WO2023029254A1

WO2023029254A1 - Data query method and apparatus, and device and medium

Info

Publication number: WO2023029254A1
Application number: PCT/CN2021/134413
Authority: WO
Inventors: 任智新; 张闯; 谢志勇; 王敏; 孙忠祥
Original assignee: 苏州浪潮智能科技有限公司
Priority date: 2021-09-03
Filing date: 2021-11-30
Publication date: 2023-03-09
Also published as: CN113468220A

Abstract

Disclosed in the present application are a data query method and apparatus, and a device and a medium. The method comprises: when a data query operation for a data block is executed, acquiring a data proportion mean value, wherein the data proportion mean value is a mean value of ratios of the amounts of data, which meets a query condition, to the total data amounts of corresponding data blocks in all previous data query operations, the data query operations are all query operations under the same data query task, the data blocks are data blocks of a target database, and the target database is a database corresponding to the data query task; determining whether the data proportion mean value is less than a data proportion threshold value; and if so, moving the current data block to an FPGA acceleration device for processing, otherwise, processing the current data block by using database management software. In this way, the efficiency of data query task processing can be improved, thereby improving the user experience.

Description

A data query method, device, equipment and medium

This application claims the priority of the Chinese patent application with the application number 202111029248.7 and the title of the invention "a data query method, device, equipment and medium" submitted to the China Patent Office on September 3, 2021, the entire contents of which are incorporated by reference in this application.

technical field

The present application relates to the technical field of data processing, and in particular to a data query method, device, equipment and medium.

Background technique

With the emergence of data-intensive applications such as the intelligent society, the cloud era, and the Internet of Things, the amount of data in the database and the number of user visits have risen sharply, and there is an urgent need to increase the processing speed of the database. In order to alleviate the computing pressure brought by the increasing amount of data to the CPU (Central Processing Unit, central processing unit), many studies have focused on hardware acceleration. Hardware acceleration has been proven to be an effective solution, which is to improve database management. The solution for data-intensive application performance in the system. Compared with high-performance GPU (Graphics Processing Unit, graphics processing unit), FPGA is more suitable for database applications with computing bottlenecks, because it has considerable computing power and higher energy efficiency ratio, which is very important for data centers. important. In addition, due to the high parallelism of FPGA itself, for the same database query task, it consumes less computing time and consumes less power than general-purpose processors such as CPU and ARM (Advanced RISC Machine, microprocessor). , so a series of FPAG-based heterogeneous acceleration solutions have appeared in the industry.

In the FPGA-based heterogeneous acceleration solution, the central idea is to offload computationally intensive tasks from the CPU to the FPGA for execution. As for task allocation, most of the tasks related to database query are directly offloaded to FPGA for execution, but for some query tasks, directly offloading tasks to FPGA will consume more time and affect user experience.

Contents of the invention

In view of this, the purpose of the present application is to provide a data query method, device, device, and medium, which can improve the processing efficiency of data query tasks, thereby improving user experience. The specific plan is as follows:

In the first aspect, the application discloses a data query method, including:

When performing a data query operation for a data block, the average value of the data ratio is obtained; wherein, the average value of the data ratio is the average value of the ratio of the amount of data that meets the query conditions in previous data query operations to the total data amount of the corresponding data block, Moreover, the data query operations are all query operations under the same data query task, the data block is a data block of a target database, and the target database is a database corresponding to the data query task;

Judging whether the average value of the data proportion is less than the data proportion threshold;

If yes, move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.

Optionally, also include:

The data query task corresponding to the test data set is processed by FPGA acceleration device and database management software;

Determining the first consuming time for processing by the FPGA acceleration device, and the second consuming time for processing by using the database management software;

determining a target proportion from all data proportions based on the first consumption time and the second consumption time, and using the target proportion as the data proportion threshold;

Wherein, the data ratio is the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block in any data query operation under the data query task, and the first consumption time corresponding to the target ratio is It is equal to the second elapsed time.

Optionally, after moving the current data block to the FPGA acceleration device for processing, or using the database management software to process the current data block, it also includes:

Median filtering is performed on the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block in this and previous data query operations, and the filtering result is used as the data proportion threshold corresponding to the next data query operation.

Based on the type of the target database, the type of data in the target database, the type of query data, the ratio of the amount of data meeting the query conditions in this and previous data query operations to the total data amount of the corresponding data block to determine the next time The data proportion threshold corresponding to the data query operation.

Calculate the average value of the ratio of the amount of data that meets the query conditions to the total data amount of the corresponding data block in this and previous data query operations, and use the average value as the average value of the data proportion corresponding to the next data query operation.

Optionally, also include:

The size of the data block corresponding to each data query operation is determined based on the data storage method and data volume of the target database, and the bandwidth and data processing capability of the FPGA acceleration device.

Optionally, also include:

When performing the data query operation for the data block under the data query task for the first time, the current data block is directly moved to the FPGA acceleration device for processing.

In a second aspect, the present application discloses a data query device, including:

The data ratio mean value acquisition module is used to obtain the data ratio mean value when performing a data query operation for the data block; wherein, the data ratio mean value is the amount of data that meets the query conditions and the corresponding data block in previous data query operations The average value of the ratio of the total data volume, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database is the corresponding data query task database;

A data proportion threshold judging module, configured to determine whether the data proportion average value is less than the data proportion threshold;

The data block processing module is used to move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.

In a third aspect, the present application discloses an electronic device, comprising:

memory for storing computer programs;

A processor, configured to execute the computer program, so as to realize the aforementioned data query method.

In a fourth aspect, the present application discloses a computer-readable storage medium for storing a computer program, and when the computer program is executed by a processor, the aforementioned data query method is implemented.

It can be seen that when the application performs a data query operation for a data block, the average value of the data ratio is obtained; wherein, the average data ratio is the ratio of the amount of data that meets the query conditions in previous data query operations and the total data amount of the corresponding data block. The average value of the ratio, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database is the database corresponding to the data query task, and then it is judged that the Whether the average value of the above data ratio is less than the data ratio threshold, if so, move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block. That is, in this application, when performing a data query operation for a data block under a data query task, the ratio of the amount of data that meets the query conditions in the previous data query operations under the data query task to the total data amount of the corresponding data block Compared with the data ratio threshold, the current data block will be moved to the FPGA acceleration device for processing when it is less than the threshold, otherwise, the current data block will be processed by the database management software, so that the data query is performed according to the amount of data that meets the query conditions Task, to determine whether to move the data block to the FPGA for processing, avoiding the problem of large data delay caused by moving the data block to the FPGA when the amount of data that meets the query conditions is large, and can improve the processing efficiency of the data query task, thereby Improve user experience.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present application, and those skilled in the art can also obtain other drawings according to the provided drawings without creative work.

Fig. 1 is a flow chart of a data query method disclosed in the present application;

Fig. 2 is a kind of concrete realization block diagram based on FPGA database heterogeneous acceleration system that the application provides;

FIG. 3 is a flow chart of a specific data query method disclosed in the present application;

FIG. 4 is a flow chart of a specific data query method disclosed in the present application;

Fig. 5 is a schematic structural diagram of a data query device disclosed in the present application;

FIG. 6 is a structural diagram of an electronic device disclosed in the present application.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

In the FPGA-based heterogeneous acceleration solution, the central idea is to offload computationally intensive tasks from the CPU to the FPGA for execution. As for task allocation, most of the tasks related to database query are directly offloaded to FPGA for execution, but for some query tasks, directly offloading tasks to FPGA will consume more time and affect user experience. For this reason, the present application provides a data query solution, which can improve the processing efficiency of data query tasks, thereby improving user experience.

Referring to Fig. 1, the embodiment of the present application discloses a data query method, including:

Step S11: When performing a data query operation for a data block, obtain the average value of the data proportion; wherein, the average data proportion is the ratio of the amount of data that meets the query conditions in previous data query operations to the total data amount of the corresponding data block , and the data query operations are all query operations under the same data query task, the data block is a data block of a target database, and the target database is a database corresponding to the data query task.

In a specific implementation manner, the size of the data block corresponding to each data query operation is determined based on the data storage method and data volume of the target database, and the bandwidth and data processing capability of the FPGA acceleration device.

It should be pointed out that in the system, the memory of the FPGA acceleration device and the memory of the host are fixed and limited. When processing large data, the data needs to be partitioned. The data transmission bandwidth and memory capacity between devices are limited, and they are also moved to FPGA acceleration devices in batches and multiple times. In the actual application scenario, the size of each data block moved can be set according to the storage method and size of the data table of the database, as well as the bandwidth and data processing capability of the FPGA acceleration device.

Specifically, when the data query request sent by the user is obtained, a corresponding data query task is generated, and the data query task includes multiple data query operations for data blocks.

Step S12: Judging whether the average data proportion is smaller than a data proportion threshold.

Step S13: If yes, move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.

And, after moving the current data block to the FPGA acceleration device for processing, or using the database management software to process the current data block, the embodiment of the present application calculates the amount of data that meets the query conditions and the corresponding data in this and previous data query operations. The average value of the ratio of the total data volume of the block is used as the average value of the data proportion corresponding to the next data query operation.

Moreover, in the embodiment of the present application, when the data query operation for the data block under the data query task is executed for the first time, the current data block is directly moved to the FPGA acceleration device for processing.

For example, as shown in Figure 2, Figure 2 is a block diagram of a specific FPGA-based database heterogeneous acceleration system provided by the embodiment of the present application, a server or service terminal device running the database system and an FPGA accelerator card that realizes intensive task offloading The devices were previously interconnected through the high-speed interface PCIe (Peripheral Component Interconnect Express, a high-speed serial computer expansion bus standard). The FPGA accelerator card includes multiple acceleration function units such as acceleration function unit 0 and acceleration function unit 1 . At present, the general processing flow of FPGA-based heterogeneous database acceleration tasks is as follows: (1) On the server side, when a user initiates a database operation request, the CPU reads the relevant original data block from the disk into the memory, parses and unloads the task , notify the FPGA accelerator card to start the task; (2) send the original data and the calculation task to the DDR ( (Double Data Rate, double rate) memory; (3) After receiving the notification, the FPGA fetches the data and caches the source data from the DDR to the internal RAM according to the assigned calculation tasks, then preprocesses the data, checks Filter the data that meets the query conditions; (4) send the qualified data to the acceleration calculation unit module, perform calculation processing on the data, and write the calculation results back to the internal RAM buffer; (5) the FPGA acceleration unit continues to write the output results Go back to the DDR on the accelerator card and notify the host to read the result, and the host reads the result back to the host-side memory through the PCIe channel in DMA mode. After completing the preceding steps, the CPU passes the calculation result data to the query request operation.

It should be pointed out that in the prior art, the FPGA accelerator card cannot directly read data from the memory in the host system. It needs to move the original data of the database to the DDR in the accelerator card through DMA first, and then transfer the result from The acceleration card is moved back to the host memory. In this process, the delay of data transmission is a non-negligible factor. The more data that meets the query conditions, the larger the amount of data moved, the more data returned, and the greater the PCIe transmission delay. When the amount of data selected is greater than a certain value, the FPGA acceleration method will cause a greater delay instead of achieving the effect of performance acceleration. At this time, you should choose to use the original software query method to perform tasks, thereby improving the overall performance. performance and improve user experience.

Therefore, in the database system based on FPGA heterogeneous acceleration, this application can dynamically select whether to offload the query task to the FPGA according to the size of the query task. When the data that meets the query conditions is less than the set threshold θ, the FPGA acceleration function is started, and the data query operation under the query task is offloaded to the FPGA for execution. Otherwise, the FPGA hardware acceleration function is invalid, and the data query operation under the query task is still in the software. implement. In this application, a task selector is designed and implemented on the host side. When the user sends a query task, the CPU makes a judgment during preprocessing, and selects whether to start the FPGA accelerator according to the threshold and the amount of data to be selected. Referring to FIG. 3 , the embodiment of the present application discloses a specific data query method flow chart. The task selector is implemented on the host side. Before the original data is moved, threshold judgment is performed. If the accelerator startup condition is met, the query task is offloaded to the FPGA for execution. Otherwise, the original process is continued, that is, the query task is executed by the CPU. The specific process includes: when the system executes the task for the first time, that is, when it executes the first data query operation under the current task, the FPGA accelerator is selected by default, and the task selector is opened; when a data block is processed, the calculation meets the query conditions The data volume of the data, and calculate the ratio of the data volume to the total data volume of the data block, denoted as θe, and use this value as the estimated value of whether to use hardware acceleration for the data query operation of the next data block, when θe is less than When θ, the query operation of the next data block still uses hardware acceleration, otherwise, the original software method is used for query. For subsequent data query operations, the value of θe uses the average value of the ratio of the data volume that satisfies the query condition in all data blocks that have been calculated to the total data volume of the corresponding data block. The amount of data is more objective and real.

It can be seen that in the embodiment of the present application, when the data query operation for the data block is executed, the average value of the data proportion is obtained; wherein, the average value of the data proportion is the amount of data that meets the query conditions in previous data query operations and the total data of the corresponding data block The average value of the ratio of the amount, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database is the database corresponding to the data query task, and then It is judged whether the average value of the data ratio is less than the data ratio threshold, if so, the current data block is moved to the FPGA acceleration device for processing, otherwise the current data block is processed by using the database management software. That is, in this application, when performing a data query operation for a data block under a data query task, the ratio of the amount of data that meets the query conditions in the previous data query operations under the data query task to the total data amount of the corresponding data block Compared with the average value of the data ratio threshold, the current data block will be moved to the FPGA acceleration device for processing when it is less than the threshold value, otherwise, the current data block will be processed by the database management software. In this way, the data query is performed according to the amount of data that meets the query conditions Task, to determine whether to move the data block to the FPGA for processing, avoiding the problem of large data delay caused by moving the data block to the FPGA when the amount of data that meets the query conditions is large, and can improve the processing efficiency of the data query task, thereby Improve user experience.

Referring to Figure 4, the embodiment of the present application discloses a specific data query method, including:

Step S21: The data query task corresponding to the test data set is processed by the FPGA acceleration device and the database management software respectively.

Step S22: Determine the first time consumption for processing by the FPGA acceleration device and the second time consumption for processing by the database management software.

Step S23: Determine the target proportion from all data proportions based on the first consumption time and the second consumption time, and use the target proportion as a data proportion threshold.

Among them, the data ratio is the ratio of the amount of data that meets the query conditions to the total data amount of the corresponding data block in any data query operation under the data query task, and the first consumption time corresponding to the target ratio is equal to the second consumption time .

It should be pointed out that, for a data proportion smaller than the target proportion, the corresponding first consumption time is shorter than the second consumption time, and for a data proportion greater than the target proportion, the corresponding first consumption time is greater than the second consumption time.

That is to say, the embodiment of the present application can determine the size of the threshold through experiments, that is, based on a certain test set, the query operation is performed in two ways of starting the accelerator and closing the accelerator respectively. The time consumption of each query method is used to find the intersection point of performance, and the percentage of the amount of data that satisfies the query condition to the total size of the data block is used as the threshold θ. This value can be used as the threshold for whether the FPGA acceleration selector is started or not in the current verified database system.

Step S24: When performing a data query operation for a data block, obtain the average value of the data proportion; wherein, the average data proportion is the average value of the ratio of the amount of data that meets the query conditions in previous data query operations to the total data amount of the corresponding data block , and the data query operations are all query operations under the same data query task, the data block is the data block of the target database, and the target database is the database corresponding to the data query task.

Step S25: Determine whether the average data proportion is greater than the data proportion threshold.

Step S26: If yes, move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.

Of course, in some other embodiments, after the current data block is moved to the FPGA acceleration device for processing, or the current data block is processed by the database management software, the data that meets the query conditions in this and previous data query operations can be The ratio of the amount to the total data amount of the corresponding data block is subjected to median filtering, and the filtering result is used as the data proportion threshold corresponding to the next data query operation, or, based on the type of the target database, in the target database The type of data, the type of query data, the ratio of the amount of data meeting the query conditions in this and previous data query operations to the total data amount of the corresponding data block determines the corresponding data proportion threshold for the next data query operation.

Of course, the data proportion threshold can also be manually determined. Specifically, a threshold determination method suitable for an actual system may be selected according to an actual application scenario, a data storage structure of a database, and query conditions.

Referring to Figure 5, the embodiment of the present application discloses a data query device, which is characterized in that it includes:

The data ratio mean value acquisition module 11 is used to obtain the data ratio mean value when performing a data query operation for a data block; wherein, the data ratio mean value is the amount of data and the corresponding data that meet the query conditions in previous data query operations The average value of the ratio of the total data volume of the block, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database corresponds to the data query task database.

The data proportion threshold judging module 12 is configured to judge whether the average data proportion is smaller than the data proportion threshold.

The data block processing module 13 is used to move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.

It can be seen that in the embodiment of the present application, when the data query operation for the data block is executed, the average value of the data proportion is obtained; wherein, the average value of the data proportion is the amount of data that meets the query conditions in previous data query operations and the total data of the corresponding data block The average value of the ratio of the amount, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database is the database corresponding to the data query task, and then It is judged whether the average value of the data ratio is less than the data ratio threshold, if so, the current data block is moved to the FPGA acceleration device for processing, otherwise the current data block is processed by using the database management software. That is, in this application, when performing a data query operation for a data block under a data query task, the ratio of the amount of data that meets the query conditions in the previous data query operations under the data query task to the total data amount of the corresponding data block Compared with the data ratio threshold, the current data block will be moved to the FPGA acceleration device for processing when it is less than the threshold, otherwise, the current data block will be processed by the database management software, so that the data query is performed according to the amount of data that meets the query conditions Task, to determine whether to move the data block to the FPGA for processing, avoiding the problem of large data delay caused by moving the data block to the FPGA when the amount of data that meets the query conditions is large, and can improve the processing efficiency of the data query task, thereby Improve user experience.

Further, the device also includes a threshold determination module, specifically for processing the data query tasks corresponding to the test data set using the FPGA acceleration device and the database management software respectively; determining the first consumption time for processing by the FPGA acceleration device, and The second consumption time for processing by using database management software; determining a target proportion from all data proportions based on the first consumption time and the second consumption time, and using the target proportion as the data proportion than the threshold;

Wherein, the data ratio is the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block in any data query operation under the data query task, and the first consumption time corresponding to the target ratio It is equal to the second elapsed time.

In some embodiments, the threshold determination module is specifically used to process the current and previous data blocks after the data block processing module 13 moves the current data block to the FPGA acceleration device for processing, or uses the database management software to process the current data block. In the query operation, the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block is subjected to median filtering, and the filtering result is used as the data proportion threshold corresponding to the next data query operation.

In some embodiments, the threshold determination module is specifically used to move the current data block to the FPGA acceleration device for processing by the data block processing module 13, or after the current data block is processed by the database management software, it also includes:

Moreover, the embodiment of the present application also includes a data ratio mean determination module, which is specifically used to calculate the current data block after the data block processing module 13 moves the current data block to the FPGA acceleration device for processing, or uses the database management software to process the current data block. The average value of the ratio of the amount of data that meets the query conditions to the total data amount of the corresponding data block in this and previous data query operations is used as the average value of the data ratio corresponding to the next data query operation.

Further, the device also includes: a data block size determination module, configured to determine the corresponding data block size of each data query operation based on the data storage method and data volume of the target database, and the bandwidth and data processing capability of the FPGA acceleration device. The size of the data block.

The data block processing module 13 is further configured to directly move the current data block to the FPGA acceleration device for processing when the data query operation for the data block under the data query task is executed for the first time.

As shown in FIG. 6, the embodiment of the present application discloses an electronic device 20, including a processor 21 and a memory 22; wherein, the memory 22 is used to store computer programs; the processor 21 is used to execute the A computer program, wherein, when the computer program is executed by a processor, the data query method disclosed in the foregoing embodiments is implemented.

Regarding the specific process of the above data query method, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not repeated here.

Moreover, the memory 22, as a resource storage carrier, may be a read-only memory, random access memory, magnetic disk or optical disk, etc., and the storage method may be temporary storage or permanent storage.

In addition, the electronic device 20 also includes a power supply 23, a communication interface 24, an input and output interface 25, and a communication bus 26; wherein the power supply 23 is used to provide operating voltages for each hardware device on the electronic device 20; the The communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows is any communication protocol applicable to the technical solution of the present application, which is not specifically limited here; The input and output interface 25 is used to obtain external input data or output data to the external, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

Further, the embodiment of the present application also discloses a computer-readable storage medium for storing a computer program, wherein, when the computer program is executed by a processor, the data query method disclosed in the foregoing embodiments is implemented.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same or similar parts of each embodiment can be referred to each other. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and for the related information, please refer to the description of the method part.

The steps of the methods or algorithms described in connection with the embodiments disclosed herein may be directly implemented by hardware, software modules executed by a processor, or a combination of both. Software modules can be placed in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other Any other known storage medium.

The above is a detailed introduction to a data query method, device, equipment and medium provided by this application. In this paper, specific examples are used to illustrate the principle and implementation of this application. The description of the above embodiments is only for helping understanding The method of this application and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation and scope of application. In summary, the content of this specification should not understood as a limitation of the application.

Claims

A data query method, characterized in that, comprising:

When performing a data query operation for a data block, the average value of the data ratio is obtained; wherein, the average value of the data ratio is the average value of the ratio of the amount of data that meets the query conditions in previous data query operations to the total data amount of the corresponding data block, Moreover, the data query operations are all query operations under the same data query task, the data block is a data block of a target database, and the target database is a database corresponding to the data query task;

Judging whether the average value of the data proportion is less than the data proportion threshold;

If yes, move the current data block to the FPGA acceleration device for processing, otherwise use the database management software to process the current data block.
The data query method according to claim 1, further comprising:

The data query task corresponding to the test data set is processed by FPGA acceleration device and database management software;

Determining the first consuming time for processing by the FPGA acceleration device, and the second consuming time for processing by using the database management software;

determining a target proportion from all data proportions based on the first consumption time and the second consumption time, and using the target proportion as the data proportion threshold;

Wherein, the data ratio is the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block in any data query operation under the data query task, and the first consumption time corresponding to the target ratio It is equal to the second elapsed time.
The data query method according to claim 1, wherein, after the current data block is moved to the FPGA acceleration device for processing, or after the current data block is processed by database management software, it also includes:

Median filtering is performed on the ratio of the amount of data meeting the query conditions to the total data amount of the corresponding data block in this and previous data query operations, and the filtering result is used as the data proportion threshold corresponding to the next data query operation.
The data query method according to claim 1, wherein, after the current data block is moved to the FPGA acceleration device for processing, or after the current data block is processed by database management software, it also includes:

Based on the type of the target database, the type of data in the target database, the type of query data, the ratio of the amount of data meeting the query conditions in this and previous data query operations to the total data amount of the corresponding data block to determine the next time The data proportion threshold corresponding to the data query operation.
The data query method according to claim 1, wherein, after the current data block is moved to the FPGA acceleration device for processing, or after the current data block is processed by database management software, it also includes:

Calculate the average value of the ratio of the amount of data that meets the query conditions to the total data amount of the corresponding data block in this and previous data query operations, and use the average value as the average value of the data proportion corresponding to the next data query operation.
The data query method according to claim 1, further comprising:

The size of the data block corresponding to each data query operation is determined based on the data storage method and data volume of the target database, and the bandwidth and data processing capability of the FPGA acceleration device.
The data query method according to claim 1, further comprising:

When performing the data query operation for the data block under the data query task for the first time, the current data block is directly moved to the FPGA acceleration device for processing.
A data query device, characterized in that it comprises:

The data ratio mean value acquisition module is used to obtain the data ratio mean value when performing a data query operation for the data block; wherein, the data ratio mean value is the amount of data that meets the query conditions and the corresponding data block in previous data query operations The average value of the ratio of the total data volume, and the data query operations are all query operations under the same data query task, the data block is a data block of the target database, and the target database is the corresponding data query task database;

A data proportion threshold judging module, configured to determine whether the data proportion average value is less than the data proportion threshold;

The data block processing module is used to move the current data block to the FPGA acceleration device for processing, or use the database management software to process the current data block.
An electronic device, characterized in that it comprises:

memory for storing computer programs;

A processor, configured to execute the computer program, so as to realize the data query method according to any one of claims 1-7.
A computer-readable storage medium, characterized in that it is used to store a computer program, and when the computer program is executed by a processor, the data query method according to any one of claims 1 to 7 is implemented.