CN114489501A - Real-time big data processing system and method - Google Patents
Real-time big data processing system and method Download PDFInfo
- Publication number
- CN114489501A CN114489501A CN202210058389.XA CN202210058389A CN114489501A CN 114489501 A CN114489501 A CN 114489501A CN 202210058389 A CN202210058389 A CN 202210058389A CN 114489501 A CN114489501 A CN 114489501A
- Authority
- CN
- China
- Prior art keywords
- data
- real
- component
- asynchronous
- data processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
- G06F3/0613—Improving I/O performance in relation to throughput
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
- G06F3/0674—Disk device
- G06F3/0676—Magnetic disk device
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a real-time big data processing system and a method thereof, wherein the method comprises the following steps: asynchronously receiving, compressing/decompressing data by adopting an asynchronous data receiving component, and sending the data to a multithread data processing component; analyzing the data messages from the asynchronous data receiving component by adopting a multithread data processing component, carrying out asynchronous processing, and inserting the data messages into a column type storage database in batches; the role-based peer-to-peer multiple column storage databases are adopted to provide a materialized view function so as to improve the storage query performance. By adopting the data asynchronous receiving and concurrent analysis processing technology and based on the column storage and related components, the application performance index data can be analyzed, processed, stored and analyzed in real time under the condition of using less system resources, and the operation and maintenance efficiency is improved.
Description
Technical Field
The invention belongs to the field of application performance monitoring, in particular to a real-time big data processing technology which is mainly used for big data processing analysis with high throughput and low delay on monitoring data.
Background
At present, in the field of application performance monitoring, a technical scheme for collecting data in real time by a related probe (agent) and efficiently receiving, processing and storing the data collected by the probe in real time has the problems of large consumption of hardware resources and poor real-time performance, and is difficult to consider both throughput rate and delay rate. Generally, the throughput rate is high, but the data delay is high, and the real-time performance is difficult to achieve; or in high real-time but with low data throughput. When both are considered, high system resources are consumed, so the solutions surrounding the conventional data reception and processing have not met the relevant performance requirements.
For example, the conventional data receiving scheme generally adopts a BIO model, that is, one thread corresponds to one request, which results in many requesting threads being in a state of waiting for io resources, and thus cannot be effectively utilized. Since creating a thread is a relatively expensive system resource, when the amount of concurrency increases, the resource load of the system becomes large, and the requirement for high throughput cannot be satisfied. In addition, the single-machine or master-slave design mode of the traditional database cannot meet the requirements of high throughput and low delay of the database.
Disclosure of Invention
In order to overcome the above drawbacks of the prior art, an object of the present invention is to provide a real-time big data processing technology, which is based on asynchronous data reception of an asynchronous event model, and can achieve a low consumption of thread resources and a high data reception capability.
The invention also aims to provide a real-time big data processing technology, and the materialized view function provided by the column database can reduce the number of data and improve the query performance.
The invention aims to provide a real-time big data processing technology, which combines asynchronous data receiving based on an asynchronous event model with materialized view functions provided by a column database, not only can consume smaller thread resources and achieve higher data receiving capacity, but also can reduce the number of data and improve query performance, thereby simultaneously achieving the purposes of low data delay, high throughput rate and less resource consumption.
In order to achieve the above purpose, the main technical scheme provided by the invention comprises:
a real-time big data processing system, comprising:
the asynchronous data receiving component is used for asynchronously receiving, compressing/decompressing data and sending the data to the multithreading data processing component, namely the distributed data analysis service component;
the multithreading data processing component, namely a distributed data analysis service component, is used for analyzing the data message from the asynchronous data receiving component, carrying out asynchronous processing and inserting the data message into the column type storage database in batches;
and the column storage databases have equal roles and provide a materialized view function.
The real-time big data processing technology provided by the invention can analyze, process, store and analyze the application performance index data in real time under the condition of utilizing less system resources through the data asynchronous receiving and concurrent analysis processing technology and based on the column type storage and related components, thereby improving the operation and maintenance efficiency.
In the real-time big data processing system according to an embodiment of the present invention, the asynchronous data receiving component asynchronously receives and compresses data by using a non-blocking io (no-blocking io) technology, so that a data receiving capability is improved, and a transmission efficiency of data in a distributed system is improved.
The real-time big data processing system of one embodiment of the invention, wherein the asynchronous data receiving component is based on an asynchronous event model, is uniformly coordinated by providing an EventLoop component, is based on event driving, provides a thread for processing when receiving an event, and can process other tasks when waiting for io. Thus a single service can achieve 10000 +/sec data reception capability.
In the real-time big data processing system according to an embodiment of the present invention, the multithread data processing component is a distributed data analysis service component, and the data can be analyzed and processed in real time in a large batch by using distributed analysis processing.
In the real-time big data processing system of one embodiment of the invention, the roles of the multiple columnar storage databases are equal, and each database can be written in, so that the resources of the server can be fully utilized. When the data volume is large, the column-type storage databases can be transversely expanded to share the pressure, and meanwhile, each column-type storage database provides a backup mechanism to guarantee high availability of the data. Preferably, the column-type database provides a materialized view function, and the query performance is improved by replacing time with space, wherein when data is written into the original table, the data can be aggregated in advance according to data characteristics, so that the number of data is greatly reduced, and the query performance is improved.
The real-time big data processing technology of a preferred embodiment of the present invention, wherein the writing mechanism of the data into the original table is specifically:
the database driving is realized through a standard jdbc standard, the connection with data is established through an http mode, a low-frequency and large-batch write strategy is adopted in a write-in mode, and the write-in frequency does not exceed 1 second every time for a single database instance. Tens of thousands of strips are written in a single batch. And writing the data into a temporary disk directory of the database, and merging the background of the database.
In the real-time big data processing technique according to a preferred embodiment of the present invention, the merging mechanism of the early aggregation specifically includes:
by adopting an LSM (Log Structured Merge Trees for short), a data file with a corresponding number is formed in a related directory every time a database is written. When a plurality of data files exist in the data directory, the background thread can merge the plurality of data files which are written for many times into one data file, and the process is accompanied with a compression process so as to achieve the effect of aggregating and merging in advance.
The invention also provides a real-time big data processing method, which comprises the following steps:
s1, asynchronously receiving, compressing and decompressing data by adopting an asynchronous data receiving component, and sending the data to a multithread data processing component;
s2, analyzing the data message from the asynchronous data receiving component by adopting a multi-thread data processing component, carrying out asynchronous processing, and inserting the data message into a column type storage database in batch;
s3, providing the materialized view function by adopting a plurality of column type storage databases with equal roles to improve the storage query performance.
In step S1, the real-time big data processing method according to an embodiment of the present invention uses a nio technique to asynchronously receive and compress data, specifically, the asynchronous data receiving component is based on an asynchronous event model, and unified coordination is performed by providing an EventLoop component, based on event driving, when an event is received, a thread is provided for processing, and when the thread waits for io, the thread can process another task.
According to the real-time big data processing method, the materialized view function of the database is stored in a column mode, the time is changed by space, the query performance is improved, and when data are written into an original table, the data can be aggregated in advance according to the data characteristics. By using the column type storage component on the bottom layer, in the aspect of query, the query performance is optimized and analyzed through a user-defined function, the total disk I/O is reduced after data compression, and the data volume needing to be loaded from a disk is reduced, so that the whole system has the capability of high throughput and low delay, and finally the operation and maintenance information can be analyzed and displayed in multiple dimensions.
The real-time big data processing method of one embodiment of the invention, wherein the writing mechanism of the data writing into the original table is specifically as follows:
the writing mechanism of data writing into the original table is specifically as follows: the database driving is realized through the standard jdbc standard, the connection with the data is established through an http mode, the data is written into the temporary disk directory of the database, and the database is merged at the background.
The real-time big data processing method of one embodiment of the invention, wherein, the merging mechanism of the advance aggregation specifically comprises:
and by adopting an LSM algorithm, a data file with a corresponding number is formed in the relevant directory every time the database is written. When a plurality of data files exist in the data directory, the background thread can merge the plurality of data files which are written for many times into one data file, and the process is accompanied with a compression process so as to achieve the effect of aggregating and merging in advance.
The invention has the beneficial effects that:
according to the invention, through the data asynchronous receiving and concurrent analysis processing technology and based on the column type storage and related components, the application performance index data can be analyzed, processed and stored in real time under the condition of using less system resources, and the operation and maintenance efficiency is improved. The large data real-time processing and analyzing technology based on column-type storage is combined with the asynchronous multithreading receiving and analyzing component, the technical bottleneck problem of data access, analysis, storage and analysis of data collected by each agent is effectively solved, real-time high-throughput data access, high-performance storage and multidimensional query and analysis functions are provided, downstream components or business analysis personnel are helped to know the application performance problem, the positioning of performance faults is facilitated, the customer experience is improved, and the performance is improved. Through actual production verification, the invention can receive, analyze and process data sent by the application probe in real time, daily data processing capacity reaches TB level, data delay is low, throughput rate is high, failure rate is low, resource consumption is low, fault transfer capability is provided, high availability can be supported by 100%, and perfect application performance analysis and diagnosis function is provided.
Drawings
FIG. 1 is a schematic diagram of an overall framework of a real-time big data processing technique according to the present invention;
FIG. 2 is a block diagram of an asynchronous data receiving component in a real-time big data processing technique according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a structure of a columnar storage database in the real-time big data processing technology according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a materialized view writing mechanism of a columnar storage database in a real-time big data processing technology according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating a data merging mechanism for materialized views of a columnar storage database in a real-time big data processing technique according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating real-time big data processing and analysis in the real-time big data processing technology according to an embodiment of the present invention;
FIG. 7 is a schematic diagram illustrating a logic flow of data analysis in a real-time big data processing technique according to an embodiment of the present invention;
fig. 8 is a schematic diagram of an interface analysis function in the real-time big data processing technology according to an application example of the present invention.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
Referring to fig. 1, a real-time big data processing technique according to an embodiment of the present invention includes:
a data receiving component ApmCollector for asynchronous data (e.g., probe data) reception, compression, etc.;
a plurality of data processing and analyzing components ApmConSumer are distributed and analyzed concurrently and are used for data cleaning;
the database columnar storage component is used for data storage and multi-dimensional query;
and the database query analysis component provides a data analysis query result required by the user.
Referring to fig. 2, a real-time big data processing technique according to an embodiment of the present invention is shown, in which:
the thread comprises 3 parts, wherein the first part is a BossEventLoopgroup thread group which is used for receiving connection;
the second part is an EventLoop thread group which is used for processing various events through polling;
the third part is WorkerEventLoopGroup, which is used as a work thread to process relevant events.
Therefore, even when the concurrency amount is increased, the resource load of the system is not too large, and the requirement of high throughput can be met.
Referring to fig. 3, a real-time big data processing technique according to an embodiment of the present invention is shown, in which:
the data writing process of the Consumer writing the data into the distributed table may be that the data is written into a temporary directory of the database first, and then the data is distributed to the corresponding database shard through the data distribution sharing thread. For example, as shown, data is written to the temporary directory tsb _ distribution of host1, and then distributed to the machines of host1 and host 3.
Therefore, the requirements on high throughput and low delay of the database can be met.
Referring to FIG. 4, a real-time big data processing technique according to an embodiment of the present invention is shown, in which data is written into a pipelined table app _ request _ all, and then an event is triggered through a predetermined mechanism to write the data into a physical table through a materialized view. The physical table comprises an aggregation process, and when the physical table is inquired, the inquiry speed can be improved by inquiring the physical table.
The materialized view is queried in the following two ways, and the result is the same:
a select xx from mv _ app _ request _ aggr _ min _ all
Mode two select xx from app _ request _ aggr _ min _ all
Wherein, whether the query mv _ app _ request _ aggr _ min _ all is the target table app _ request _ aggr _ min _ all or the query is.
Referring to fig. 5, a real-time big data processing technique according to an embodiment of the present invention is shown, in which: a data aggregation process is shown, as shown in the figure, two pieces of data (the two pieces of data are identical except for a current _ time field) are written into app _ request _ all at 15:00:00 and 15:10:00 respectively, the running chart stores two pieces of records, after the corresponding materialized views (which can be aggregation granularity of 15 minutes) are aggregated, the materialized view chart only needs to store one piece of data record, and the time field time _ slot of the record is 15:00: 00. The data size is reduced, so that the data query speed can be improved.
The real-time big data processing technology adopts an asynchronous data receiving component to asynchronously receive, compress/decompress data and send the data to a multithread data processing component; analyzing the data messages from the asynchronous data receiving component by adopting a multithread data processing component, carrying out asynchronous processing, and inserting the data messages into a column type storage database in batches; the method adopts a plurality of column type storage databases with equivalent roles to provide a materialized view function to improve the storage and query performance, and can analyze, process and store the application performance index data in real time under the condition of utilizing less system resources by adopting a data asynchronous receiving and concurrent analysis processing technology and based on column type storage and related components, thereby improving the operation and maintenance efficiency.
The technique of the present invention will be described in detail below as an application example of the present invention.
See fig. 6-8, where:
fig. 6 shows that real-time big data processing can be well completed through three layers of a data receiving layer, a data processing layer and a data storage layer. Fig. 7 and 8 are data display effect diagrams, and relevant information can be acquired through data display.
In conclusion, the real-time big data processing technology of the invention combines asynchronous multithreading receiving and analyzing components and the column-type storage big data real-time processing and analyzing technology, effectively solves the technical bottleneck problem of data access, analysis, storage and analysis of data collected by each agent, solves the problems of large hardware resource consumption and poor real-time performance, can simultaneously consider the throughput rate and the delay rate, provides the functions of real-time high-throughput data access, high-performance storage and multidimensional query and analysis, helps downstream components or service analysts to know the application performance problem, is helpful for positioning performance faults, improves the customer experience and improves the performance.
Claims (10)
1. A real-time big data processing system, comprising:
the asynchronous data receiving component is used for asynchronously receiving, compressing/decompressing data and sending the data to the multithreading data processing component, namely the distributed data analysis service component;
the multithreading data processing component, namely a distributed data analysis service component, is used for analyzing the data message from the asynchronous data receiving component, carrying out asynchronous processing and inserting the data message into the column type storage database in batches;
and the column storage databases have equal roles and provide a materialized view function.
2. The real-time big data processing system of claim 1, wherein the asynchronous data reception component asynchronously receives, compresses data using a nio technique.
3. The real-time big data processing system of claim 2, wherein the asynchronous data receiving component comprises a unified coordination by providing an EventLoop component based on an asynchronous event model, and providing a thread to process when an event is received based on event driving.
4. The real-time big data processing system according to claim 3, wherein the thread can process another task while the thread waits for io.
5. The real-time big data processing system of any one of claims 1 to 4, wherein the multi-threaded data processing component is a distributed data parsing services component.
6. The real-time big data processing system of claim 5, wherein the materialized view functions of the columnar storage database include: when data is written into the original table, the data is aggregated in advance according to the data characteristics.
7. A real-time big data processing method is characterized by comprising the following steps:
s1, asynchronously receiving, compressing and decompressing data by adopting an asynchronous data receiving component, and sending the data to a multithread data processing component;
s2, analyzing the data message from the asynchronous data receiving component by adopting the multithread data processing component, carrying out asynchronous processing, and inserting the data message into the column type storage database in batch;
s3, providing the materialized view function by adopting a plurality of column type storage databases with equal roles to improve the storage query performance.
8. The method as claimed in claim 7, wherein the step S1 utilizes nio technology to asynchronously receive and compress data, and in particular, the asynchronous data receiving component is based on an asynchronous event model, and is uniformly coordinated by providing an EventLoop component, and based on event driving, when an event is received, a thread is provided for processing, and when the thread waits for io, the thread can process other tasks.
9. The real-time big data processing method according to claim 7, wherein the column stores materialized view functions of the database to replace time by space to improve query performance, and data are aggregated in advance according to data characteristics when written into the original table.
10. The real-time big data processing method according to claim 9, wherein:
the writing mechanism of data writing into the original table is specifically as follows: the database driving is realized through the standard jdbc standard, the connection with the data is established through an http mode, the data is written into a temporary disk directory of the database, and the database is subjected to background merging; and/or
The merging mechanism of the advanced aggregation is specifically as follows: and when the data directory has a plurality of data files, the background threads can combine the data files which are written for many times into one data file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210058389.XA CN114489501A (en) | 2022-01-19 | 2022-01-19 | Real-time big data processing system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210058389.XA CN114489501A (en) | 2022-01-19 | 2022-01-19 | Real-time big data processing system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114489501A true CN114489501A (en) | 2022-05-13 |
Family
ID=81472599
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210058389.XA Pending CN114489501A (en) | 2022-01-19 | 2022-01-19 | Real-time big data processing system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114489501A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699172A (en) * | 2021-01-06 | 2021-04-23 | 中车青岛四方机车车辆股份有限公司 | Data processing method and device for railway vehicle |
| CN115277361A (en) * | 2022-06-29 | 2022-11-01 | 国家电网公司华中分部 | Intelligent information system performance analysis method based on big data |
-
2022
- 2022-01-19 CN CN202210058389.XA patent/CN114489501A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112699172A (en) * | 2021-01-06 | 2021-04-23 | 中车青岛四方机车车辆股份有限公司 | Data processing method and device for railway vehicle |
| CN115277361A (en) * | 2022-06-29 | 2022-11-01 | 国家电网公司华中分部 | Intelligent information system performance analysis method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114489501A (en) | Real-time big data processing system and method | |
| JP6388655B2 (en) | Generation of multi-column index of relational database by data bit interleaving for selectivity | |
| US11841862B2 (en) | Query execution via virtual segments | |
| CN102831245B (en) | A kind of real-time data memory of relevant database and read method | |
| US20200117663A1 (en) | Managed query execution platform, and methods thereof | |
| CN113360554B (en) | Method and equipment for extracting, converting and loading ETL (extract transform load) data | |
| Hao et al. | Ts-benchmark: A benchmark for time series databases | |
| CN107346270B (en) | Method and system for real-time computation based radix estimation | |
| CN109669975B (en) | Industrial big data processing system and method | |
| CN111914013B (en) | Data management method, system, terminal and medium based on pandas database and InfluxDB database | |
| CN113312376A (en) | Method and terminal for real-time processing and analysis of Nginx logs | |
| CN117056303B (en) | Data storage method and device suitable for military operation big data | |
| CN116680315A (en) | Data offline processing method and device, electronic equipment and storage medium | |
| CN107590199B (en) | Memory-oriented multithreading database design method | |
| CN116431635A (en) | Lake and warehouse integrated-based power distribution Internet of things data real-time processing system and method | |
| CN113656370B (en) | Data processing method and device for electric power measurement system and computer equipment | |
| CN116700917A (en) | Data decision platform and use method | |
| CN115994144A (en) | Data storage method and device, storage medium and electronic equipment | |
| Munirathinam et al. | Design and implementation of manufacturing data lake in Hadoop | |
| Jiadi et al. | Research on Data Center Operation and Maintenance Management Based on Big Data | |
| CN111641874A (en) | Distributed computing method, system and readable storage medium | |
| CN115952200B (en) | MPP architecture-based multi-source heterogeneous data aggregation query method and device | |
| CN117390040B (en) | Service request processing method, device and storage medium based on real-time wide table | |
| Zhao et al. | Analyzing and Applying Big Data Technologies Based on Industrial Internet Data | |
| CN112131302B (en) | Commercial data analysis method and platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |