CN113282585B

CN113282585B - Report calculation method, device, equipment and medium

Info

Publication number: CN113282585B
Application number: CN202110593382.3A
Authority: CN
Inventors: 徐会
Original assignee: Inspur General Software Co Ltd
Current assignee: Inspur General Software Co Ltd
Priority date: 2021-05-28
Filing date: 2021-05-28
Publication date: 2023-12-29
Anticipated expiration: 2041-05-28
Also published as: CN113282585A

Abstract

The embodiment of the specification discloses a report calculation method, which is used for reducing memory occupation of an application server and reducing database pressure. The method comprises the following steps: determining the identification of the current node when carrying out report calculation; if corresponding first report data exists in the database according to the identification of the current node and a preset configuration file, acquiring the first report data, and completing report calculation according to the first report data; and if the corresponding second report data does not exist in the database according to the identification configuration file of the current node, acquiring the corresponding second report data from the pre-deployed relational database, and completing report calculation according to the second report data.

Description

Report calculation method, device, equipment and medium

Technical Field

The present disclosure relates to the field of big data technologies, and in particular, to a report calculation method, apparatus, device, and medium.

Background

Databases can be divided into two classes according to query types, one is a relational database based on SQL (Structured QueryLanguage ) queries, and the other is a non-relational database based on Key-Value (K-V for short) queries, or called Key-Value database. The relational database is commonly used for data storage and inquiry of complex business relations, and has the advantage of being capable of carrying out aggregation treatment on various data; the non-relational databases are commonly used for simple query, and have the advantages of high speed and common non-relational databases such as HBase and Redis. The Redis is characterized in that the memory is used for data access, so that the data reading and writing speed is greatly improved, and meanwhile, the distributed design enables the Redis to utilize the performance of the computer cluster, so that the problem that the capacity of the single memory is possibly insufficient is solved.

Generally, when report calculation is performed, information such as report format needs to be queried from a relational database, and when a plurality of users calculate simultaneously, the situation that report data used by a current node and a subsequent node are partially repeated can occur. At this time, the problem that the report calculation performance is easily reduced because the memory occupation of the application server is high due to the fact that data are repeatedly called into the relational database through SQL sentences often occurs.

Disclosure of Invention

One or more embodiments of the present disclosure provide a report calculation method, apparatus, device, and medium, which are used to solve the following technical problems: a report calculation method for reducing the request times of a relational database and reducing the memory occupation of a server is needed.

One or more embodiments of the present disclosure adopt the following technical solutions:

one or more embodiments of the present specification provide a report calculation method, including:

determining the identification of the current node when carrying out report calculation;

if the corresponding first report data exists in the database according to the identification of the current node and the preset configuration file, acquiring the first report data, and completing the report calculation according to the first report data;

And if the fact that the corresponding second report data does not exist in the database according to the identification of the current node and the configuration file is determined, acquiring the corresponding second report data through the pre-deployed relational database, and completing the report calculation according to the second report data.

Optionally, in one or more embodiments of the present disclosure, before determining the identity of the current node when performing report calculation, the method further includes:

presetting a configuration file of the key value to a database through an application server; the key value is preset in a database server;

carrying out instance numbering on a database according to a key value which is pre-deployed in a database server according to the configuration file, and setting an IP address and a port number for the server where the database is located for the key value;

selecting whether to enable the key value to the database or not through a cache setting mechanism of the configuration file, and setting cache valid time for the key value to the database;

and if the effective time is exceeded, clearing the report data cached in the database by the key value, and re-acquiring the data information needed by the computing node based on the relational database.

Optionally, in one or more embodiments of the present specification, before the obtaining the first report data, the method further includes:

determining whether the key value needs to cache the report data of the current node or not according to the historical data of the report calculation;

if the fact that the report data need to be cached by the key value to the database is determined, caching the report data with the timestamp; wherein, the time stamp is used for recording the storing time and the effective time of the report data so as to update and delete the report data;

and if the report data is determined not to be cached by the key value to the database, postponing the time point of storing the report data into the key value to the database.

Optionally, in one or more embodiments of the present disclosure, before the obtaining, by the pre-deployed relational database, the corresponding second report data, the method includes:

sending a data query request and an SQL query statement to the relational database so that the current node can acquire second report data; wherein the data query request includes an identification of the current node;

the specific object of the second report data is obtained by converting the second report data into a data type corresponding to the key value stored in the database;

And caching the second report data into the key value database.

Optionally, in one or more embodiments of the present disclosure, before the caching the second report data into the key-value database, the method further includes:

determining whether the sum of the size of the second report data to be stored and the size of the cached report data in the database exceeds a preset limit value of the database by the key value based on a judging mechanism;

if the number of the latest use times of the cached report data exceeds the preset limit value, determining the latest least use times of the cached report data according to a cache replacement algorithm, and dividing the weight value of the cached data by combining the reuse distance of the report data;

and deleting the cached data in sequence according to the weight value until the preset limit value of the key value to the database is not exceeded, so as to ensure the normal process of report calculation.

Optionally, in one or more embodiments of the present disclosure, the determining, according to a cache replacement algorithm, a least recently used number of times of the cached report data, and dividing a weight value of the cached data in combination with a reuse distance of the report data specifically includes:

Analyzing the cached report data based on a pre-filtered replacement algorithm to obtain report data which is least recently used in a database by the key value;

positioning an acquisition instruction of report data in a report calculation process according to a branch prediction technology, and further predicting the reuse distance of the report data;

and obtaining the weight value of the cached report data in the database by the key value based on the least recently used report data and the reuse distance.

Optionally, in one or more embodiments of the present specification, the method further includes:

and if the format information of the report is changed or the record function of the database is actively triggered to be emptied by the key value, an emptying interface is called to empty the data information of the database by the key value, and the data information of the database by the key value is updated according to the format information of the report.

One or more embodiments of the present specification provide a report calculation apparatus, the apparatus including:

the identification determining unit is used for determining the identification of the current node when the report calculation is performed;

the first report data acquisition unit is used for acquiring the first report data and completing the report calculation according to the first report data if the corresponding first report data exists in the database according to the identification of the current node and the preset configuration file;

And the second report data acquisition unit is used for acquiring corresponding second report data through the pre-deployed relational database and completing report calculation according to the second report data if the pre-deployed key value is determined to not have the corresponding second report data according to the identification of the current node and the configuration file.

Optionally, in one or more embodiments of the present specification, the apparatus further includes:

the file configuration unit is used for presetting a configuration file of the key value to the database through the application server; the key value is preset in a database server; carrying out instance numbering on a database according to a key value which is pre-deployed in a database server according to the configuration file, and setting an IP address and a port number for the server where the database is located for the key value; selecting whether to enable the key value to the database or not through a cache setting mechanism of the configuration file, and setting cache valid time for the key value to the database; and if the effective time is exceeded, clearing the report data cached in the database by the key value, and re-acquiring the data information needed by the computing node based on the relational database.

the data caching unit is used for determining whether the key value needs to cache the report data of the current node or not according to the historical data of the report calculation; if the fact that the report data need to be cached by the key value to the database is determined, caching the report data with the timestamp; wherein, the time stamp is used for recording the storing time and the effective time of the report data so as to update and delete the report data; and if the report data is determined not to be cached by the key value to the database, postponing the time point of storing the report data into the key value to the database.

the data acquisition unit is used for sending a data query request and an SQL query statement to the relational database so that the current node can acquire second report data; wherein the data query request includes an identification of the current node; the specific object of the second report data is obtained by converting the second report data into a data type corresponding to the key value stored in the database; and caching the second report data into the key value database.

the buffer memory replacing unit is used for determining whether the sum of the size of the second report data to be stored and the size of the buffered report data in the database exceeds the preset limit value of the database by the key value based on a judging mechanism; if the number of the latest minimum use times of the cached report data exceeds the preset limit value, determining the weight value of the cached report data according to a cache replacement algorithm and combining the reuse distance of the report data; and deleting the cached report data in turn according to the weight value until the preset limit value of the key value to the database is not exceeded, so as to ensure the normal process of report calculation.

the weight setting unit is used for analyzing the cached report data based on a pre-filtering replacement algorithm so as to obtain the report data which is least recently used in the database by the key value; positioning an acquisition instruction of report data in a report calculation process according to a branch prediction technology, and further predicting the reuse distance of the report data; and obtaining the weight value of the cached report data in the database by the key value based on the least recently used report data and the reuse distance.

and the emptying updating unit is used for calling an emptying interface to empty the data information of the key value to the database when the format information of the report is changed or the recording function of the key value to the database is actively triggered, and updating the data information of the key value to the database according to the format information of the report.

One or more embodiments of the present specification provide a report computing device, the device comprising:

at least one processor, and

a memory communicatively coupled to the at least one processor, wherein the memory stores execution instructions for the at least one processor to cause the at least one processor to execute, the execution instructions comprising:

One or more embodiments of the present specification provide a non-volatile storage medium storing executable instructions for a computer, the executable instructions comprising:

The above-mentioned at least one technical scheme that this description embodiment adopted can reach following beneficial effect:

the characteristic grammar provided by the key value to the database is recorded in the form of key-value for different types of report format information in the database. When the report is calculated, the relational database is not required to be frequently inquired, the pressure of the database is reduced, multiple users can share the key value to the report format information in the database, the occupation of the memory of the application server is reduced, and the effect of optimizing the calculation performance of the report is achieved. Based on a replacement caching algorithm and a branch prediction technology, the cached report data in the database is subjected to weight distribution and deletion of the report data with low weight by the key value, so that the memory of the database by the key value is ensured not to exceed a preset value, and meanwhile, the problem of out-of-order execution in the report calculation process is avoided.

Drawings

In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some of the embodiments described in the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

In the drawings:

fig. 1 is a method flowchart of a report calculation method according to an embodiment of the present disclosure;

fig. 2 is a schematic flow chart of a report calculation method in an application scenario provided in the embodiment of the present disclosure;

fig. 3 is a schematic diagram of an internal structure of a report calculation method according to an embodiment of the present disclosure;

fig. 4 is a schematic diagram of an internal structure of a report calculation method according to an embodiment of the present disclosure;

fig. 5 is a schematic diagram of an internal structure of a nonvolatile storage medium according to an embodiment of the present disclosure.

Detailed Description

The embodiment of the specification provides a method, a device and equipment for realizing query dynamic columns.

In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.

In the operation process of a daily enterprise, a large amount of data needs to be processed and calculated, and a report is generated and displayed by utilizing the functions of computer processing data and interface design. Redis (Remote Dictionary Server) is a kind of Key-Value pair database, namely a remote dictionary service, which is an open-source log-type, key-Value database written in ANSI C language, supported by network, capable of being based on memory and persistent, and provides multiple language APIs. Currently, most systems use a memory to cache frequently used data, and the size of the memory determines the amount of data to be cached. In general statistical report calculation, information about report formats and the like needs to be queried from a relational database. Moreover, when a plurality of users calculate simultaneously, the condition that SQL sentences are repeatedly executed and the memory occupation of an application server is high often occurs, so that the performance of report calculation is reduced.

The invention records the report format information of different types in the database in the form of key-value by means of the special grammar of the key-value on the database itself. The report data is cached by using the key value to the database, and the report data is acquired by combining the relational database, so that the report data can be directly acquired by the data in a key value matching mode. And, when the report data is needed to be used in the next calculation or other user calculation, the report data can be directly obtained from the key value database. The request times of the relational database are reduced, the occupation of the memory of the server is reduced, the report calculation performance is optimized, and the report calculation speed is improved. The data cached in the database is judged by the cache replacement algorithm, so that the report data cached in the database can be replaced orderly by the key value, the efficiency of report calculation is improved, and the problem that unnecessary data occupy a large amount of memory is solved.

The execution body of one or more embodiments of the present disclosure is an execution unit where a current node is located.

Fig. 1 is a flow chart of a report calculation method according to one or more embodiments of the present disclosure, and as can be seen from fig. 1, the method includes the following steps:

s101, determining the identification of the current node when carrying out report calculation.

When report calculation is performed, the identification of the current node needs to be determined.

In one or more embodiments of the present disclosure, before determining the identity of the current node when performing report calculation, the method further includes:

Before the current node performs report calculation, a database and related information for performing report data caching need to be determined, and in one or more embodiments of the present disclosure, an execution unit of the current node needs to dispose a key value in a database server in advance, and preset a configuration file in an application server to set the key value to the database and a database server where the key value is located. For example, when the Redis database is selected as a key value pair database required in report calculation, a configuration file StaticReportRedis Config. Xml can be deployed in each application server to perform relevant configuration on the Redis database and a database server where the Redis database is located, and the main content of the configuration can be realized by the following codes:

<？xml version＝"1.0"？>

</CacheSetting>

it should be noted that: before the report calculation of the current node is performed, the key value needs to be preconfigured on the database and the database server where the key value is located. The execution unit of the current node numbers the database instance by using the AppCode in the application server, for example: the database in the above example is numbered MT88; setting an IP address and a port number for a database server through IPAndPort, for example: the IP address configured in the above configuration content is 127.0.0.1, and the port number of the database server is 6379.

In real-world applications, report calculations performed by some users are not conventional, and report data that needs to be used is not multiplexed by itself or by other users for a long period of time. At this time, if the report data required by the user is added to the database by the key value, not only the storage space of the key value to the data is occupied, but also the overall performance and the operation speed of the report calculation are adversely affected. Therefore, the configuration file sets whether to enable the key value to the database through State so as to avoid misuse of the key value to the database. Wherein, when State is set to 1, it represents the database using key values, and when State is set to 0, it represents the database not applicable to key values. Likewise, to ensure the key value's data updating capability to the database, the key value's effective time to the database is set in the configuration file by VaildTime. The effective time can be correspondingly set according to specific report calculation business. If the effective time is exceeded, the key value clears the data cached by the database, and the data required by report calculation is acquired from the relational database again.

It should be noted that Redis is a key-value storage system. To ensure efficiency, data is buffered in memory. The Redis can periodically write updated data into the disk or write modification operation into the additional record file, so that the cache can be recovered from the disk file after power failure and restarting.

S102, if the corresponding first report data exists in the database according to the identification of the current node and the preset configuration file, acquiring the first report data, and completing the report calculation according to the first report data.

In one or more embodiments of the present disclosure, before the obtaining the first report data, the method includes:

And if the corresponding report data can be matched in the database according to the identification of the current node, directly acquiring the report data. Taking the Redis database in the database as an example of the key value as in step 101, if State in the configuration file is set to 1, that is, the Redis database is used in the preset configuration file. Inquiring whether corresponding report data exists in a Redis database according to the identification of the current node, wherein the main codes of the inquiring step are as follows:

public static bool ContainKey(string dicName,string key)

{

if(State＝＝"1")

{

return＝RedisHelper.HExists(dicName,key)；

}

If it is determined that report data of the current node is cached in the Redis database according to the above-mentioned query step, report data corresponding to the identification of the current node may be directly obtained from the Redis database. The main codes for acquiring report data are as follows:

if the report data required by the current node can be directly obtained from the key value pair database, the key value pair database caches the related report data before the current node, but in order to ensure the capacity of the memory, the efficiency of report calculation is improved, and the key value may have capacity limitation on the report data cached in the database. Therefore, in order to ensure that report calculation can reduce the number of times report data is called from the relational database, in one or more embodiments of the present disclosure, machine learning may be performed according to the history data of report calculation, so as to obtain report data required in the report calculation process, and cache the required report data into the database with a key value.

When the information format of the report data is updated continuously and the report data of the current node is determined to be cached, a time stamp containing the storing time and the effective time is attached to the report data and cached in the key value database. The report data in the database can be continuously updated by the key value based on the limit of the effective time, so that the method is suitable for different format information required by different users in report calculation, and the continuous updating and expanding of the key value to the database are ensured.

And when the report data of the current node is determined according to the historical data and the data does not need to be cached by the key value. When the historical data of machine learning is analyzed, and the report data needed by the current node is determined to have no multiplexing or extremely low multiplexing probability, caching the report data is refused or the time point of storing the report data into the database by the key value is deferred. The report data stored in the key value database is repeatedly called data with high probability through the process, so that the effect of relieving the pressure of the memory is achieved.

And S103, if the fact that the corresponding second report data does not exist in the database according to the identification of the current node and the configuration file is determined, acquiring the corresponding second report data through the pre-deployed relational database, and completing report calculation according to the second report data.

In one or more embodiments of the present disclosure, before the obtaining, by the pre-deployed relational database, the corresponding second datagram table information, the method includes:

and caching the second report data into the key value database.

When the current node acquires second report data required by report calculation, if State in the configuration file is set to 0, namely the configuration file is configured to not use the Redis database, or when the corresponding second report data cannot be queried in the Redis database according to the identification of the current node, a data query request needs to be sent to the relational database, and the required second report data is acquired through SQL query statements. Meanwhile, in order to facilitate report calculation in the subsequent nodes, the second report data obtained from the relational database can be counted into the Redis database according to the situation. Because the data in the Redis database is stored in the key-value form, the second report data acquired from the relational database needs to be converted into the corresponding DataTable, list < string > data type, and then recorded in the Redis database in the form of the identifier-data object. So that when the second report data is needed to be used in the next calculation or other user calculation, the second report data can be directly obtained from the Redis database according to the identification inquiry of the current node, so that the number of times of requesting the data from the relational database again is reduced, and the report calculation efficiency and performance are improved. Through the combination of the relational database and the key value, the query calculation of SQL can be carried out with high concurrency and high performance during report calculation, and the performance of report calculation is improved.

In one or more embodiments of the present disclosure, before the caching the second report data in the key-value database, the method further includes:

if the number of the latest minimum use times of the cached report data exceeds the preset limit value, determining the weight value of the cached report data according to a cache replacement algorithm and combining the reuse distance of the report data;

and deleting the cached report data in turn according to the weight value until the preset limit value of the key value to the database is not exceeded, so as to ensure the normal process of report calculation.

In one or more embodiments of the present disclosure, the determining, according to a cache replacement algorithm, a least recently used number of times of the cached report data, and dividing a weight value of the cached report data in combination with a reuse distance of the report data specifically includes:

After the current node needing report calculation obtains a second report database cached in the database for the key value according to the relational database, in order to facilitate report calculation of other users or subsequent nodes, the obtained second report data needs to be considered to be recorded in the database for the key value. In one or more embodiments of the present disclosure, in order to ensure that a key value limits the data capacity that can be cached in the database for the service processing speed of the database, so as to ensure that the running speed thereof improves the efficiency of report calculation. When the second report data is stored in the key value pair database, whether the sum of the size of the second report data with the stored data plus the cached report data exceeds the preset limit value of the key value pair database is judged.

When the limit value is exceeded, one or more cached report data in the database needs to be deleted by the key value pair, so that the second report data can be successfully recorded. In one or more embodiments of the present disclosure, a cache replacement optimization algorithm based on pre-filtering is adopted to obtain report data with a key value least recently used in a database, and an acquisition instruction of the report data in a predicted report calculation process is positioned by combining a branch prediction technology, so as to obtain a reuse distance of the report data, and calculate a weight value of the cached report data. The reuse distance of the report data is obtained based on the branch prediction technology, the disorder execution of the report data and the interference of cache prefetching are avoided, the uncertainty faced in the prediction process is reduced, and the precision of the cache data replacement process is improved.

In one or more embodiments of the specification, an average of the least recently used times and the 1/reuse distance of the cached report data is used as a weight value of the cached report data, wherein a smaller weight indicates a lower availability of the data. And deleting the numerical values with small weights in sequence according to the obtained weight values until the second report data are stored and do not exceed the preset limit value of the database by the key values.

In one or more embodiments of the present specification, the method further comprises:

If the format information of the report changes when the current node performs report calculation, or if the user starts a record function of clearing the Redis database in the previous node, the clearing Redis data interface is called to clear the content in the Redis database, wherein the main codes of clearing the Redis database are as follows:

when different users perform report calculation or multiple threads perform report calculation, key values can be shared between different users and multiple threads for data in the database, so that the repeated query and call processes of the data from the relational data are reduced, repeated query sentences are saved, the calculation speed of the report is improved, and the calculation performance of the report is optimized. When the multi-application server configures and deploys the key value to the database and the server where the key value is located, the key value can update and acquire the database at any time, so that the situation of inconsistent data among all servers is avoided, and the situation of inaccurate report calculation is avoided.

As shown in fig. 2, one or more embodiments of the present disclosure provide a flowchart of report calculation in an application scenario.

As shown in FIG. 2, in one or more embodiments of the present disclosure, a current node may deploy a Redis service through a database server and a configuration file based on an application server prior to performing report calculations. When the calculation is started, each node to be calculated generates a unique node identifier, and whether report data needed at the moment is contained in the Redis database is judged according to the identifier. If the Redis database is a record, the record is required to be inquired out of the relational database through SQL sentences and is recorded in the Redis database; if the report data exists in the Redis database, the report data is directly acquired, the business operation corresponding to the current node is completed by acquiring the corresponding report data, and the steps are circulated to calculate the next calculation node until the calculation task of the report is completed.

By combining the key values with the relational database, the relational database does not need to be frequently queried to acquire report data when the report is calculated, and the pressure of the database is reduced. And multiple users can directly share report format information in Redis, so that the memory occupation of an application server is reduced, and the aim of optimizing the calculation performance of the report is fulfilled.

As shown in fig. 3, one or more embodiments of the present disclosure provide an internal structural schematic diagram of a report computing device, which, as shown in fig. 3, includes:

the identification determining unit 301 is configured to determine an identification of a current node when performing report calculation;

a first report data obtaining unit 302, configured to obtain first report data if it is determined that, according to the identifier of the current node and a preset configuration file, a corresponding first report data exists in the database by using a pre-deployed key value, and complete the report calculation according to the first report data;

and the second report data obtaining unit 302 is configured to obtain, if it is determined, according to the identifier of the current node and the configuration file, that the pre-deployed key value does not have corresponding second report data in the database, the corresponding second report data through the pre-deployed relational database, and complete the report calculation according to the second report data.

In one or more embodiments of the present description, the apparatus further comprises:

As shown in fig. 4, one or more embodiments of the present specification provide a report computing device, the device comprising:

at least one processor 401; the method comprises the steps of,

a memory 402 communicatively coupled to the at least one processor; wherein,

the memory 402 stores instructions executable by the at least one processor 401, the instructions being executable by the at least one processor 401 to enable the at least one processor 401 to:

As shown in fig. 5, one or more embodiments of the present description provide a non-volatile storage medium storing computer-executable instructions 501, the computer-executable instructions 501 comprising:

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, devices, non-volatile computer storage medium embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the section of the method embodiments being relevant.

The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present description, is intended to be included within the scope of the claims of the present description.

Claims

1. A report calculation method, the method comprising:

if the fact that the corresponding second report data does not exist in the database according to the identification of the current node and the configuration file is determined, acquiring the corresponding second report data through the pre-deployed relational database, and completing report calculation according to the second report data;

before the corresponding second report data is acquired through a pre-deployed relational database, the method comprises the following steps:

Caching the second report data into the key value database;

before the first report data is acquired, the method further includes:

determining whether the key value needs to cache the report data of the current node or not according to the historical data of the report calculation; the determining whether the key value needs to cache the report data of the current node according to the historical data of the report calculation specifically includes:

the historical data calculated by the report is machine-learned to determine the multiplexing probability of the report data of the current node;

if the multiplexing probability is smaller than a preset threshold, the report data of the current node does not need to be cached, and if the multiplexing probability is larger than the preset threshold, the report data of the current node needs to be cached;

if the report data is determined not to be cached by the key value to the database, deferring the time point of storing the report data into the key value to the database;

Before the second report data is cached in the key value database, the method further comprises:

sequentially deleting the cached report data according to the weight value until the preset limit value of the database by the key value is not exceeded, so as to ensure the normal process of report calculation;

the method for determining the least recently used times of the cached report data according to the cache replacement algorithm and dividing the weight value of the cached report data by combining the reuse distance of the report data specifically comprises the following steps:

2. The report calculation method according to claim 1, wherein before determining the identity of the current node when performing report calculation, the method further comprises:

3. The report calculation method of claim 1, further comprising:

4. A report computing device, the device comprising:

the second report data acquisition unit is used for acquiring corresponding second report data through a pre-deployed relational database if the corresponding second report data does not exist in the pre-deployed key value database according to the identification of the current node and the configuration file, and completing report calculation according to the second report data; before the corresponding second report data is obtained through the pre-deployed relational database, the method further comprises the following steps:

caching the second report data into the key value database;

the apparatus further comprises:

the buffer memory determining unit is used for determining whether the key value needs to buffer the report data of the current node to the database according to the historical data of the report calculation; the determining whether the key value needs to cache the report data of the current node according to the historical data of the report calculation specifically includes: the historical data calculated by the report is machine-learned to determine the multiplexing probability of the report data of the current node; if the multiplexing probability is smaller than a preset threshold, the report data of the current node does not need to be cached, and if the multiplexing probability is larger than the preset threshold, the report data of the current node needs to be cached;

The updating and caching unit is used for caching the report data with the timestamp if the fact that the report data need to be cached by the key value to the database is determined; wherein, the time stamp is used for recording the storing time and the effective time of the report data so as to update and delete the report data;

the deferred caching unit is used for deferring the time point of storing the report data into the key value database if the report data is determined not to be cached by the key value database;

the apparatus further comprises:

the limit value determining unit is used for determining whether the sum of the size of the second report data to be stored and the size of the cached report data in the database exceeds the preset limit value of the database by the key value based on a judging mechanism;

the weight dividing unit is used for determining the least recently used times of the cached report data according to a cache replacement algorithm if the weight dividing unit exceeds the preset limit value, and dividing the weight value of the cached report data by combining the reuse distance of the report data;

the deleting unit is used for sequentially deleting the cached report data according to the weight value until the preset limit value of the key value to the database is not exceeded, so as to ensure the normal process of report calculation;

The weight dividing unit is specifically configured to:

5. A reporting computing device, the device comprising:

at least one processor; the method comprises the steps of,

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to:

before the corresponding second report data is acquired through the pre-deployed relational database, the method comprises the following steps:

caching the second report data into the key value database;

before the first report data is acquired, the method further includes:

6. A non-volatile storage medium storing executable instructions for a computer, the executable instructions comprising:

caching the second report data into the key value database;

before the first report data is acquired, the method further includes: