WO2023125380A1

WO2023125380A1 - Data management method and corresponding apparatus

Info

Publication number: WO2023125380A1
Application number: PCT/CN2022/141828
Authority: WO
Inventors: 井天成
Original assignee: 华为技术有限公司
Priority date: 2021-12-31
Filing date: 2022-12-26
Publication date: 2023-07-06
Also published as: CN116418893A

Abstract

Disclosed in the present application is a data management method, which is applied to a network device, the network device being a switching device or a network interface apparatus, such as a network adapter, an accelerator card, etc., installed in a switching device or a data node. The method comprises: a network device analyzes a service message by means of a message analysis rule in an acquired data management rule, so as to determine target data associated with the service message and further acquire a statistical result of a service request related to the target data; and when the statistical result of the service request related to the target data meets a response condition, the network device performs, according to a response policy, corresponding operations on the service message received by the network device and/or the target data (hot data). In the present solution, the network device manages the hot data, does not create a resource conflict with a service related to the target data, and moreover, reduces the probability of heavy pressure on a service device caused by overloaded access to the hot data.

Description

A data management method and corresponding device

This application claims the priority of the Chinese patent application with the application number 202111679430.7 and the title of the invention "a data management method and corresponding device" filed with the China Patent Office on December 31, 2021, the entire contents of which are incorporated herein by reference Applying.

technical field

The present application relates to the fields of communication technology and data storage technology, and in particular to a data management method and a corresponding device.

Background technique

With the massive increase in data volume, for centralized storage, distributed storage, distributed database, centralized database and other service systems, some data identification and fast access have always been key indicators of business performance, such as: Data identification and fast access directly affect the user's business experience. In the process of file storage and database storage, hierarchical storage can also be performed according to the frequency of data access, so as to reduce storage costs on the basis of guaranteed performance.

In a hierarchical storage scenario, some commonly used data will be stored in the cache, which facilitates quick access to these commonly used data. However, if a large number of users suddenly need to access the same data, the traffic will be too concentrated, and many physical resources such as network bandwidth, physical storage space, and database connections cannot be supported. At present, it is impossible to quickly identify these data with too concentrated access, which affects business operations.

Contents of the invention

The present application provides a data management method for quickly identifying target data in a network and performing corresponding operations. The embodiment of the present application also provides a corresponding device, a computer-readable storage medium, a computer program product, and the like.

The first aspect of the present application provides a data management method, the method is applied to network equipment, the network equipment is a switching equipment or a network interface device installed in the switching equipment or data node, the method includes: the network equipment obtains data management rules, Data management rules include message analysis rules, response conditions, and response strategies; network devices analyze service messages from clients according to message analysis rules to determine the target data associated with service messages, and the target data is application layer data; network The device counts the service requests related to the target data, and the service requests are obtained from the service messages; when the statistical results of the service requests related to the target data meet the response conditions, the network device will respond to the service messages received by the network device according to the response strategy. And/or act accordingly on the target data.

In this application, the data system where the network device is located may be a distributed system or a centralized system, the distributed system may include a distributed database service system, the centralized system may include a centralized database service system, and other related underlying storage system.

In this application, the switching device may be a switch, a router or a gateway device, etc. used to implement data forwarding in the network, such as: storage area network (storage area network, SAN) switch, Ethernet switch, fiber optic switch or other physical switches The device can also be a virtual switch. A data node can be a physical machine, a virtual machine (virtual machine, VM) or other forms of resources for data storage. The network interface device may be a network card, an accelerator card, a data processing unit (data processing unit, DPU) or a network interface, such as a switch interface. The client may be a terminal device or an application installed on the terminal device.

In this application, the data management rules may be pre-configured on the network equipment, or may be received from a rule management device or a service server related to the target data.

In this application, a service packet refers to a packet of data related to a service. Service messages can be transmitted in the network using Transmission Control Protocol/Internet Protocol (Transmission Control Protocol/Internet Protocol, TCP/IP), or remote direct memory access (remote direct memory access, RDMA) protocol, or other communication protocols.

In this application, the target data associated with the service message can be directly parsed from the service message, or can be searched according to the identifier (such as: index information or address information of the target data) parsed from the service message arrived.

In this application, the service request related to the target data refers to a request for reading or writing the target data, etc., and the type of the service request can be parsed from the service message.

In this application, the message parsing rules refer to the rules for parsing business messages, which can be: the content that needs to be parsed in the first step, the content that needs to be parsed in the second step, etc. The rules of message parsing can specify the required Parsed fields.

In this application, the response condition refers to the condition that needs to be met to trigger the response policy to take effect.

In this application, the response strategy refers to a quality of service (Qos) adjustment strategy or a resource adjustment strategy made for service packets or target data.

It should be noted that, the service quality in the embodiment of the present application refers to the network service quality.

It can be seen from the above first aspect that the network equipment can identify and process the target data, and will not occupy the resources of the service server used to provide the target data-related services, and will not cause resource conflicts with the target data-related services, and, through the network equipment It can be identified in advance when the business request related to the target data is still being transmitted in the network, and then take corresponding actions, which can reduce the probability that the target data will be overloaded and accessed, resulting in excessive pressure on the business server.

In a possible implementation manner of the first aspect, the response condition includes a statistical rule and a response relationship; wherein, the statistical rule includes a statistical time length, and the response relationship includes a relationship between the number of business requests counted according to the statistical time length and the target threshold. The statistical results of the business requests related to the target data include: the number of the business requests related to the target data within the statistical time period.

In another possible implementation of the first aspect, the statistical result meeting the response condition includes: according to the indication of the response relationship, the number of business requests related to the target data within the statistical time exceeds the target threshold, if the number exceeds the target threshold, and the target data is not existing hot data, it indicates that the target data is newly generated hot data; or, the statistical result meets the response condition including: according to the indication of the response relationship, the number of business requests related to the target data is low within the statistical period At the target threshold, if the quantity is lower than the target threshold and the target data is existing hot data, it indicates that the hot attribute of the target data disappears.

In this application, the statistical duration refers to the length of time used to count service requests, such as: 1 minute, 2 minutes, 5 minutes or other values. The response relationship refers to the relationship between the number of business requests counted according to the statistical duration and the target threshold (such as: exceeding the target threshold or lower than the target threshold. It should be noted that there is only one relationship in a response relationship, or it is greater than the target threshold threshold, or less than the target threshold). When taking the response relationship greater than the target threshold as an example, for example: the statistical duration is 1 minute, and the target threshold is 10,000 times. If the statistical result is that the number of business requests related to the target data in one minute is 12,000 times, If the target data is not existing hot data, it means that the target data is newly generated hot data, and the statistical result satisfies the response condition. When the response relationship is less than the target threshold as an example, if the statistical result is that the number of business requests related to the target data in one minute is 7,000 times, and when the target data is existing hot data, it means that the statistics The result also satisfies the response condition.

In this application, the process of hot data generation or hot data disappearance can be counted. If the statistics show that the service request for target data that is not hot data exceeds the target threshold within one minute, it can be understood as the process of hot data generation. If the statistics are counted If the one-minute business request of target data that is already hot data is lower than the target threshold, it can be understood as the process of hot data disappearing. Of course, the target thresholds in the response relationship used for hotspot data generation and hotspot data disappearance may be different.

In this application, hot data refers to data whose active volume exceeds a certain threshold for a period of time, such as: data that has been visited, clicked, browsed, followed, and downloaded many times. Hot data can be hot products, Various forms of data such as hot news, hot applications, hot programs, and node music.

In this application, statistical hotspot data can be used, such as: least recently used (least recently used, LRU), recently least frequently used (least frequently used, LFU), and first in first out (first in first out, FIFO), etc. algorithm.

In this possible implementation mode, the network device can quickly identify and process the hot data, and will not occupy the resources of the service server used to provide the target data-related services, and will not cause resource conflicts with the target data-related services. Moreover, through The network device can identify the hotspot data in advance and take corresponding actions when the hotspot data is still being transmitted in the network, which can reduce the probability that the hotspot data will be overloaded and accessed, resulting in excessive pressure on the business server.

In a possible implementation of the first aspect, the response conditions include multiple sets of statistical rules and response relationships, the statistical durations in different statistical rules are different, and the target thresholds in the response relationships corresponding to different statistical rules are different; different Different adjustment strategies in response strategies corresponding to statistical rules.

In this application, the statistical duration included in the statistical rule and the target threshold in the response relationship usually correspond. For example: the statistics time is 1 minute, the target threshold is 10,000 times, the statistics time is 2 minutes, the target threshold is 20,000 times, the statistics time is 5 minutes, and the target threshold is 50,000 times. Of course, this is just an example. In fact, the relationship between the statistical duration in different statistical rules and the target threshold in the corresponding response relationship does not necessarily change proportionally. Although it is only introduced here that the statistical rules include the statistical duration, in fact, the statistical rules can also include other parameters, such as: the identification of the statistical object. The identification of the statistical object can be, for example, the external business address of the target data corresponding business. For example: it can be the address of the device where the target data is located, or the virtual interface address of the VM, or other information that can represent the business address, such as the key value (key) of redis, or the url address of the video , the index of the file, etc. When different response conditions are used for data on different devices, the statistical object identifier can be set in the statistical rule, and if the same set of response conditions is used for all data, the statistical object identifier may not be set.

In this possible implementation, by configuring multiple sets of statistical rules and response relationships in the response conditions, different adjustment strategies can be effectively enabled. The data access volume is more matched, which can avoid the problem of business overload or downtime due to insufficient resources, and can also improve resource utilization.

In a possible implementation of the first aspect, the response strategy includes: a service quality adjustment strategy for service packets with target priority, and an adjustment strategy for resource expansion or resource shrinkage for target data; wherein, target priority level is included in the response policy.

In a possible implementation of the first aspect, the above steps: the network device performs corresponding operations on the service packets received by the network device and/or on the target data according to the response policy, including: when the target data is a newly generated hotspot data, the network device lowers the service quality of target priority service packets according to the service quality adjustment policy, and/or performs resource expansion operations on the target data according to the resource expansion adjustment policy; when the hotspot attribute of the target data disappears The network device adjusts the service quality of the service packets with the priority of the target service according to the service quality adjustment strategy, and/or performs resource shrinkage operations on the target data according to the resource shrinkage adjustment strategy.

In this application, the network equipment can receive service messages with different priorities. When hot data is generated, in order to avoid the impact of hot data on high priority services, the high priority service messages can be processed first. In order to ensure The processing of high-priority service packets can reduce the service quality of target priority service packets according to the target priority specified in the response policy (for example: reduce the bandwidth, delay or packet loss rate, etc.). And/or, do resource expansion for hot data, that is, increase hot data resources, so that the impact of hot data on high-priority services can be alleviated.

In this application, resources for expansion or contraction of hotspot data include network resources, computing resources, and storage resources.

In this application, when the hotspot data disappears, the disappearance of the hotspot data means that the resources occupied by the hotspot data will be vacated. At this time, the service quality of the service message of the target priority can be improved, which is beneficial to the priority of these targets. better processing of business packets. And/or, shrink resources for hot data, that is, reduce hot data resources, which can improve resource utilization.

In a possible implementation of the first aspect, the above steps: the network device parses the service message from the client according to the message parsing rules to determine the target data associated with the service message, including: the network device analyzes the service message according to the message The parsing rule parses the five-tuple of the service message to determine the service type of the service message; the network device parses the indication field according to the message parsing rule to determine the target data of the service type or the index information of the target data, and the index information is used to determine target data.

In this application, the five-tuple includes the destination address, and the service type of the service message can be determined through the destination address, such as: the destination address can be the IP address of the game server, the IP address of the video server, or the IP address of the music server , and the IP address of the e-commerce server. The indication field in the service message can be located in the message header or in the message load. The indication field can be the index information of a certain target data. The index information can be the specific address of the target data or the address of the target data. name or logo etc.

In a possible implementation of the first aspect, when the response strategy indicates that the control node used to manage the data node executes the adjustment strategy for resource expansion or resource reduction; Perform resource expansion operations, including: the network device sends an expansion notice to the control node, and the expansion notice is used to instruct the control node to perform resource expansion for the target data; or, the above steps: make resource shrinkage for the target data according to the resource shrinkage adjustment strategy The operation includes: the network device sends a capacity reduction notification to the control node, and the capacity reduction notification is used to instruct the control node to perform resource reduction for the target data.

In this possible implementation, the control node refers to a device that globally manages the resources in the system where the network device is located. The control node can expand the resource capacity of the target data according to the received The capacity notification performs resource shrinkage for the target data, because the control node coordinates the resources of the entire system, and can make better resource expansion or resource shrinkage for the target data.

In a possible implementation manner of the first aspect, the data node further includes a first control device; when the response strategy indicates that the first control device executes an adjustment strategy for resource expansion or resource shrinkage; the above steps: Adjusting the policy to perform resource expansion operations on the target data includes: the network device sends a capacity expansion notice to the first control device, and the capacity expansion notice is used to instruct the first control device to perform resource expansion for the target data; or, the above steps: The operation of adjusting the strategy to shrink resources for the target data includes: the network device sends a shrink notification to the first control device, and the shrink notification is used to instruct the first control device to shrink resources for the target data.

In this possible implementation manner, the process of resource expansion or resource reduction for target data is completed by the first control device in the data node.

In a possible implementation of the first aspect, the network device is a network interface device installed on the switching device; the switching device further includes a second control device, and when the response policy indicates that the resource expansion or resource reduction is performed by the control device When adjusting the strategy; the above steps: perform resource expansion operations on the target data according to the resource expansion adjustment strategy, including: the network interface device sends a capacity expansion notification to the second control device, and the capacity expansion notification is used to instruct the second control device to perform resource expansion for the target data. capacity expansion; or, the above step: performing a resource reduction operation on the target data according to the resource reduction adjustment strategy, including: the network interface device sends a capacity reduction notification to the second control device, and the capacity reduction notification is used to instruct the second control device to be Target data for resource shrinkage.

In this possible implementation manner, the second control device on the switching device completes the resource expansion or resource reduction process of the target data, which eliminates the need to configure a control node in the system and simplifies the network structure.

In a possible implementation manner of the first aspect, the above step: the network device obtains the data management rules includes: the network device obtains the data management rules from a service server or a rule management device, and the service server is a server related to the target data.

A second aspect of the present application provides a network device, where the network device has a function of implementing the first aspect or the method of any possible implementation manner of the first aspect. This function may be implemented by hardware, or may be implemented by executing corresponding software on the hardware. The hardware or software includes one or more modules corresponding to the above-mentioned functions, for example: an acquisition unit, a first processing unit, a second processing unit, and an operation unit, wherein the first processing unit and the second processing unit can also be processed through a processing unit to achieve.

A third aspect of the present application provides a network device, the network device includes at least one processor, a memory, an input/output (input/output, I/O) interface, and a computer executable stored in the memory and operable on the processor Instructions, when the instruction executed by the network device is executed by the processor, the processor executes the method in the first aspect or any one possible implementation manner of the first aspect.

The fourth aspect of the present application provides a computer-readable storage medium that stores one or more computer-executable instructions. When the computer-executable instructions are executed by a processor, one or more processors execute any of the above-mentioned first aspect or first aspect. A method of one possible implementation.

The fifth aspect of the present application provides a computer program product that stores one or more computer-executable instructions. When the computer-executable instructions are executed by one or more processors, one or more processors execute the above-mentioned first aspect or first A method for any one of the possible implementations of the aspect.

The sixth aspect of the present application provides a network card, the network card includes an input interface, an output interface and at least one processor, the input interface is used to receive service messages, and at least one processor is used to support the network card to realize the first aspect or the first aspect For the functions involved in any possible implementation, the output interface is used to send the notification of capacity expansion or capacity reduction. In a possible design, the network card may also include memory, memory, and necessary program instructions and data for network equipment.

Description of drawings

FIG. 1 is a schematic structural diagram of a distributed system provided by an embodiment of the present application;

Fig. 2 is a schematic structural diagram of the centralized system provided by the embodiment of the present application;

Fig. 3 is a schematic diagram of an embodiment of a data management method provided by an embodiment of the present application;

Fig. 4 is a schematic diagram of another embodiment of the data management method provided by the embodiment of the present application;

FIG. 5 is a schematic structural diagram of a service message provided by an embodiment of the present application;

Fig. 6 is a schematic diagram of another embodiment of the data management method provided by the embodiment of the present application;

Fig. 7 is a schematic diagram of another embodiment of the data management method provided by the embodiment of the present application;

Fig. 8 is a schematic diagram of another embodiment of the data management method provided by the embodiment of the present application;

Fig. 9 is a schematic diagram of another embodiment of the data management method provided by the embodiment of the present application;

FIG. 10 is another schematic structural diagram of a service message provided by an embodiment of the present application;

FIG. 11 is a schematic structural diagram of a network device provided by an embodiment of the present application;

FIG. 12 is another schematic structural diagram of a network device provided by an embodiment of the present application.

Detailed ways

Embodiments of the present application are described below in conjunction with the accompanying drawings. Apparently, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Those of ordinary skill in the art know that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

The terms "first", "second" and the like in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

An embodiment of the present application provides a data management method for quickly identifying target data in a network and performing corresponding operations. The embodiment of the present application also provides a corresponding device, a computer-readable storage medium, a computer program product, and the like. Each will be described in detail below.

The method provided by the embodiment of this application can be implemented on a network device. The data system where the network device is located can be a distributed system or a centralized system. The distributed system can include a distributed database service system, and the centralized system can include a centralized database service system, and other systems involving underlying storage.

FIG. 1 is a schematic diagram of a scenario of a distributed system provided by an embodiment of the present application.

As shown in FIG. 1 , in the distributed system scenario provided by the embodiment of the present application, a client can access the distributed system through a network, and the distributed system includes a service entry node, a switching device, a control node, and multiple data nodes. Wherein, the service entry node is in communication connection with the switching device, and the switching device is in communication connection with the control node and each data node. It should be noted that the distributed system may not include a control node, and the function of the control node is implemented by a switching device.

The service entry node is equivalent to the interface connecting the client network and the distributed system. A service entry node can be connected to multiple switching devices (only one switching device is shown in Figure 1), and the service entry node is used to receive business from the client. message, and deliver the service message to the switching device, and the switching device distributes it to different data nodes. The service entry node can realize the load balancing of each switching device. Data nodes can communicate with control nodes through switching devices, and control nodes can also send messages to each data node through switching devices.

A service packet refers to a packet of data related to a service. Service packets can be transmitted in the network using Transmission Control Protocol/Internet Protocol (Transmission Control Protocol/Internet Protocol, TCP/IP), or remote direct memory access (remote direct memory access, RDMA) protocol, or other communication protocols.

The switching device can be a switch, a router or a gateway device, etc. used to implement data forwarding in the network, such as: a storage area network (storage area network, SAN) switch, an Ethernet switch, a fiber switch or other physical switching devices, or It is a virtual device, such as a virtual switch.

Data nodes can be physical machines, virtual machines (virtual machine, VM) or other forms of resources for data storage, such as: data storage (data storage) or non-volatile memory, the data storage includes but not Limited to solid state drives (SSD), disk arrays, or other types of non-transitory computer-readable media, it can also include traditional dynamic random access memory (dynamic random access memory, DRAM) and new persistent memory ( persistent memory, PMem). Non-volatile memory includes semiconductor memory devices such as erasable programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM) and flash memory (flash memory) ; Disks, such as internal hard disk or removable disk, magneto optical disk, and CD ROM and DVD-ROM.

The control node can be a physical machine or a virtual machine, and is used to manage global resources in a distributed system, including computing resources, storage resources, and network resources.

The client may be a terminal device or an application installed on the terminal device. Terminal equipment (also called user equipment (UE)) is a device with wireless transceiver function, which can be deployed on land, including indoor or outdoor, handheld or vehicle-mounted; it can also be deployed on water (such as ships etc.); can also be deployed in the air (such as aircraft, balloons and satellites, etc.). The terminal can be mobile phone (mobile phone), tablet computer (pad), computer with wireless transceiver function, virtual reality (virtual reality, VR) terminal, augmented reality (augmented reality, AR) terminal, industrial control (industrial control) Wireless terminals, wireless terminals in self driving, automatic driving devices, wireless terminals in remote medical, wireless terminals in smart grid, wireless in transportation safety Terminals, wireless terminals in smart cities, wireless terminals in smart homes, etc.

FIG. 2 is a schematic diagram of a scenario of a centralized system provided by an embodiment of the present application.

As shown in Figure 2, in the scenario of the centralized system provided by the embodiment of the present application, the client can access the centralized system through the network, and the centralized system includes multiple centralized services, wherein each centralized service includes a The service entry node and a data node manage the computing resources, storage resources and network resources in the centralized service.

The service entry node is equivalent to the interface connecting the network of the client and the centralized system. The service entry node is used to receive the business message from the client and pass the service message to the corresponding data node.

The structures of the distributed system and the centralized system have been introduced above. The data management method provided by the embodiment of the present application can be applied to the distributed system shown in FIG. 1 and the centralized system shown in FIG. 2 , for example. The following describes the data management method provided by the embodiment of the present application with reference to the accompanying drawings. The data management method is applied to a network device, and the network device may be a switching device in the above-mentioned distributed system or a centralized system, or a network interface device installed in a switching device or a data node.

Wherein, the network interface device may be a network card, an accelerator card, a data processing unit (data processing unit, DPU) or a network interface, such as a switch interface.

As shown in Figure 3, an embodiment of the data management method provided by the embodiment of the present application includes:

101. The network device acquires data management rules, and the data management rules include message parsing rules, response conditions, and response strategies.

The data management rules may be pre-configured on the network equipment, or may be received from a rule management device or a service server related to the target data, where the service server is a server related to the target data.

The message parsing rules refer to the rules for parsing business messages, which can be: the content that needs to be parsed in the first step, the content that needs to be parsed in the second step, etc. The fields that need to be parsed can be specified in the message parsing rules.

Response conditions refer to the conditions that need to be met to trigger the response policy to take effect.

The response strategy refers to a quality of service (Qos) adjustment strategy or a resource adjustment strategy made for service packets or target data.

102. The network device parses the service message from the client according to the message parsing rules to determine the target data associated with the service message, and the target data is application layer data.

The target data associated with the service message can be directly parsed from the service message, or can be found according to the identifier (such as: index information or address information of the target data) parsed from the service message.

This step 102 may include: the network device parses the five-tuple of the service message according to the message parsing rule to determine the service type of the service message; the network device parses the indication field according to the message parsing rule to determine the target data or Index information of the target data, the index information is used to determine the target data.

The five-tuple includes the destination address, through which the business type of the service message can be determined, for example: the destination address can be the IP address of the game server, the IP address of the video server, the IP address of the music server, and the e-commerce The IP address of the server, etc. The indication field in the service message can be located in the message header or in the message load. The indication field can be the index information of a certain target data. The index information can be the specific address of the target data or the address of the target data. name or logo etc.

103. The network device counts service requests related to the target data, and the service requests are obtained from service packets.

The service request related to the target data refers to a request for reading or writing the target data, etc., and the type of the service request can be parsed from the service message.

104. When the statistical result of the service request related to the target data satisfies the response condition, the network device performs corresponding operations on the service message received by the network device and/or on the target data according to the response strategy.

In the embodiment of the present application, the network device can identify and process the target data without occupying the resources of the service server used to provide the target data-related services, and will not cause resource conflicts with the target data-related services. Moreover, the network device can When the business request related to the target data is still transmitted in the network, it can be identified in advance and then take corresponding actions, which can reduce the probability of causing the target data to be overloaded and accessed, resulting in excessive pressure on the business server.

Optionally, in this embodiment of the application, the response condition includes a statistical rule and a response relationship; wherein, the statistical rule includes a statistical time length, and the response relationship includes a relationship between the number of service requests counted according to the statistical time length and the target threshold.

The statistical results of the business requests related to the target data include: the number of the business requests related to the target data within the statistical time period.

Statistical results satisfying the response conditions include: according to the indication of the response relationship, the number of business requests related to the target data exceeds the target threshold within the statistical period. If the number exceeds the target threshold and the target data is not existing hot data, indicate the target data It is newly generated hotspot data; or, the statistical results meet the response conditions include: according to the indication of the response relationship, the number of business requests related to the target data within the statistical period is lower than the target threshold, if the number is lower than the target threshold, and the target data If it is existing hotspot data, the hotspot attribute indicating the target data disappears.

Wherein, the statistical duration refers to the length of time used for statistical service requests, such as: 1 minute, 2 minutes, 5 minutes or other values.

The response relationship refers to the relationship between the number of business requests counted according to the statistical duration and the target threshold (such as: exceeding the target threshold or lower than the target threshold. It should be noted that there is only one relationship in a response relationship, or it is greater than the target threshold threshold, or less than the target threshold). When the response relationship is greater than the target threshold as an example, for example: the statistical duration is 1 minute, and the target threshold is 10,000 times. If the statistical result is that the number of business requests related to the target data in one minute is 12,000 times , and the target data is not existing hot data, it means that the target data is newly generated hot data, and the statistical result satisfies the response condition. When the response relationship is less than the target threshold as an example, if the statistical result is that the number of business requests related to the target data in one minute is 7,000 times, and when the target data is existing hot data, it means the The statistical results also meet the response conditions.

A record table of hot data can be maintained on the network device, and data information that has become hot data is marked in the record table.

In the embodiment of the present application, in combination with the above example, the relationship between statistical rules, response relationships, statistical results, and whether the statistical results meet the response conditions can be understood by referring to Table 1 below.

Table 1:

It can be seen from the above Table 1 that although the statistical time length is the same, the response relationship can be different, and whether the response condition is met or not needs to be considered whether the target data is an existing hot data. The statistical rules can be marked as the rules used for the existing hotspot data, or the rules used for the non-existing hotspot data. And Table 1 is just an example. In fact, in the scene where the hotspot data disappears, the statistical duration can be set to be longer than the scene where the hotspot data is generated. For example, the statistical duration of the scene where the hotspot data is generated is set to 1 minute. , the statistical duration of the scene where the hotspot data disappears is set to 3 minutes or 5 minutes, of course, the specific value is not limited in this embodiment of the application. In addition, the detection cycle can also be set for the scene where the hot data disappears, and the resource shrinking process or other service quality adjustments are made to the data when the business requests for the hot data are detected to be declining for several consecutive cycles.

In the embodiment of this application, the process of hot data generation or hot data disappearance can be counted. If the statistics show that the service request for target data that is not hot data exceeds the target threshold within one minute, it can be understood as the process of hot data generation. If It can be understood as the process of hot data disappearing if the business requests within one minute of the target data that are already hot data are lower than the target threshold. Of course, the target thresholds in the response relationship used for hotspot data generation and hotspot data disappearance may be different.

In the embodiment of this application, hot data refers to data whose activity exceeds a certain threshold for a period of time, such as: data that has been accessed, clicked, browsed, followed, and downloaded many times. Hot data can be hot Various forms of data such as commodities, hot news, hot applications, hot programs, and node music. Statistical hotspot data can be used, such as: least recently used (least recently used, LRU), least recently used (least frequently used, LFU), and first in first out (first in first out, FIFO) and other algorithms.

Optionally, in this embodiment of the application, the response conditions include multiple sets of statistical rules and response relationships. The statistical durations in different statistical rules are different, and the target thresholds in the response relationships corresponding to different statistical rules are different; different statistical rules correspond to Different adjustment strategies in the response strategy.

In this embodiment of the present application, the statistical duration included in the statistical rule and the target threshold in the response relationship usually correspond. For example: the statistics time is 1 minute, the target threshold is 10,000 times, the statistics time is 2 minutes, the target threshold is 20,000 times, the statistics time is 5 minutes, and the target threshold is 50,000 times. Of course, this is just an example. In fact, the relationship between the statistical duration in different statistical rules and the target threshold in the corresponding response relationship does not necessarily change proportionally. Although it is only introduced here that the statistical rules include the statistical duration, in fact, the statistical rules can also include other parameters, such as: the identification of the statistical object. The identification of the statistical object can be, for example, the external business address of the target data corresponding business. For example: it can be the address of the device where the target data is located, or the virtual interface address of the VM, or other information that can represent the business address, such as the key value (key) of redis, or the url address of the video , the index of the file, etc. When different response conditions are used for data on different devices, the statistical object identifier can be set in the statistical rule, and if the same set of response conditions is used for all data, the statistical object identifier may not be set.

In the embodiment of this application, by configuring multiple sets of statistical rules and response relationships in the response conditions, different adjustment strategies can be effectively enabled. In this way, the adjusted resources for processing target data and the current client access to target data The volume is more matched, and the utilization rate of resources is improved.

Optionally, in this embodiment of the application, the response strategy includes: a service quality adjustment strategy for service packets with a target priority, and an adjustment strategy for resource expansion or resource shrinkage for target data; wherein, the target priority is included in response strategy.

In combination with the response strategy, the above step 104 may include two cases, one is the processing process when the hotspot data is generated, and the other is the processing process when the hotspot data disappears, which will be introduced respectively below.

1. Generate hotspot data.

When the target data is newly generated hotspot data, the network device lowers the service quality of the target priority service packets according to the service quality adjustment strategy, and/or performs resource expansion operations on the target data according to the resource expansion adjustment strategy;

In the embodiment of this application, the network device can receive service messages of different priorities. When hot data is generated, in order to avoid the impact of hot data on high priority services, the high priority service messages can be processed first. To ensure the processing of high-priority service packets, the service quality of target priority service packets can be lowered according to the target priority specified in the response policy (for example: reduce the bandwidth, time delay or packet loss rate, etc.). And/or, do resource expansion for hot data, that is, increase hot data resources, so that the impact of hot data on high-priority services can be alleviated.

In the embodiment of the present application, resources for expansion or contraction of hotspot data include network resources, computing resources, and storage resources.

2. Hotspot data disappears.

When the hotspot attribute of the target data disappears, the network device adjusts the quality of service of the service packets with the priority of the target business according to the service quality adjustment strategy, and/or performs resource shrinkage operations on the target data according to the resource shrinkage adjustment strategy .

In the embodiment of the present application, when the hot data disappears, the disappearance of the hot data means that the resources occupied by the hot data before will be vacated. At this time, the service quality of the service message with the target priority can be improved, which is beneficial to these targets. Priority service packets are better processed. And/or, shrink resources for hot data, that is, reduce hot data resources, which can improve resource utilization.

As mentioned above, when the response condition includes multiple sets of statistical rules and response relationships, different sets of statistical rules may correspond to adjustment strategies in different response strategies, and the corresponding relationship may be understood with reference to the example in Table 2 below.

Table 2:

It can be seen from the examples in Table 2 that when the number of service requests counted by different statistical time periods exceeds the corresponding target threshold, different response strategies are executed. Of course, the content in Table 1 here is just an example, and only provides an idea that different statistical rules correspond to different response strategies. Other implementation methods that can be within this idea can be understood as the solutions of this application.

It should be noted that in the above Table 2, when performing resource expansion or resource shrinkage, different statistical rules can also be used to perform resource expansion or resource shrinkage according to different specifications. For example, the statistics time is 2 minutes. Add 1 data node to process hot data, and the statistical time is 5 minutes. When performing resource expansion, you can add 2 data nodes to process hot data. Of course, this is just an idea, specific resource expansion or resource shrinking The specification of the content is not limited in this application.

In addition, in the embodiment of the present application, the management process of the target data is slightly different in different scenarios, and the management process of the target data in each scenario will be introduced respectively below.

In the scenario of the distributed system shown in Figure 1 above, the network device can be a switching device or a network interface device installed on the switching device or data node. In Figure 4, the network device is installed on the network card of the data node as an example. Introduction, as shown in Figure 4, another embodiment of the data management method provided by the embodiment of the present application includes:

201. The network card receives a service packet from the client.

The network card has pre-acquired the data management rules described above. As shown in FIG. 4 , the data management rules have been pre-acquired from the rule management device.

202. The network card parses the service packets from the client according to the packet parsing rules, so as to determine the target data associated with the service packets.

The message parsing process can be understood by referring to Figure 5. As shown in Figure 5, the service message includes the message header (Eth Header, IP Header, and TCP Header) and the message payload (RESP payload). The text parsing rule parses the packet header, determines the destination address of the data requested by the service packet according to the IP address, and then parses the packet payload to determine that the packet payload includes the read request for the target data and the target data According to the index information, the target data can be obtained from the server at the IP address.

203. The network card counts the service requests related to the target data according to the statistical duration in the statistical rule.

When the target data is not recorded in the hot data record table, it means that the target data is not hot data, then you can choose the response condition of not hot data, or the statistical rule of not hot data and the corresponding response relationship. Then count the number of business requests related to the target data during the statistical period. When the number is greater than the target threshold, it means that the target data is newly generated hot data.

When the target data has been recorded in the hot data record table, it means that the target data is hot data, then you can choose the response condition of hot data, or the statistical rules of hot data and the corresponding response relationship. Then count the number of service requests related to the target data during the statistical period. When the data is less than the target threshold, it means that the hotspot attribute of the target data has disappeared.

204. When the statistical result of the service request related to the target data satisfies the response condition, the network card lowers the service quality of the service message with the target priority, or increases the service quality of the service message with the target priority.

For the newly generated hot data, the network card can lower the service quality of the target priority business packets. For the target data whose hotspot attribute has disappeared, the network card can increase the service quality of the target priority business packets.

205. The network card sends a capacity expansion notice or a capacity reduction notice to the control node through the switching device.

The capacity expansion notification is used to instruct the control node to perform resource expansion for the target data.

The capacity reduction notification is used to instruct the control node to perform resource reduction for the target data.

If it is a scene where hot data is generated, the NIC sends a capacity expansion notice; if it is a scene where hot data disappears, the network card sends a capacity reduction notice.

206. The control node performs resource expansion for the target data according to the capacity expansion notification, or, the control node performs resource reduction for the target data according to the capacity reduction notification.

The resource expansion process may be that the control node sends a notification to a certain data node, indicating that the data node is also a node for processing target data.

The resource shrinking process may be that the control node sends a notification to a certain data node, indicating that the data node is no longer a node for processing target data.

Because the control node coordinates the resources of the entire network, it can make better resource expansion or resource reduction for the target data.

The above describes the process in which the network card on the data node is used as a network device to manage the above-mentioned hot data in conjunction with FIG. 4 . When the network device is a network interface device on a switching device, the above process can be understood with reference to FIG. 6 .

The following describes the hotspot data management process when the network device is a network interface on a switching device, such as a switch interface, in a distributed system scenario with reference to FIG. 6 .

As shown in Figure 6, another embodiment of the data management method provided by the embodiment of the present application includes:

301. A network interface on the switching device receives a service packet from a client.

302. The network interface parses the service packets from the client according to the packet parsing rules, so as to determine the target data associated with the service packets.

303. The network interface counts service requests related to the target data according to the counting duration in the counting rule.

304. When the statistical result of the service request related to the target data satisfies the response condition, the network interface lowers the service quality of the service message with the target priority, or increases the service quality of the service message with the target priority.

305. The network interface sends a capacity expansion notice or a capacity reduction notice to the control node.

306. The control node performs resource expansion for the target data according to the capacity expansion notification, or, the control node performs resource reduction for the target data according to the capacity reduction notification.

Relevant content in the above steps performed by the network interface in FIG. 6 can be understood by referring to related explanations in steps 201 to 206 in FIG. 4 , and will not be repeated here.

The above describes the process of managing the above-mentioned hotspot data by using the network interface on the switching device as a network device in conjunction with Fig. 6. When the network device is a switching device, the process is the same as the process in Fig. 6 above, and you can refer to the process in Fig. 6 for understanding .

The solution introduced in FIG. 6 above is a process of implementing resource expansion or resource reduction through the control node when the network device is the network interface of the switching device. Optionally, a control device may also be installed on the switching device, and the resource expansion or resource reduction process may be implemented through the control device. The process of this scenario may be understood by referring to FIG. 7 .

As shown in FIG. 7 , steps 401 to 404 in this scenario can be understood by referring to steps 301 to 304 in FIG. 6 , and steps 405 and 406 are different from FIG. 6 .

405. The network interface sends a capacity expansion notice or a capacity reduction notice to the control device.

406. The control device performs resource expansion for the target data according to the capacity expansion notification, or, the control device performs resource reduction for the target data according to the capacity reduction notification.

The control device in the distributed scenario may also be referred to as a second control device.

The specific implementation process of step 406 can be understood by referring to the corresponding description in the preceding step 206 .

The above describes the management process of hotspot data when the network device in the distributed system is the network card on the data node, the network interface on the switching device, and the switching device. data management process.

As shown in FIG. 8 , in the scenario of a centralized system, the network device is a network interface device installed on a data node, and the data node also includes a control device.

As shown in Figure 8, another embodiment of the data management method provided by the embodiment of the present application includes:

Steps 501 to 504 in this embodiment can be understood by referring to steps 201 to 204 in FIG. 4 , except for steps 505 and 506 .

505. The network card sends a capacity expansion notice or a capacity reduction notice to the control device.

506. The control device performs resource expansion for the target data according to the capacity expansion notification, or, the control device performs resource reduction for the target data according to the capacity reduction notification.

The control device in the centralized scenario may also be referred to as a first control device.

In this centralized system, the amount of resources of each data node is usually very large, and the idle resources or business resources on the data node can be adjusted to serve the hot data, or the idle resources generated due to the disappearance of hot data can be recovered.

The solution provided by the embodiment of the present application can also be applied to the Redis business scenario. The application process in the Redis business scenario can be understood by referring to FIG. 9. As shown in FIG. Proxy nodes (one of which can be selected as the control node) and Redis data nodes. Among them, the service entry node is mainly responsible for business distribution and processing, the business proxy node is mainly responsible for business link management and data caching, and the Redis data node mainly provides data read and write services.

In this Redis cluster, taking the video scene as an example, the data of video A, video B and video C are respectively distributed in three Redis data nodes, and each Redis data node is performed by a network card to perform data identification and statistics The process, the specific process includes:

601. The network card in the Redis data node of video A receives the service packet.

602. The network card first parses the service packets according to the packet parsing rules.

The format of the service message can be understood by referring to Figure 10. The network card first analyzes the message header, determines the IP address of the server where the video A is located, and then analyzes the message load to determine the index information of the video A. , according to the index information, the specific content in the corresponding video A can be found.

603. The network card counts the service requests related to video A according to the statistical rules in the response condition.

604. When the video A is newly generated hotspot data, the network card lowers the service quality of the service packets with the service priority specified in the response policy.

605. The network card sends a capacity expansion notice to the service proxy node serving as the control node.

606. The control node sends a notification to another idle Redis data node, instructing the Redis data node to provide video A with data processing services.

It should be noted that the above-mentioned Figures 4 to 9 all involve the process of expanding or shrinking the target data, and the device performing the expansion or shrinking can be the control node, or the first control device or the second control device , the process of which device will perform capacity expansion or contraction can be indicated in the response policy, and the network device can send a capacity expansion notice or a capacity contraction notice to the device for performing capacity expansion or contraction indicated in the response policy, and then the device to complete the expansion or reduction process.

The above provides centralized and distributed storage or database services, which changes the traditional method of hotspot statistics in the main control business process, and sends the statistics and identification of hotspots to network devices for operation. In addition to other basic services, if certain characteristics of data identification and statistics are required, the technology of this application can be used to replace them. On the premise of reducing business computing resource conflicts, network resources can be fully utilized for statistical identification of data. In the process of AI model training, the model data can also be identified and counted through this solution.

The data management method provided by the embodiment of the present application is described above in combination with different scenarios, and the network device provided by the embodiment of the present application is described below in conjunction with the accompanying drawings.

As shown in FIG. 11, the network device 70 provided by the embodiment of the present application may be a switching device or a network interface device installed in a switching device or a data node. An embodiment of the network device 70 includes:

The acquiring unit 701 is configured to acquire data management rules, where the data management rules include message parsing rules, response conditions, and response policies. The obtaining unit 701 may execute step 101 in the foregoing method embodiments.

The first processing unit 702 is configured to parse the service message from the client according to the message parsing rules acquired by the acquiring unit 701 to determine target data associated with the service message, where the target data is application layer data. The first processing unit 702 may execute

steps

102, 202, 302, 402, and 502 in the foregoing method embodiments.

The second processing unit 703 is configured to count service requests related to the target data analyzed by the first processing unit 702, and the service requests are obtained from service packets. The second processing unit 703 may execute steps 103, 203, 303, 403, and 503 in the foregoing method embodiments.

The operation unit 704 is configured to respond to the service message received by the network device and/or to the target data according to the response strategy when the statistical result of the service request related to the target data collected by the second processing unit 703 satisfies the response condition operate. The operation unit 704 may execute steps 104, 105, 204, 205, 304, 305, 404, 405, 504, and 505 in the foregoing method embodiments.

Optionally, the response condition includes a statistical rule and a response relationship; wherein, the statistical rule includes a statistical time length, and the response relationship includes a relationship between the number of business requests counted according to the statistical time length and the target threshold.

Optionally, the statistical result meeting the response condition includes: according to the indication of the response relationship, the number of business requests related to the target data exceeds the target threshold within the statistical period, if the number exceeds the target threshold, and the target data is not existing hot data, It indicates that the target data is newly generated hot data; or, the statistical result meets the response condition including: according to the indication of the response relationship, the number of business requests related to the target data within the statistical time is lower than the target threshold, if the number is lower than the target threshold , and the target data is existing hotspot data, it indicates that the hotspot attribute of the target data disappears.

Optionally, the response conditions include multiple sets of statistical rules and response relationships, different statistical rules have different statistical durations, and different statistical rules correspond to different target thresholds in the response relationships.

Different statistical rules correspond to different adjustment strategies in the response strategy.

Optionally, the response strategy includes: a service quality adjustment strategy for service packets with a target priority, and an adjustment strategy for resource expansion or resource shrinkage for target data; wherein the target priority is included in the response strategy.

Optionally, the operation unit 704 is used to:

When the target data is newly generated hot data, lower the service quality of the target priority service packets according to the service quality adjustment strategy, and/or perform resource expansion operations on the target data according to the resource expansion adjustment strategy.

When the hotspot attribute of the target data disappears, according to the service quality adjustment policy, the service quality of the service message with the target business priority is increased, and/or, according to the resource shrinkage adjustment policy, the resource shrinking operation is performed on the target data.

Optionally, the first processing unit 702 is configured to: parse the five-tuple of the service message according to the message parsing rule to determine the service type of the service message; parse the indication field according to the message parsing rule to determine the target of the service type Index information of the data or target data, the index information is used to determine the target data.

Optionally, the operation unit 704 is configured to send a capacity expansion notice to the control node when the response policy indicates that the control node for managing the data node executes the resource expansion or resource contraction adjustment policy, and the capacity expansion notice is used to instruct the control node to be The resource expansion is performed on the target data; or, a capacity reduction notification is sent to the control node, and the capacity reduction notification is used to instruct the control node to perform resource reduction for the target data.

Optionally, the operation unit 704 is used for the data node to further include a first control device; when the response strategy indicates that the first control device executes an adjustment strategy for resource expansion or resource shrinkage; send a capacity expansion notification to the first control device, and the capacity expansion The notification is used to instruct the first control device to perform resource expansion for the target data; or, send a capacity reduction notification to the first control device, and the capacity reduction notification is used to instruct the first control device to perform resource reduction for the target data.

Optionally, the operation unit 704 is used for the network device to be a network interface device installed on the switching device; the switching device further includes a second control device, and when the response policy indicates that the resource expansion or resource shrinkage adjustment is performed by the second control device In the strategy, send a capacity expansion notice to the second control device, and the capacity expansion notice is used to instruct the second control device to expand the resource capacity for the target data; or, send a capacity reduction notice to the second control device, and the capacity reduction notice is used to instruct the second control device Resource shrinking for target data.

Optionally, the obtaining unit 701 is configured to obtain data management rules from a service server or a rule management device, where the service server is a server related to target data.

Above, the network device provided in the embodiment of the present application can be understood by referring to the corresponding content in the foregoing method embodiment, and details are not repeated here.

FIG. 12 is a schematic diagram of a possible logical structure of the network device 80 involved in the above-mentioned embodiments provided by the embodiments of the present application. The network device 80 includes: a processor 801 , a communication interface 802 , a memory 803 and a bus 804 . The processor 801 , the communication interface 802 and the memory 803 are connected to each other through a bus 804 .

The communication interface 802 is used to support the network device 80 to communicate, and the communication interface 802 may be a network card or a network interface. Such as executing steps 101, 201, 301, 401 and 501 in Fig. 3 to Fig. 10 .

In the embodiment of the present application, the processor 801 is used to control and manage the actions of the network device 80, for example, the processor 801 is used to execute steps 102 to 106, 202 to 206, 302 to 306, 402 to 406, 502 to 506 in FIGS. 3 to 10 and/or other processes for the techniques described herein.

The memory 803 is used for storing program codes and data of the network device 80 .

Wherein, the processor 801 may be a central processing unit, a general purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute the various illustrative logical blocks, modules and circuits described in connection with the present disclosure. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, and the like. The bus 804 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA) bus, etc. The bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in FIG. 12 , but it does not mean that there is only one bus or one type of bus.

In another embodiment of the present application, a computer-readable storage medium is also provided, and computer-executable instructions are stored in the computer-readable storage medium. When at least one processor of the device executes the computer-executable instructions, the device executes the above diagram. 3 to the method described in some embodiments in FIG. 10 .

In another embodiment of the present application, a computer program product is also provided, the computer program product includes computer-executable instructions, and the computer-executable instructions are stored in a computer-readable storage medium; Reading the storage medium reads the computer-executable instructions, and at least one processor executes the computer-executable instructions so that the device executes the methods described in some embodiments in FIG. 3 to FIG. 10 .

In another embodiment of the present application, a network card is also provided, the network card includes an input interface, an output interface and at least one processor, the input interface is used to receive service packets, and at least one processor is used to support the network device to realize the above-mentioned In the methods described in some embodiments in Figures 3 to 7, the output interface is used to send a notification of capacity expansion or capacity reduction when the statistics show that the service packet is hot data. In a possible design, the network card may also include memory, memory, and necessary program instructions and data for network devices.

Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the embodiments of the present application.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in the embodiments of the present application, it should be understood that the disclosed systems, devices and methods may be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated. to another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

A unit described as a separate component may or may not be physically separated, and a component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.

If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods in the various embodiments of the embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

Claims

A data management method, characterized in that the method is applied to a network device, the network device is a switching device or a network interface device installed in the switching device or a data node, and the method includes:

The network device acquires data management rules, and the data management rules include message parsing rules, response conditions, and response strategies;

The network device parses the service message from the client according to the message parsing rule, so as to determine the target data associated with the service message, and the target data is application layer data;

The network device counts service requests related to the target data, and the service requests are obtained from the service packets;

When the statistical result of the service request related to the target data satisfies the response condition, the network device performs corresponding operations on the service message received by the network device and/or on the target data according to the response policy .
The method according to claim 1, wherein the response condition includes a statistical rule and a response relationship; wherein the statistical rule includes a statistical duration, and the response relationship includes the number of service requests counted according to the statistical duration The relationship with the target threshold that should be met;

The statistical result of the service requests related to the target data includes: the number of service requests related to the target data within the statistical period.
The method according to claim 2, characterized in that,

The statistical result meeting the response condition includes: according to the indication of the response relationship, the number of service requests related to the target data within the statistical time span exceeds the target threshold, if the number exceeds the target threshold, and the target data is not existing hotspot data, indicating that the target data is newly generated hotspot data; or,

The statistical result meeting the response condition includes: according to the indication of the response relationship, the number of service requests related to the target data within the statistical time period is lower than the target threshold, if the number is lower than the specified When the target threshold is set and the target data is existing hotspot data, it indicates that the hotspot attribute of the target data disappears.
The method according to claim 2 or 3, wherein the response conditions include multiple sets of statistical rules and response relationships, the statistical durations in different statistical rules are different, and the target thresholds in the response relationships corresponding to different statistical rules are different. same;

Different statistical rules correspond to different adjustment strategies in the response strategy.
The method according to any one of claims 2-4, wherein the response strategy includes: a service quality adjustment strategy for service packets with a target priority, and resource expansion or resource reduction for the target data A content adjustment strategy; wherein, the target priority is included in the response strategy.
The method according to claim 5, wherein the network device performs corresponding operations on the service message received by the network device and/or on the target data according to the response policy, including:

When the target data is newly generated hotspot data, the network device lowers the service quality of service packets with the target priority according to the service quality adjustment policy, and/or, according to the resource expansion adjustment The strategy performs resource expansion operations on the target data;

When the hot spot attribute of the target data disappears, the network device adjusts the service quality of service packets with the target service priority according to the service quality adjustment policy, and/or adjusts according to the resource shrinkage The strategy performs resource shrinking operations on the target data.
The method according to any one of claims 1-6, wherein the network device parses the service message from the client according to the message analysis rule, so as to determine the target data associated with the service message ,include:

The network device parses the quintuple of the service message according to the message parsing rule to determine the service type of the service message;

The network device parses the indication field according to the packet parsing rule to determine the target data of the service type or index information of the target data, where the index information is used to determine the target data.
The method according to claim 6, wherein when the response strategy indicates that the control node for managing the data node executes the resource expansion or resource shrinkage adjustment strategy;

The resource expansion operation for the target data according to the resource expansion adjustment strategy includes:

The network device sends a capacity expansion notice to the control node, where the capacity expansion notice is used to instruct the control node to perform resource expansion for the target data;

or,

The resource shrinking operation on the target data according to the resource shrinking adjustment strategy includes:

The network device sends a capacity reduction notification to the control node, where the capacity reduction notification is used to instruct the control node to perform resource reduction for the target data.
The method according to claim 6, wherein the data node further comprises a first control device; when the response strategy indicates that the resource expansion or resource shrinkage adjustment strategy is executed by the first control device ;

The resource expansion operation for the target data according to the resource expansion adjustment strategy includes:

The network device sends a capacity expansion notice to the first control device, where the capacity expansion notice is used to instruct the first control device to perform resource expansion for the target data;

or,

The resource shrinking operation on the target data according to the resource shrinking adjustment strategy includes:

The network device sends a capacity reduction notification to the first control device, where the capacity reduction notification is used to instruct the first control device to perform resource reduction for the target data.
The method according to claim 6, wherein the network device is a network interface device installed on the switching device; the switching device further includes a second control device, when the response policy indicates that the When the control device executes the resource expansion or resource reduction adjustment strategy;

The resource expansion operation for the target data according to the resource expansion adjustment strategy includes:

The network interface device sends a capacity expansion notification to the second control device, and the capacity expansion notification is used to instruct the second control device to perform resource expansion for the target data;

or,

The resource shrinking operation on the target data according to the resource shrinking adjustment strategy includes:

The network interface device sends a capacity reduction notification to the second control device, where the capacity reduction notification is used to instruct the second control device to perform resource reduction for the target data.
The method according to any one of claims 1-10, wherein the acquisition of data management rules by the network device includes:

The network device obtains the data management rules from a service server or a rule management device, and the service server is a server related to the target data.
A network device, characterized in that the network device is a switching device or a network interface device installed in the switching device or a data node, and the network device includes:

An acquisition unit, configured to acquire data management rules, where the data management rules include message parsing rules, response conditions, and response strategies;

The first processing unit is configured to analyze the service message from the client according to the message analysis rule acquired by the acquisition unit, so as to determine the target data associated with the service message, and the target data is application layer data;

The second processing unit is configured to count service requests related to the target data parsed by the first processing unit, where the service requests are obtained from the service packets;

An operation unit, configured to, when the statistical result of the service request related to the target data collected by the second processing unit satisfies the response condition, perform the service message received by the network device and/or according to the response policy Or perform corresponding operations on the target data.
The network device according to claim 12, wherein the response condition includes a statistical rule and a response relationship; wherein the statistical rule includes a statistical time length, and the response relationship includes statistics of service requests according to the statistical time length The relationship between the quantity and the target threshold should be met;

The statistical result of the service requests related to the target data includes: the number of service requests related to the target data within the statistical period.
The network device according to claim 13, characterized in that,

The statistical result meeting the response condition includes: according to the indication of the response relationship, the number of service requests related to the target data within the statistical time span exceeds the target threshold, if the number exceeds the target threshold, and the target data is not existing hotspot data, it indicates that the target data is newly generated hotspot data; or,

The statistical result meeting the response condition includes: according to the indication of the response relationship, the number of service requests related to the target data within the statistical time period is lower than the target threshold, if the number is lower than the specified When the target threshold is set and the target data is existing hotspot data, it indicates that the hotspot attribute of the target data disappears.
The network device according to claim 13 or 14, wherein the response conditions include multiple sets of statistical rules and response relationships, the statistical time lengths in different statistical rules are different, and the target thresholds in the response relationships corresponding to different statistical rules Are not the same;

Different statistical rules correspond to different adjustment strategies in the response strategy.
The network device according to any one of claims 13-15, wherein the response policy includes: a service quality adjustment policy for service packets with target priority, and resource expansion or resource allocation for the target data. A shrinkage adjustment strategy; wherein, the target priority is included in the response strategy.
The network device according to claim 16, characterized in that,

The operating unit is used for:

When the target data is newly generated hotspot data, lower the service quality of service packets with the target priority according to the service quality adjustment policy, and/or adjust the resource expansion policy according to the resource expansion adjustment policy. The target data performs resource expansion operations;

When the hot spot attribute of the target data disappears, according to the quality of service adjustment strategy, the service quality of the service packet with the priority of the target service is increased, and/or, according to the resource shrinkage adjustment strategy, the The target data performs resource shrinking operations.
The network device according to any one of claims 12-17, characterized in that,

The first processing unit is used for:

Parsing the quintuple of the service message according to the message parsing rules to determine the service type of the service message;

Parsing the indication field according to the packet parsing rules to determine the target data of the service type or index information of the target data, where the index information is used to determine the target data.
The network device according to claim 17, wherein the operation unit is configured to, when the response strategy indicates that the control node for managing the data node executes the resource expansion or resource shrinkage adjustment strategy , sending a capacity expansion notice to the control node, where the capacity expansion notice is used to instruct the control node to perform resource expansion for the target data; or, sending a capacity reduction notice to the control node, where the capacity reduction notice is used to Instructing the control node to perform resource shrinkage for the target data.
The network device according to claim 17, wherein the operation unit is used for the data node and further includes a first control device; when the response strategy indicates that the resource expansion is performed by the first control device or resource shrinkage adjustment strategy, send a capacity expansion notice to the first control device, where the capacity expansion notice is used to instruct the first control device to perform resource expansion for the target data; or, to the first control device The device sends a capacity reduction notification, where the capacity reduction notification is used to instruct the first control device to perform resource reduction for the target data.
The network device according to claim 17, wherein the operation unit is used for the network device to be a network interface device installed on the switching device; the switching device further includes a second control device, when When the response strategy indicates that the resource expansion or resource reduction adjustment strategy is executed by the second control device, a capacity expansion notification is sent to the second control device, and the capacity expansion notification is used to instruct the second control device performing resource expansion for the target data; or sending a capacity reduction notification to the second control device, where the capacity reduction notification is used to instruct the second control device to perform resource reduction for the target data.
The network device according to any one of claims 12-21, characterized in that,

The obtaining unit is configured to obtain the data management rules from a service server or a rule management device, the service server being a server related to the target data.
A network device, characterized by including one or more processors and a storage medium storing computer programs;

The computer program implements the method according to any one of claims 1-11 when executed by the one or more processors.
A computer-readable storage medium on which a computer program is stored, wherein the computer program implements the method according to any one of claims 1-11 when executed by one or more processors.
A computer program product, characterized by comprising a computer program, the computer program is used to implement the method according to any one of claims 1-11 when executed by one or more processors.
A data system, characterized in that the data system includes a switching device, a data node, and a network interface device is installed on the switching device and the data node, and the switching device, the switching device or the data The network interface device installed on the node is used to implement the method according to any one of claims 1-11.