0147] 2. After network device A detects that its performance data meets the conditions indicated by the energy-saving strategy, it selects N ports from the ports between network device A and network device B to perform energy-saving operations (M>N).

[0148] a) Network device A stops sending traffic on these N ports by modifying the routing or traffic load balancing configuration, and then shuts down the sending direction of these N ports, thus entering a power-saving state.

[0149] b) If the receiving side of the corresponding port of network device B detects a port abnormality, then network device B will stop sending traffic on N ports by modifying the routing or traffic load balancing configuration, and then shut down the sending direction of N ports.

[0150] c) Network device B detects an anomaly on N ports of network device A, and then the receiving side of the N ports of network device B that detected the anomaly is shut down, so that the N ports of network device A completely enter the power-saving state.

[0151] d) If network device B determines that energy-saving measures are also needed for the ports between network device A and network device B based on the load, then network device B will shut down the receiving side of the N ports that are in an abnormal state.

[0152] 3. If network device A and network device B detect that the bidirectional traffic load between them exceeds the power-up threshold, then power-up operations will be performed on the N interfaces between network device A and network device B that are in power-saving mode:

[0153] a) Network device A will turn on the interfaces that were turned off for energy saving.

[0154] b) Network device B will turn on the interfaces that were turned off for energy saving.

[0155] It is understood that, in order to achieve the functions in the above embodiments, the controller includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0156] The above text combines Figures 1 to 13 The control method provided according to this application is described in detail below, in conjunction with... Figure 14, describes a control device provided by the present application.

[0157] Figure 14 The control device provided by the present application is shown in a structural diagram. The control device can be used to realize the functions of the controller in the method embodiments described above, and thus can also realize the beneficial effects possessed by the method embodiments described above. In the present embodiment, the control device can be a network device as shown in Figure 3 , or a module (such as a chip) applied to a server.

[0158] As shown in Figure 14 , the control device 1400 includes a communication module 1410, an energy consumption control module 1420, and a storage module 1430. The control device 1400 is used to realize the functions of the network device in the method embodiments shown in Figure 3 .

[0159] The communication module 1410 is used to obtain an energy saving policy of the network device, and the energy saving policy is used to indicate the conditions of opening or closing of a port or a channel corresponding to the port of the network device. For example, the communication module 1410 is used to perform step 310 in Figure 3 .

[0160] The energy consumption control module 1420 is used to obtain performance data of the network device, and adjust the state of the port or the channel corresponding to the port according to the energy saving policy and the performance data, the state including opening or closing, and the performance data is used to indicate the running characteristics of the port of the network device in a current period. For example, the energy consumption control module 1420 is used to perform steps 320 and 330 in Figure 3 .

[0161] The storage module 1430 is used to store the energy saving policy, the performance data, and the state, so as to facilitate the energy consumption control module 1420 to adjust the state of the port or the channel corresponding to the port according to the energy saving policy and the performance data.

[0162] It should be understood that the control device 1400 of the present embodiment can be realized by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), and the above-mentioned PLD can be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. The control device 1400 can also be realized by software. Figure 3The control method shown, and each module thereof, can also be a software module. The control device 1400 and each module thereof can also be a software module.

[0163] The control device 1400 according to the embodiments of the present application can correspond to performing the methods described in the embodiments of the present application, and the above and other operations and / or functions of each unit in the control device 1400 are respectively implemented to realize the corresponding flow of each method in the embodiments of the present application. For brevity, they will not be repeated here. Figure 3 The control device 1400 according to the embodiments of the present application can correspond to performing the methods described in the embodiments of the present application, and the above and other operations and / or functions of each unit in the control device 1400 are respectively implemented to realize the corresponding flow of each method in the embodiments of the present application. For brevity, they will not be repeated here.

[0164] Figure 15 A structure schematic diagram of a network device 1500 according to the embodiments of the present application is provided. As shown in the figure, the network device 1500 includes a control chip 1510, a bus 1520, a memory 1530, a communication interface 1540, a memory unit 1550 (which can also be referred to as a main memory unit), and a switching chip 1560 including a serial-parallel converter 1561. The control chip 1510, the switching chip 1560, the memory 1530, the memory unit 1550, and the communication interface 1540 are connected through the bus 1520.

[0165] It should be understood that in the embodiments, the control chip 1510 can be a CPU, and the control chip 1510 can also be other general-purpose processors, digital signal processors (DSPs), ASICs, FPGAs, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0166] The processor can also be a graphics processing unit (GPU), a neural network processing unit (NPU), a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of programs of the present application scheme.

[0167] The communication interface 1540 is used to realize the communication between the network device 1500 and external devices or instruments. In the embodiments, the network device 1500 is used to realize the functions of the network device shown, and the communication interface 1540 is used to obtain energy saving strategies and performance data, so that the control chip 1510 adjusts the state of the port or the channel corresponding to the port according to the energy saving strategies and the performance data, and the state includes opening or closing. Figure 3

[0168] ​The bus 1520 can include a pathway for transferring information between the above-mentioned components (e.g., the control chip 1510, the memory unit 1550, the storage 1530, and the switching chip 1560). The bus 1520 can include a data bus, a power supply bus, a control bus, and a status signal bus, etc. in addition to the data bus. However, for the purpose of clarity, all the buses are marked as the bus 1520 in the figure. The bus 1520 can be a Peripheral Component Interconnect Express (PCIe) bus, or an extended industry standard architecture (EISA) bus, a unified bus (Ubus or UB), a compute express link (CXL), a cache coherent interconnect for accelerators (CCIX), etc. The bus 1520 can be divided into an address bus, a data bus, a control bus, etc.

[0169] As an example, the control chip 1510 can be a multi-CPU control chip. The control chip here can refer to one or more devices, circuits, and / or computing units for processing data (e.g., computer program instructions).

[0170] It is worth noting that, Figure 15 In the network device 1500, only one control chip 1510 and one storage 1530 are taken as an example. Here, the control chip 1510 and the storage 1530 are respectively used to indicate a type of device or equipment. In specific embodiments, the number of each type of device or equipment can be determined according to business needs.

[0171] The memory unit 1550 can be a pool of volatile memory or a pool of non-volatile memory, or can include both volatile and non-volatile memory. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM) used as an external cache. By way of example and not limitation, many forms of RAM can be used, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM). The memory unit 1550 is used to store energy saving strategies, performance data, state, and the like.

[0172] The memory 1530 can correspond to the storage medium used to store energy saving strategies, performance data, state, and the like in the above method embodiments, for example, a disk such as a mechanical hard disk or a solid state disk.

[0173] The above network device 1500 can be a general-purpose device or a special-purpose device. For example, the network device 1500 can be an edge device (for example, a box carrying a chip with processing capability) and the like. Alternatively, the network device 1500 can also be a server or other device with computing capability.

[0174] It should be understood that the network device 1500 according to the present embodiment can correspond to the control apparatus 1400 in the present embodiment, and can correspond to the corresponding subject performing any method according to the present embodiment, and the above and other operations and / or functions of each module in the control apparatus 1400 are respectively for realizing the corresponding flow of each method in the present embodiment, and for brevity, will not be repeated here. Figure 3 Figure 3 It should be understood that the network device 1500 according to the present embodiment can correspond to the control apparatus 1400 in the present embodiment, and can correspond to the corresponding subject performing any method according to the present embodiment, and the above and other operations and / or functions of each module in the control apparatus 1400 are respectively for realizing the corresponding flow of each method in the present embodiment, and for brevity, will not be repeated here.

[0175] ​The embodiment of the application provides a chip, comprising: a processor and a power supply circuit; wherein the power supply circuit is used for supplying power for the processor; and the processor is used for executing the operation steps of the control method in the method embodiment.

[0176] The method steps in the embodiment can be realized by a hardware mode or a mode of executing software instructions by a control chip. The software instructions can be composed of corresponding software modules, and the software modules can be stored in a random access memory (RAM), a flash memory, a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a register, a hard disk, a mobile hard disk, a CD-ROM or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a computing device. Of course, the processor and the storage medium can also exist as discrete components in the computing device.

[0177] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer programs or instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are performed. The computer can be a general purpose computer, a special purpose computer, a computer network, a network device, a user equipment or other programmable apparatus. The computer programs or instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer programs or instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through wired or wireless manner. 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, data center and the like integrated with one or more available media. The available media can be a magnetic medium, for example, a floppy disk, a hard disk, a magnetic tape; or an optical medium, for example, a digital video disc (digital video disc, DVD); or a semiconductor medium, for example, a solid state disk (solid state drive, SSD). The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A control method characterized by, The method is performed by a first chip of a network device, and the method comprises: obtaining an energy saving policy of the network device, the energy saving policy being used to indicate a condition of turning on or off of a port of the network device or a channel corresponding to the port; obtaining performance data of the network device, the performance data being used to indicate a running feature of the port of the network device in a current period; adjusting a state of the port or the channel corresponding to the port according to the energy saving policy and the performance data, the state comprising turning on or off.

2. The method of claim 1, wherein, The adjusting of the state of the port or the channel corresponding to the port according to the energy saving policy and the performance data comprises: when the performance data satisfies a turning-off condition indicated by the energy saving policy, adjusting the state of at least one port of the network device or the state of the channel corresponding to the port according to the performance data.

3. The method of claim 2, wherein, The performance data comprises at least one of an interface bandwidth occupancy and a queue depth.

4. The method according to any one of claims 1-3, characterized in that, The adjusting of the state of the channel corresponding to the port comprises: migrating a state of a serial-parallel converter corresponding to the channel from an active state to an energy saving state.

5. The method of claim 4, wherein, The network device further comprises a second chip, the port is used to send data, and the adjusting of the state of the channel corresponding to the port comprises: indicating the second chip to stop sending a training code stream, and adjusting the state of the channel corresponding to the port.

6. The method of claim 4, wherein, The port is used to receive data, and the adjusting of the state of the channel corresponding to the port comprises: when no training code stream is received, adjusting the state of the channel corresponding to the port.

7. The method according to any one of claims 1-5, characterized in that, The port is used to send data, and the adjusting of the state of the channel corresponding to the port comprises: when a link layer stops transmitting data, adjusting the state of the channel corresponding to the port.

8. The method according to any one of claims 1 to 7, characterized in that, The state of the port or the channel corresponding to the port is turned off, and the method further comprises: when the performance data satisfies a turning-on condition indicated by the energy saving policy, turning on the port or the channel corresponding to the port.

9. The method of claim 8, wherein, The turning on of the state of the channel corresponding to the port comprises: migrating a state of a serial-parallel converter corresponding to the channel from an energy saving state to an active state.

10. The method according to any one of claims 1-9, characterized in that, The obtaining of the energy saving policy of the network device comprises: obtaining the energy saving policy of the network device according to a feature of the network device, the feature of the network device comprising at least one of a position of the network device in a network, a scale of the network device, a configuration of the network device, and performance data of the network device.

11. A control device characterized by comprising: The control device comprises a module for performing the operation steps of the method in any one of claims 1-10.

12. A chip, characterized by The chip comprises a processor and a power supply circuit; the power supply circuit is used to supply power to the processor; and the processor is used to perform the operation steps of the method in any one of claims 1-10.

13. A network device, comprising: The network device comprises a memory and a processor, the memory is used to store a set of computer instructions; and when the processor executes the set of computer instructions, the operation steps of the method in any one of claims 1-10 are performed.

14. A communication system, characterized by The communication system comprises a first network device configured to perform the operation steps of the method of any one of claims 1-5, 7-10, and a second network device configured to perform the operation steps of the method of any one of claims 1-4, 6-10.

15. The system of claim 14, wherein, The communication system further comprises a controller configured to send a power saving policy to the first network device and the second network device, the power saving policy being configured to indicate a condition for turning on or off a port of the first network device and the second network device or a lane corresponding to the port.

Images