CN117194172B

CN117194172B - Network card power supply control method and related device

Info

Publication number: CN117194172B
Application number: CN202311314198.6A
Authority: CN
Inventors: 孙文锋; 张艳平
Original assignee: Zhuhai Shininda Technology Co ltd
Current assignee: Zhuhai Shininda Technology Co ltd
Priority date: 2023-10-11
Filing date: 2023-10-11
Publication date: 2024-03-22
Anticipated expiration: 2043-10-11
Also published as: CN117194172A

Abstract

The application discloses a network card power supply control method and a related device, wherein the method comprises the following steps: performing a first response process when triggered based on a timer signal of a first preset period; performing a second power supply mode adjustment process by combining the temperature information set and the current scheduling request quantity; the first response processing performed when triggered by the timer signal based on the first preset period specifically includes: acquiring a task request queue to be scheduled, wherein the task request queue to be scheduled comprises a plurality of data transmission requests; generating a time zone request subset based on the task request queue to be scheduled to update the time zone request set; and performing first power supply mode adjustment processing based on the current scheduling request quantity. The current scheduling request quantity is judged, the power supply mode of the first intelligent network card in the first server is adjusted in time, and the dormancy switching is performed on the network card ID in time by judging the high-temperature duration, so that the reliability and the accuracy of the network card power supply control are improved.

Description

Network card power supply control method and related device

Technical Field

The present disclosure relates to the field of server technologies, and in particular, to a network card power supply control method and a related device.

Background

With the continuous development of computer network technology, a network card has become an important interface for connecting a computer with a network. However, the inaccurate network card power control results in unstable network connection, and further, data packets are lost due to network connection interruption, which is a trouble for servers for large-scale data processing. In the traditional network card power supply technology, once the support of the network card power supply cannot be accurately controlled, the working state of the network card is easily unstable, so that the stability and the reliability of network connection are affected. Accordingly, there is a need for improved network card power control to provide more accurate control in response to the network card power control issues.

Disclosure of Invention

In order to overcome the defects and shortcomings in the prior art, the application provides a network card power supply control method and a related device, and reliability and accuracy of network card power supply control are improved.

In order to achieve the above purpose, the present application adopts the following technical scheme:

according to one aspect of the application, a network card power supply control method is provided and applied to a server, wherein the server is a server cluster formed by a first server and a plurality of second servers, the first server is respectively connected with the plurality of second servers, the first server comprises a first power supply controller and a plurality of first intelligent network cards, each second server comprises a second power supply controller and a plurality of second intelligent network cards, the first server is used for receiving all data transmission requests initiated by a terminal device and carrying out flow scheduling on all data transmission requests and then distributing the data transmission requests to the designated second server, the second server is used for responding to the distributed data transmission requests, the first power supply controller is used for controlling a power supply mode of each first intelligent network card, and the second power supply controller is used for controlling a power supply mode of a second intelligent network card belonging to the same second server;

The method comprises the following steps: performing a first response process when triggered based on a timer signal of a first preset period; performing a second power supply mode adjustment process by combining the temperature information set and the current scheduling request quantity; the step of performing the first response processing when triggered by the timer signal based on the first preset period specifically includes: acquiring a task request queue to be scheduled, wherein the task request queue to be scheduled comprises a plurality of data transmission requests, each data transmission request corresponds to one piece of flow information, each piece of flow information comprises a user ID, a request time and an application identifier, and deleting queue elements after each data transmission request is scheduled to finish updating; generating a time zone request subset based on a task request queue to be scheduled so as to update the time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval; and performing first power supply mode adjustment processing based on the current scheduling request quantity.

Preferably, the step of performing the first power supply mode adjustment processing based on the current number of scheduling requests specifically includes: aiming at a first intelligent network card with the function type of management network card identification:

if the number of the current scheduling requests is smaller than the number of the first preset scheduling requests, setting the first intelligent network card with the function type of the management network card identifier as an energy-saving power supply mode;

if the current scheduling request number exceeds the first preset scheduling request number and is smaller than the second preset scheduling request number, setting a first intelligent network card with the function type of management network card identification as a first load power supply mode;

if the number of the current scheduling requests exceeds the number of the second preset scheduling requests, setting the first intelligent network card with the function type of the management network card identifier as a second load power supply mode;

the power supply power in the energy saving power supply mode is smaller than the power supply power in the first load power supply mode, and the power supply power in the first load power supply mode is smaller than the power supply power in the second load power supply mode.

Preferably, the step of performing the second power supply mode adjustment processing by combining the temperature information set and the current scheduling request number specifically includes: if the number of the current scheduling requests exceeds the second preset scheduling request number, calculating the period continuous times, wherein the period continuous times are used for indicating the statistics times that the number of the current scheduling requests exceeds the second preset scheduling request number, and taking a first preset period as one statistics time, and the number of the current scheduling requests is the number of flow information in a time zone request subset closest to the current time;

If the period continuous times exceeds the preset times, acquiring a temperature information set, wherein the temperature information set is composed of a plurality of temperature information, each temperature information corresponds to a time value, the temperature information comprises all network card IDs and network card detection temperatures corresponding to the network card IDs, each first intelligent network card association or each second intelligent network card association is respectively associated with a unique network card ID, and further, different network cards are positioned according to the network card IDs;

traversing the temperature information set, screening out network card IDs of which the network card detection temperature exceeds a preset temperature threshold based on the current time value, judging whether the high-temperature duration exceeds a preset high-temperature duration threshold, and performing dormancy switching on the network card IDs of which the high-temperature duration exceeds the preset high-temperature duration threshold;

if the period continuous times do not exceed the preset times, the processing is not performed.

Preferably, the step of performing sleep switching on the network card ID with the high-temperature duration exceeding the preset high-temperature duration threshold specifically includes: if the network card ID corresponds to the attribution identifier of the first server, switching the first intelligent network card with the function type of redundant network card identifier and the first intelligent network card corresponding to the network card ID, and setting a dormancy identifier and a first preset dormancy time;

If the network card ID has the attribution identification of the second server, determining a second server serial number where the network card ID is located according to the attribution identification of the second server, mutually switching a second intelligent network card which belongs to the second server serial number and has the function type of redundant network card identification and a second intelligent network card corresponding to the network card ID, and setting a dormancy identification and a second preset dormancy time;

in the first server, the function types of the first intelligent network card comprise a management network card identifier, a public network card identifier, an intranet card identifier and a redundant network card identifier, and in any one of the second servers, the function types of the second intelligent network card comprise the public network card identifier, the intranet card identifier and the redundant network card identifier, wherein the management network card identifier is an identifier for indicating that the network card is used for scheduling, the public network card identifier is an identifier for indicating that the network card is used for accessing an external network, the intranet card identifier is an identifier for indicating that the network card is used for accessing an internal network, and the redundant network card identifier is an identifier for indicating that the network card is used for backup.

Preferably, for a first intelligent network card with a sleep identifier, when the sleep identifier is set, a first timer is set according to a first preset sleep time, and when the first preset sleep time is triggered, the sleep identifier and the first preset sleep time are cleared for the first intelligent network card by the first timer, so that the first intelligent network card is timely dormant;

And setting a second timer according to a second preset dormancy time when the dormancy identification is set for the second intelligent network card with the dormancy identification, and clearing the dormancy identification and the second preset dormancy time through the second timer when the second preset dormancy time is triggered, so that the second intelligent network card is dormant in time.

Preferably, the method further comprises:

acquiring application operation interval information of all users, wherein the application operation interval information of each user contains a plurality of groups of application operation interval sub-information, the plurality of groups of application operation interval sub-information is obtained by classifying the application operation interval information based on application identifiers, and each group of application operation interval sub-information comprises a plurality of protocol average end times;

when the timer signal based on the second preset period is triggered, second response processing is carried out, and therefore network card power supply conditions of all second servers are timely adjusted;

the second preset period is set according to the link bandwidth occupancy rate of each second server, namely:

wherein θ is _k Represents the link bandwidth occupancy of the kth second server, P represents the number of second servers,accumulated value representing link bandwidth occupancy of P second servers, +. >Represents average of the link bandwidth occupancy rate of each second server, T represents a second preset period, θ _ref For a preset reference period threshold value, T ₁ The link bandwidth occupancy rate of each second server at average for the second preset period is smaller than the preset parameterAssignment of test period threshold value, T ₂ And assigning a value for the second preset period when the link bandwidth occupancy rate of each second server exceeds a preset reference period threshold value.

Preferably, the average ending time of the protocol is an average time difference between the protocol initiating time and the application ending time of any kind of protocol, and is specifically expressed as:

wherein the method comprises the steps ofProtocol average initiation time value, T, expressed as the ith user ID triggering the nth class protocol at the kth use of the mth class application identity _k’ (i, m) is expressed as a value of a moment when the ith user ID is finished at the kth use of the mth class application identifier, and L (i, m) is a total number of times the ith user ID is used at the mth class application identifier>And the average protocol ending time of the ith user ID when the mth class application identifier triggers the nth class protocol is represented.

Preferably, the second response processing is performed when triggered by a timer signal based on a second preset period, and specifically includes:

Traversing all the second servers, and acquiring a plurality of request processing lists of the second server traversed currently, wherein each request processing list corresponds to one second intelligent network card of the second server and is used for managing data transmission requests of the corresponding second intelligent network card;

extracting the protocol type of the data transmission request with the first requirement identifier from the request processing list to determine the sequence numbers of all the protocols of the data transmission request with the first requirement identifier;

based on the sequence numbers of the protocols of the data transmission requests, screening the maximum value of the average ending time of the protocols of the data transmission requests with the first requirement identifiers in each request processing list respectively to obtain a plurality of application ending estimated times, wherein each second intelligent network card in the second server traversed at present corresponds to one application ending estimated time one by one;

if the application ending estimated time is smaller than a first preset waiting threshold, setting a second intelligent network card corresponding to the application ending estimated time as an energy-saving power supply mode, if the application ending estimated time exceeds the first preset waiting threshold and is smaller than a second preset waiting threshold, setting the second intelligent network card corresponding to the application ending estimated time as a third load power supply mode, if the application ending estimated time exceeds the second preset waiting threshold, setting the second intelligent network card corresponding to the application ending estimated time as a fourth load power supply mode, wherein the first preset waiting threshold is smaller than the second preset waiting threshold, the power supply power in the energy-saving power supply mode is smaller than the power supply power in the third load power supply mode, the power supply power in the third load power supply mode is smaller than the power supply power in the fourth load power supply mode, the application ending estimated time is used for reflecting the ending time of a data transmission request in a request processing list, and if the application ending estimated time is longer, the second intelligent network card corresponding to the application ending estimated time is larger, the subsequent data transmission request in the request processing list is continuously reflected.

Preferably, in the process of the first server performing traffic scheduling, the method specifically includes:

classifying all data transmission requests based on application identifiers, if the application identifiers of the data transmission requests correspond to first-class traffic demands, setting first demand identifiers for the data transmission requests, and if not, setting second demand identifiers for the data transmission requests, wherein the first-class traffic demands are demand states requiring a large number of continuous data transmission processes in a network link, and the second-class traffic demands are demand states requiring a small number of short-time data transmission processes in the network link;

classifying all the second servers based on the link bandwidth occupancy rate, setting a first scheduling identifier if the link bandwidth occupancy rate of the second servers is lower than the preset bandwidth occupancy rate, otherwise setting a second scheduling identifier;

summarizing all second servers with the first scheduling identification to form a first circular distribution queue, and summarizing all second servers with the second scheduling identification to form a second circular distribution queue;

and scheduling all the data transmission requests in sequence, judging whether the data transmission requests have first demand identifiers or not if the data transmission requests are first demand conditions, distributing the data transmission requests through a first circular distribution queue if the data transmission requests have the first demand identifiers, distributing the data transmission requests through a second circular distribution queue if the data transmission requests lack the first demand identifiers, distributing the data transmission requests according to a second server distributed when the data transmission requests are first scheduling conditions if the data transmission requests are not first demand conditions, wherein the first demand conditions are protocol requests initiated by a user when the user starts to use the application, and resetting the first demand conditions after the user uses the application.

According to another aspect of the present application, there is provided a network card power supply control device, including: the power supply mode association adjustment processing module comprises a cycle continuous time calculation sub-module, a temperature information set acquisition sub-module and a dormancy switching sub-module;

the first response processing module is used for performing first response processing when triggered by a timer signal based on a first preset period;

the request queue obtaining submodule is used for obtaining a task request queue to be scheduled;

the time zone request subset generating sub-module is used for generating a time zone request subset based on a task request queue to be scheduled by a user so as to update a time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval;

The power supply mode adjustment processing submodule is used for performing first power supply mode adjustment processing based on the current scheduling request quantity;

the power supply mode association adjustment processing module is used for carrying out second power supply mode adjustment processing by combining the temperature information set and the current scheduling request quantity;

the period continuous time calculation sub-module is used for calculating period continuous times when judging that the current scheduling request number exceeds the second preset scheduling request number, wherein the period continuous times are used for representing the statistics times that the current scheduling request number exceeds the second preset scheduling request number, and the current scheduling request number is the number of flow information in the time zone request subset closest to the current time by taking a first preset period as one statistics time;

the temperature information set obtaining submodule is used for obtaining a temperature information set if the period continuous times exceed preset times, if the period continuous times do not exceed the preset times, the temperature information set is not processed, the temperature information set is composed of a plurality of temperature information, each temperature information corresponds to a time value, the temperature information comprises all network card IDs and network card detection temperatures corresponding to each network card ID, each first intelligent network card association or each second intelligent network card association is respectively associated with a unique network card ID, and further different network cards are positioned according to the network card IDs;

The dormancy switching sub-module is used for traversing the temperature information set, screening out network card IDs of which the network card detection temperature exceeds a preset temperature threshold value based on the current time value, judging whether the high-temperature duration exceeds a preset high-temperature duration threshold value, and performing dormancy switching on the network card IDs of which the high-temperature duration exceeds the preset high-temperature duration threshold value;

the application operation interval information acquisition module is used for acquiring application operation interval information of all users, wherein the application operation interval information of each user contains a plurality of groups of application operation interval sub-information, the plurality of groups of application operation interval sub-information is obtained by classifying the application operation interval information based on application identifiers, and each group of application operation interval sub-information comprises a plurality of protocol average ending times;

the second response processing module is used for performing second response processing when triggered based on a timer signal of a second preset period, so as to timely adjust network card power supply conditions of all the second servers.

Compared with the prior art, the application has the following advantages and beneficial effects:

(1) The current scheduling request quantity is judged by combining the first preset scheduling request quantity and the second preset scheduling request quantity, so that the power supply mode of the first intelligent network card with the management network card identifier in the first server is timely adjusted, the self-adaptive adjustment level of the power supply mode of the network card of the first server when the first server requests scheduling is improved, and the accuracy of the power supply control of the network card is improved.

(2) The traditional network card power supply technology does not fully consider the overheat problem of the network card, the normal operation of the network card is easy to be influenced when the network card is operated in a high-temperature environment for a long time, the network card IDs with the high-temperature duration exceeding the preset high-temperature duration threshold value are subjected to dormant switching, the function types corresponding to the two network card IDs are mutually exchanged through the mutual switching and the function types are used as redundant network card identifiers for switching, so that the first intelligent network card or the second intelligent network card corresponding to the network card ID of the original function type executes the dormant power supply mode, the subsequent working state of the first intelligent network card or the second intelligent network card is prevented from being influenced by the continuous high temperature, reliable overheat protection measures are adopted for all intelligent network cards, the unstable working state of the network card caused by network card overheat is avoided, and the reliability of network card power supply control is improved.

(3) According to the use conditions of different applications on the terminal equipment by a user, more accurate power supply control is realized on a second intelligent network card of a second server in time by analyzing the average ending time of a protocol, the power supply mode of the second intelligent network card is adjusted in time, the third load power supply mode or the fourth load power supply mode is utilized to provide enough power supply support, the energy-saving power supply mode is utilized to reduce the power supply consumption of the network card, the self-adaptive regulation level of the power supply control of the second server is improved, and the stability and the reliability of the second server in responding to a data transmission request are improved.

Drawings

Fig. 1 is a flow chart of a network card power supply control method provided in embodiment 1 of the present application;

FIG. 2 is a schematic diagram showing steps of a first response process in embodiment 1 of the present application;

fig. 3 is a schematic diagram of steps of performing a first power supply mode adjustment process based on the current number of scheduling requests in embodiment 1 of the present application;

fig. 4 is a schematic diagram of steps of performing a second power supply mode adjustment process according to embodiment 1 of the present application by combining a temperature information set and a current scheduling request number;

FIG. 5 is a schematic block diagram of the temperature information set in embodiment 1 of the present application;

fig. 6 is a schematic diagram of an improved flow chart of a network card power supply control method provided in embodiment 2 of the present application;

FIG. 7 is a schematic block diagram of application of operation interval information in embodiment 2 of the present application;

fig. 8 is a schematic block diagram of a network card power supply control device provided in embodiment 3 of the present application;

fig. 9 is a schematic structural diagram of a terminal provided in embodiment 4 of the present application;

fig. 10 is a schematic structural diagram of a computer device provided in embodiment 5 of the present application.

Detailed Description

In the description of the present disclosure, it is to be noted that embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure. It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in and/or in parallel. Furthermore, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect. It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units. It should be noted that references to "a" and "an" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be construed as "one or more" unless the context clearly indicates otherwise.

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

Example 1

A schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application is applied is described herein, where the system architecture of the present application may include a plurality of terminal devices, a network, and a server. The terminal device includes one or more of a smart phone and a computer, which may be a tablet, a portable computer, a desktop computer, or the like, and the network is used to provide a medium for a communication link between the terminal device and the server. The network may include various connection types, such as wired communication links, wireless communication links, and the like. It should be understood that the number of terminal devices, networks and servers may have any number of terminal devices, networks and servers, as desired for implementation. A user may interact with a server through a network using a terminal device to receive or send messages, etc.

It should be noted that, the network card power supply control method provided in this embodiment of the present application is executed by a server, where the server is based on an SDN architecture and is a server cluster formed by a first server and a plurality of second servers, the first server is connected to the plurality of second servers, the first server includes a first power controller and a plurality of first intelligent network cards, each second server includes a second power controller and a plurality of second intelligent network cards, the first server is configured to receive all data transmission requests initiated by a terminal device and perform flow scheduling on all data transmission requests, and then distribute the data transmission requests to a designated second server, the second server is configured to respond to the allocated data transmission requests, the first power controller is configured to control a power supply mode of each first intelligent network card, and the second power controller is configured to control a power supply mode of a second intelligent network card belonging to the same second server. In practical application, the first power supply controller or the second power supply controller changes the power supply by changing the current, so as to control different power supply modes.

As shown in fig. 1, embodiment 1 provides a network card power supply control method, which is applied to a server, and includes:

s100, performing first response processing when a timer signal based on a first preset period is triggered;

as shown in fig. 2, the first response process is performed when the timer signal based on the first preset period triggers, and specifically includes:

s101, acquiring a task request queue to be scheduled; in this embodiment, a task request queue to be scheduled is configured to implement load balancing of a scheduling request in a first server, where the task request queue to be scheduled includes a plurality of data transmission requests, each data transmission request corresponds to one piece of traffic information, each piece of traffic information includes a user ID, a request time, and an application identifier, and after each data transmission request is scheduled, a queue element is deleted to complete updating, where the data transmission request is initiated by a terminal device and is differentiated according to the user ID, different terminal devices run different applications and differentiate different applications according to the application identifier, and then traffic information is generated when the terminal device initiates the request. The first preset period is set to, for example, a time period value of 1 second or more, for example, 1 second, 5 seconds, 10 seconds, etc., and specific values are not limited herein.

S102, generating a time zone request subset based on a task request queue to be scheduled so as to update a time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval;

s103, performing first power supply mode adjustment processing based on the current scheduling request quantity; in this embodiment, only one first intelligent network card with a function type of the management network card identifier is in the first server, and the first intelligent network card with the function type of the management network card identifier is used for performing proxy scheduling on a task request queue to be scheduled for distribution to the second server.

In this embodiment, as shown in fig. 3, the first power supply mode adjustment processing is performed based on the current scheduling request number, and specifically includes:

aiming at a first intelligent network card with the function type of management network card identification:

s1031, if the number of the current scheduling requests is smaller than the number of the first preset scheduling requests, setting the first intelligent network card with the function type of the management network card identifier as an energy-saving power supply mode;

S1032, if the number of the current scheduling requests exceeds the first preset scheduling request number and is smaller than the second preset scheduling request number, setting the first intelligent network card with the function type of the management network card identifier as a first load power supply mode;

s1033, if the number of the current scheduling requests exceeds the number of second preset scheduling requests, setting the first intelligent network card with the function type of the management network card identifier as a second load power supply mode;

in this embodiment, the power supply power in the energy saving power supply mode is smaller than the power supply power in the first load power supply mode, and the power supply power in the first load power supply mode is smaller than the power supply power in the second load power supply mode. It should be noted that the actual power may be different according to the actual network card product, and specific values are not limited herein.

In actual application, the current scheduling request quantity is judged by combining the first preset scheduling request quantity and the second preset scheduling request quantity, so that the power supply mode of the first intelligent network card with the management network card identifier in the first server is timely adjusted, and the self-adaptive adjustment level of the power supply mode of the network card of the first server when scheduling is requested is improved.

S200, carrying out second power supply mode adjustment processing by combining the temperature information set and the current scheduling request quantity;

in this embodiment, as shown in fig. 4, the second power supply mode adjustment processing is performed by combining the temperature information set and the current scheduling request number, and specifically includes:

s201, if the number of the current scheduling requests exceeds the number of the second preset scheduling requests, calculating the period continuous times, wherein the period continuous times are used for representing the statistics times that the number of the current scheduling requests exceeds the number of the second preset scheduling requests, and taking a first preset period as one statistics time; in this embodiment, the current number of scheduling requests is the number of traffic information in the subset of time zone requests closest to the current time.

S202, if the period continuous times exceed the preset times, acquiring a temperature information set.

Referring to fig. 5, the temperature information set is composed of a plurality of temperature information, each temperature information corresponds to a time value, the temperature information includes all network card IDs and network card detection temperatures corresponding to each network card ID, each first intelligent network card association or each second intelligent network card association is respectively associated with a unique network card ID, and further, different network cards are positioned according to the network card IDs; in this embodiment, a temperature sensor is set in each of the first intelligent network card and the second intelligent network card to collect temperature data, so that each network card ID has a network card detection temperature. In the present embodiment, the preset number of times may be set to 20 times, 50 times, 100 times, etc., and is not limited herein.

S203, traversing the temperature information set, screening out network card IDs of which the network card detection temperature exceeds a preset temperature threshold value based on the current time value, judging whether the high-temperature duration exceeds a preset high-temperature duration threshold value, and performing dormancy switching on the network card IDs of which the high-temperature duration exceeds the preset high-temperature duration threshold value; in this embodiment, the high temperature duration is obtained by filtering the network card detected temperature exceeding the preset temperature threshold, and calculating the difference between the maximum time value and the minimum time value.

In this embodiment, performing sleep switching on a network card ID with a high-temperature duration exceeding a preset high-temperature duration threshold specifically includes:

if the network card ID corresponds to the attribution identifier of the first server, switching the first intelligent network card with the function type of redundant network card identifier and the first intelligent network card corresponding to the network card ID, and setting a dormancy identifier and a first preset dormancy time; in this embodiment, the function types corresponding to the two network card IDs are mutually exchanged by mutually switching, and the function types are used as redundant network card identifiers for switching, so that the first intelligent network card corresponding to the network card ID of the original function type executes the sleep power supply mode based on the first preset sleep time, thereby avoiding the continuous high temperature from influencing the subsequent working state of the first intelligent network card.

If the network card ID has the attribution identification of the second server, determining a second server serial number where the network card ID is located according to the attribution identification of the second server, mutually switching a second intelligent network card which belongs to the second server serial number and has the function type of redundant network card identification and a second intelligent network card corresponding to the network card ID, and setting a dormancy identification and a second preset dormancy time; in this embodiment, the function types corresponding to the two network card IDs are mutually exchanged by mutually switching, and the function types are used as redundant network card identifiers for switching, so that the second intelligent network card corresponding to the network card ID of the original function type executes the sleep power supply mode based on the second preset sleep time, thereby avoiding the continuous high temperature from influencing the subsequent working state of the second intelligent network card.

In this embodiment, in the first server, the function types of the first intelligent network card include a management network card identifier, a public network card identifier, an intranet card identifier, and a redundant network card identifier. In any one of the second servers, the function types of the second intelligent network card comprise a public network card identifier, an intranet card identifier and a redundant network card identifier. The management network card identifier is an identifier for indicating that the network card is used for scheduling, the public network card identifier is an identifier for indicating that the network card is used for accessing an external network, the intranet card identifier is an identifier for indicating that the network card is used for accessing an internal network, and the redundant network card identifier is an identifier for indicating that the network card is used for backup. By way of example, the following illustrates network cards of the respective functional types: when the function type is the management network card, the function type is generally used for the management and maintenance of the server, can be accessed through an intranet, is generally fixed in IP address, and can be connected to a management interface of the server to perform the operation such as the configuration and monitoring of the server. When the function type is a public network card, the connection is usually used for the office of a company or a mechanism or an external network, can be connected to the public network and provide network services, and needs to be connected with an external Internet server. When the function type is an internal network card, the network environment, such as a cluster or a local area network, which is usually used for connecting with the inside of the server, needs a separate IP address and can only access network resources inside the server. When the function type is the internal network card, the redundant network card plays a role of network card backup in the server, and when other network cards of the server fail or have problems, the redundant network card is used for automatically taking over service and continuously providing network service for clients, so that the stability and reliability of the server are improved, and the influence caused by service interruption is reduced.

In this embodiment, for a first intelligent network card with a sleep identifier, when the sleep identifier is set, a first timer is set according to a first preset sleep time, and when the first preset sleep time is triggered, the sleep identifier and the first preset sleep time are cleared for the first intelligent network card by the first timer, so that the first intelligent network card obtains timely sleep. And setting a second timer according to a second preset dormancy time when the dormancy identification is set for the second intelligent network card with the dormancy identification, and clearing the dormancy identification and the second preset dormancy time through the second timer when the second preset dormancy time is triggered, so that the second intelligent network card is dormant in time.

S204, if the period continuous times do not exceed the preset times, not processing.

In actual application, by judging the duration of high temperature, sleep processing is timely performed on an intelligent network card or a second intelligent network card which is in a high temperature condition Kuang Di, so that the first intelligent network card or the second intelligent network card is in a more reasonable working state, and damage to network card hardware equipment caused by high temperature is reduced.

Example 2

Based on the foregoing embodiment 1, the embodiment 2 implements more accurate power supply control on the second intelligent network card of the second server in time by analyzing the average ending time of the protocol according to the use condition of the user on the terminal device for different applications, so as to improve the stability and reliability of the second server when responding to the data transmission request.

As shown in fig. 6, the method for controlling power supply of a network card provided in this embodiment further includes:

s300, acquiring application operation interval information of all users.

In this embodiment, as shown in fig. 7, application operation interval information of different users is distinguished based on user IDs, the application operation interval information of each user contains multiple sets of application operation interval sub-information, and the multiple sets of application operation interval sub-information are obtained by classifying the application operation interval information based on application identifiers, and each set of application operation interval sub-information includes multiple protocol average end times;

in this embodiment, the average protocol ending time is an average time difference between a protocol initiating time and an application ending time for any type of protocol, and is specifically expressed as:

It should also be noted that the present embodiment is directed toThe relevant data collection of each protocol is processed under the authorized state of the user, the process from running to finishing of each application is regarded as one-time use times, L (i, m) can also keep the fixed times of the latest use and delete the redundant historical use times so as to enable the data to be more suitable for the latest application use habit of the user,the method comprises the steps that an average initiation time value of a protocol of an nth class protocol is triggered when an ith user ID is used for the kth time of an mth class application identifier, and the initiation time value belonging to the same class protocol is obtained by means of average calculation of the ith user ID when the ith user ID is used for the kth time of the mth class application identifier, and different classes of protocols are distinguished by different protocol IDs.

The protocols are distinguished according to different application functions, for example, in chat applications, protocols triggered by controls such as "send", "view details", "comment" all generate different types of protocols, and then a process from opening the application to closing the application once is a process from one start to end, and according to different actual operations of a user, the different types of protocols are triggered, and further the different types of protocols are analyzed to obtain average end time of the protocols.

S400, performing second response processing when the timer signal based on the second preset period is triggered, and further adjusting network card power supply conditions of all second servers in time.

In this embodiment, the second preset period is set according to the average link bandwidth occupancy rate of each second server, that is:

wherein θ is _k Represents the link bandwidth occupancy of the kth second server, P represents the number of second servers,accumulated value representing link bandwidth occupancy of P second servers，/>Represents average of the link bandwidth occupancy rate of each second server, T represents a second preset period, θ _ref For a preset reference period threshold value, T ₁ Assigning a value T to the second preset period when the link bandwidth occupancy rate of each second server is smaller than the preset reference period threshold value ₂ And assigning a value for the second preset period when the link bandwidth occupancy rate of each second server exceeds a preset reference period threshold value.

Exemplary, when θ _ref Set to 50%, if40%, the second preset period T is set as T ₁ If->60%, the second preset period T is set to T ₂ ；θ _ref The specific value of (C) may be set to 20% to 80% according to the actual situation, T ₁ Greater than T ₂ ，T ₁ Or T ₂ For example, 15 minutes, 20 minutes, 30 minutes, 1 hour, etc. may be set, and the present invention is not limited thereto.

In this embodiment, the second response process is performed when triggered by a timer signal based on a second preset period, and specifically includes:

s401, traversing all the second servers, and acquiring a plurality of request processing lists of the second server traversed currently, wherein each request processing list corresponds to one second intelligent network card of the second server and is used for managing data transmission requests of the corresponding second intelligent network card;

s402, extracting the protocol type of the data transmission request with the first requirement identifier from the request processing list to determine the sequence numbers of all the protocols of the data transmission request with the first requirement identifier;

s403, screening maximum values of average end time of the protocol of the data transmission request with the first requirement mark in each request processing list based on the sequence number of the protocol of the data transmission request to obtain a plurality of application end estimated times, wherein each second intelligent network card in the second server traversed at present corresponds to one application end estimated time one by one;

in this embodiment, the average end time of the protocol of the data transmission request with the first requirement identifier in the request processing list is screened to a maximum value, so as to determine the estimated duration time of the second intelligent network card corresponding to the request processing list, that is, the estimated application end time.

S404, if the application ending estimated time is smaller than a first preset waiting threshold, setting a second intelligent network card corresponding to the application ending estimated time as an energy-saving power supply mode, if the application ending estimated time exceeds the first preset waiting threshold and is smaller than a second preset waiting threshold, setting the second intelligent network card corresponding to the application ending estimated time as a third load power supply mode, if the application ending estimated time exceeds the second preset waiting threshold, setting the second intelligent network card corresponding to the application ending estimated time as a fourth load power supply mode, wherein the first preset waiting threshold is smaller than the second preset waiting threshold, the power supply in the energy-saving power supply mode is smaller than the power supply in the third load power supply mode, and the power supply in the third load power supply mode is smaller than the power supply in the fourth load power supply mode;

in this embodiment, the application ending estimated time is used to reflect the ending time of the data transmission request in the request processing list, and the larger the application ending estimated time is, the longer the subsequent data transmission request in the request processing list is, and by timely adjusting the power supply mode of the second intelligent network card, sufficient power supply support is provided by using the third load power supply mode or the fourth load power supply mode, and the power supply consumption of the network card is reduced by using the energy-saving power supply mode, so that the self-adaptive adjustment level of power supply control of the second server is improved.

When the intelligent network card is in actual application, the range of the power supply power of the first intelligent network card is 20W-80W, when the intelligent network card is in an energy-saving power supply mode, the power supply power is adjusted to be 20W through the first power supply controller, when the intelligent network card is in a first load power supply mode, the power supply power is adjusted to be 50W through the first power supply controller, and when the intelligent network card is in a second load power supply mode, the power supply power is adjusted to be 80W through the first power supply controller; the range of the power supply power of the second intelligent network card is 10W-80W, when the intelligent network card is in the dormant power supply mode, the power supply power is adjusted to 10W through a second power supply controller belonging to the same second server, when the intelligent network card is in the energy-saving power supply mode, the power supply power is adjusted to 20W through the second power supply controller belonging to the same second server, when the intelligent network card is in the third load power supply mode, the power supply power is adjusted to 50W through the second power supply controller belonging to the same second server, and when the intelligent network card is in the fourth load power supply mode, the power supply power is adjusted to 80W through the second power supply controller belonging to the same second server.

In the embodiment, the network card serves as a physical interface between the computer and the network cable, and converts digital signals in the computer into optical signals; when running different programs, for example, network text chatting or web page browsing, only intermittent data downloading and uploading are needed, online video watching needs to download and decode audio in real time, the network data time intervals needed by the programs are different, and through judging the average ending time of the protocol, the network card power supply can be reduced when the network flow data requirement is less, so that the effect of saving electricity is achieved.

In an application scenario of this embodiment, in a process of performing traffic scheduling by the first server, the method specifically includes:

classifying all data transmission requests based on application identifiers, setting a first requirement identifier for the data transmission requests if the application identifiers of the data transmission requests correspond to first-class traffic requirements, otherwise, setting a second requirement identifier for the data transmission requests; in this embodiment, the first type of traffic demand is a demand state that requires a large, continuous data transfer process in the network link. The first type of traffic demands are exemplified by data migration of virtual machines, large-scale algorithmic graphics processing based on Mapreduce, etc., word processing, and parallel computing tasks of large-scale datasets. The second type of traffic demand is a demand state that requires a small, short-term data transfer process in the network link. The second category of traffic demands is, for example, email, web browsing, social chat App.

Classifying all the second servers based on the link bandwidth occupancy rate, setting a first scheduling identifier if the link bandwidth occupancy rate of the second servers is lower than the preset bandwidth occupancy rate, otherwise setting a second scheduling identifier; in this embodiment, the preset bandwidth occupancy rate may be set to a value of 40% or more, for example, 40%,45%,50%,60%, etc., which is not limited herein.

Summarizing all second servers with the first scheduling identification to form a first circular distribution queue, and summarizing all second servers with the second scheduling identification to form a second circular distribution queue; in actual application, the first circular distribution queue and the second circular distribution queue are annular queues, and distribution is carried out by sequentially designating a second server in scheduling, and the first circular distribution queue or the second circular distribution queue determines the next second server to be distributed after successful scheduling each time;

scheduling all data transmission requests in sequence, judging whether the data transmission requests have first demand identifiers or not if the data transmission requests are first demand conditions, distributing the data transmission requests through a first circular distribution queue if the data transmission requests have the first demand identifiers, distributing the data transmission requests through a second circular distribution queue if the data transmission requests lack the first demand identifiers, and distributing the data transmission requests according to a second server distributed when the data transmission requests are first scheduling conditions if the data transmission requests are not first demand conditions;

in this embodiment, the first request condition is a protocol request initiated by the user when starting to use the application, and if the user uses up the application, the first request condition is reset.

When the first circular distribution queue is not an empty queue, judging whether the number of times of completing the scheduling of the first circular distribution queue reaches a first preset scheduling reset number of times, if so, re-acquiring the link bandwidth occupancy rate of each second server, re-classifying all the second servers based on the link bandwidth occupancy rate, re-forming the first circular distribution queue and the second circular distribution queue, and if not, not processing, wherein the first scheduling reset number of times is the number of members of the first circular distribution queue, and updating after re-forming the first circular distribution queue each time; the first circular distribution queue includes 15 second servers with the first scheduling identifier, that is, the number of members of the first circular distribution queue is 15, and when the number of times of scheduling completed by the first circular distribution queue reaches 15, the judgment condition is met, and the second servers are classified according to the link bandwidth occupancy rate of each second server, so that the specific value of the first scheduling reset number is redetermined.

When the second circulation distribution queue is not empty, judging whether the number of times of completing the second circulation distribution queue reaches a second preset number of times of dispatching, if so, re-acquiring the link bandwidth occupancy rate of each second server, re-classifying all the second servers based on the link bandwidth occupancy rate, and re-forming a first circulation distribution queue and a second circulation distribution queue, if not, not processing, wherein the second dispatching number of times of dispatching is the number of members of the second circulation distribution queue, and updating after re-forming the second circulation distribution queue each time. The second circular distribution queue includes 20 second servers with second scheduling identifications, that is, the number of members of the second circular distribution queue is 20, and when the number of times of scheduling completed by the second circular distribution queue reaches 20, the second circular distribution queue meets the judgment condition, and is classified again according to the link bandwidth occupancy rate of each second server, so as to redetermine the specific value of the second scheduling reset number.

In actual application, the number of times of scheduling completed by the first circular distribution queue or the second circular distribution queue is judged, so that each data transmission request can be distributed to a proper second server for response, and the effect of load balancing when a plurality of second servers respond to the requests is improved.

Example 3

As shown in fig. 8, embodiment 3 provides a network card power supply control device, which corresponds to the network card power supply control method, on the basis of embodiment 2 above, and the device is applied to the server. The device comprises: the system comprises a first response processing module, a power supply mode association adjustment processing module, an application operation interval information acquisition module and a second response processing module, wherein the first response processing module comprises a request queue acquisition sub-module, a time zone request subset generation sub-module and a power supply mode adjustment processing sub-module, and the power supply mode association adjustment processing module comprises a period continuous time calculation sub-module, a temperature information set acquisition sub-module and a dormancy switching sub-module;

the request queue acquisition sub-module is used for acquiring a task request queue to be scheduled;

The time zone request subset generating sub-module is used for generating a time zone request subset based on a task request queue to be scheduled by a user to update a time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval;

the power supply mode adjustment processing sub-module is used for performing first power supply mode adjustment processing based on the current scheduling request quantity;

the period continuous time calculation sub-module is used for calculating period continuous times when the current scheduling request number exceeds the second preset scheduling request number, wherein the period continuous times are used for representing the statistics times that the current scheduling request number exceeds the second preset scheduling request number, and the current scheduling request number is the number of flow information in the time zone request subset closest to the current time by taking a first preset period as one statistics time.

The temperature information collection acquisition sub-module is used for judging that if the period continuous times exceeds the preset times, acquiring a temperature information collection, if the period continuous times do not exceed the preset times, not processing, wherein the temperature information collection is composed of a plurality of temperature information, each temperature information corresponds to a time value, the temperature information comprises all network card IDs and network card detection temperatures corresponding to each network card ID, each first intelligent network card association or each second intelligent network card is respectively associated with a unique network card ID, and further, different network cards are positioned according to the network card IDs;

the application operation interval information acquisition module is used for acquiring application operation interval information of all users, wherein the application operation interval information of each user contains a plurality of groups of application operation interval sub-information, the plurality of groups of application operation interval sub-information is obtained by classifying the application operation interval information based on application identifiers, and each group of application operation interval sub-information comprises a plurality of protocol average end times;

Example 4

As shown in fig. 9, the present embodiment provides a terminal including: at least one memory and at least one processor; the at least one memory is configured to store program codes, and the at least one processor is configured to invoke the program codes stored in the at least one memory to execute any one of the network card power supply control methods in the above embodiments.

Example 5

The present embodiment provides a computer device, which may be a server, and an internal structure thereof may be as shown in fig. 10. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is a physical layer for storing various databases. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a network card power control method.

It will be appreciated by those skilled in the art that the structure shown in fig. 10 is merely a block diagram of some of the structures associated with the present application and is not limiting of the computer device to which the present application may be applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

Example 6

The present embodiment provides a storage medium for storing program codes for executing the above-described network card power supply control method.

The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).

The embodiments described above are preferred embodiments of the present application, but the embodiments of the present application are not limited to the embodiments described above, and any other changes, modifications, substitutions, combinations, and simplifications that do not depart from the spirit and principles of the present application should be made by equivalent substitution methods, and are included in the scope of the present application.

Claims

1. The network card power supply control method is characterized by being applied to a server, wherein the server is a server cluster formed by a first server and a plurality of second servers, the first server is respectively connected with the plurality of second servers, the first server comprises a first power supply controller and a plurality of first intelligent network cards, each second server comprises a second power supply controller and a plurality of second intelligent network cards, the first server is used for receiving all data transmission requests initiated by a terminal device and carrying out flow scheduling on all data transmission requests and then distributing the data transmission requests to the designated second server, the second server is used for responding to the distributed data transmission requests, the first power supply controller is used for controlling the power supply mode of each first intelligent network card, and the second power supply controller is used for controlling the power supply mode of the second intelligent network card belonging to the same second server;

The method comprises the following steps:

performing a first response process when triggered based on a timer signal of a first preset period;

performing a second power supply mode adjustment process by combining the temperature information set and the current scheduling request quantity;

the step of performing the first response processing when triggered by the timer signal based on the first preset period specifically includes:

acquiring a task request queue to be scheduled, wherein the task request queue to be scheduled comprises a plurality of data transmission requests, each data transmission request corresponds to one piece of flow information, each piece of flow information comprises a user ID, a request time and an application identifier, and deleting queue elements after each data transmission request is scheduled to finish updating;

generating a time zone request subset based on a task request queue to be scheduled so as to update the time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval;

performing first power supply mode adjustment processing based on the current scheduling request quantity;

The step of performing the first power supply mode adjustment processing based on the current scheduling request number specifically includes:

the power supply power in the energy-saving power supply mode is smaller than the power supply power in the first load power supply mode, and the power supply power in the first load power supply mode is smaller than the power supply power in the second load power supply mode;

the step of performing the second power supply mode adjustment processing by combining the temperature information set and the current scheduling request number specifically includes:

if the number of the current scheduling requests exceeds the second preset scheduling request number, calculating the period continuous times, wherein the period continuous times are used for representing the statistics times that the number of the current scheduling requests exceeds the second preset scheduling request number, and taking a first preset period as one time for statistics, and the number of the current scheduling requests is the number of flow information in a time zone request subset closest to the current time;

2. The method of claim 1, wherein the step of performing sleep switching on the network card ID with the high temperature duration exceeding the preset high temperature duration threshold specifically comprises:

if the network card ID corresponds to the attribution identifier of the first server, switching the first intelligent network card with the function type of redundant network card identifier and the first intelligent network card corresponding to the network card ID, and setting a dormancy identifier and a first preset dormancy time;

3. The method of claim 2, wherein for a first intelligent network card having a sleep identifier, setting a first timer according to a first preset sleep time when the sleep identifier is set, and clearing the sleep identifier and the first preset sleep time for the first intelligent network card by the first timer when the first preset sleep time is triggered, thereby enabling the first intelligent network card to get timely sleep;

4. The method as recited in claim 1, further comprising:

；

wherein the method comprises the steps ofIndicate->Link bandwidth occupancy of the second server, < >>Representing the number of second servers, +. >Representation->Accumulated value of link bandwidth occupancy of second server, +.>Representing average link bandwidth occupancy per second server,/for each second server>Representing a second preset period,/->For a preset reference period threshold, +.>Assigning a value for the second preset period when the average link bandwidth occupancy of each second server is less than the preset reference period threshold value,/for the second preset period>And assigning a value for the second preset period when the link bandwidth occupancy rate of each second server exceeds a preset reference period threshold value.

5. The method of claim 4, wherein the protocol average end time is an average time difference between a protocol initiation time and an application end time for any type of protocol, specifically expressed as:

；

wherein the method comprises the steps ofDenoted as +.>The individual user ID is at->Class application identifier->Triggering the +.>Protocol average initiation time value of class protocol, +.>Denoted as +.>The individual user ID is at->Class application identifier->Time value of application end at secondary use, < >>Is->The individual user ID is at->Total number of uses of class application identifier, +.>Indicate->The individual user ID is at->Class application identifier triggering +.>Protocol average end time at class protocol.

6. The method according to claim 5, wherein the second response processing is performed when triggered by a timer signal based on a second preset period, specifically comprising:

7. The method of claim 6, wherein during the traffic scheduling performed by the first server, specifically comprising:

classifying all data transmission requests based on application identifiers, setting a first requirement identifier for the data transmission requests if the application identifiers of the data transmission requests correspond to first-class traffic requirements, otherwise, setting a second requirement identifier for the data transmission requests;

8. A network card power supply control device, comprising: the power supply mode association adjustment processing module comprises a cycle continuous time calculation sub-module, a temperature information set acquisition sub-module and a dormancy switching sub-module;

the time zone request subset generating sub-module is used for generating a time zone request subset based on a task request queue to be scheduled so as to update a time zone request set, wherein the time zone request set comprises a plurality of time zone request subsets, each time zone request subset corresponds to a time interval, each time interval corresponds to a time interval formed by a first preset period, and each time zone request subset contains all flow information in the corresponding time interval;

the period continuous time calculation sub-module is used for calculating period continuous times when judging that the current scheduling request number exceeds the second preset scheduling request number, wherein the period continuous times are used for representing the statistics times that the current scheduling request number exceeds the second preset scheduling request number and taking a first preset period as one time for statistics, and the current scheduling request number is the number of flow information in a time zone request subset closest to the current time;

the second response processing module is used for performing second response processing when triggered based on a timer signal of a second preset period, so as to timely adjust network card power supply conditions of all second servers;

the first power supply mode adjustment process based on the current scheduling request number specifically includes: