CN111209414B - Method for realizing cold-hot separation storage of data based on image data calling business scene - Google Patents
Method for realizing cold-hot separation storage of data based on image data calling business scene Download PDFInfo
- Publication number
- CN111209414B CN111209414B CN202010004175.5A CN202010004175A CN111209414B CN 111209414 B CN111209414 B CN 111209414B CN 202010004175 A CN202010004175 A CN 202010004175A CN 111209414 B CN111209414 B CN 111209414B
- Authority
- CN
- China
- Prior art keywords
- image data
- data
- hot
- cold
- area
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/41—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
- G06F16/9554—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL] by using bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a method for realizing cold and hot separation storage of data based on a service scene called by image data, belongs to the technical field of cold and hot separation storage of image data, and solves the problems of low accuracy of cold and hot data prejudgment, long calling response time and the like in the prior art. The invention processes the image data in the temporary preheating area, the hot data area or/and the cold data area by carrying out cold-hot separation processing on the image data in the temporary preheating area, the hot data area or/and the cold data area based on whether the acquired image data is submitted for the first time or/and when the next business to be processed is obtained based on prediction, and stores the separated data in the hot data area and the cold data area respectively. The invention is used for cold-hot separation storage of images.
Description
Technical Field
A method for realizing cold-hot separation storage of data based on image data calling service scene is used for cold-hot separation storage of images and belongs to the technical field of cold-hot separation storage of image data.
Background
With the development of the internet, big data and electronic commerce, internet finance is rapidly developed, more and more financial business processes realize internetworking, online business processes generate a large amount of image data, and how to rapidly, efficiently and reasonably store, retrieve and manage the image data is particularly important. The existing image cold-hot separation method basically carries out cold-hot judgment on data through calling times, and then carries out layered storage on the cold-hot data. Because of the large volume of image data, cold data is often stored in some storage media that are less efficient to query. In some business scenarios calling for video, the access frequency of video material is low but the storage efficiency is required to be long. If the user accesses an account, the identity card photo uploaded by the user on line is stored after the user transacts the business, and the user may retrieve the identity card photo after long-term online authentication, and the image is defined as cold data, which takes a long time to retrieve.
The existing cold and hot data prejudging method mainly has the following problems:
1. the cold and hot data are pre-judged through the retrieval times, the cold data can be suddenly changed into the hot data in a business handling scene, and the hot data can not be invoked for a long time after the business handling is completed, so that the accuracy of the retrieval times for pre-judging the cold and hot data is low.
2. And the cold data is stored in a layered manner, and when the business is transacted, the data is preheated and then is called, so that the response time is long.
3. The query efficiency of cold data is generally slow and the response time is long.
Disclosure of Invention
Aiming at the problems of the researches, the invention aims to provide a method for realizing cold and hot separation storage of data based on calling service scenes of image data, which solves the problems of low accuracy of cold and hot data prejudgment, long calling response time and the like in the prior art.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a method for realizing cold and hot separation storage of data based on calling business scenes of image data comprises the following steps:
step 1, acquiring user behavior data in real time according to a point burying technology in an application scene, acquiring image data submitted based on the user behavior data, judging whether the acquired image data is submitted for the first time, if not, turning to step 2, if so, storing in a hot data area, and turning to step 2, wherein the image data is unstructured data, including pictures, videos and PDF;
step 2, predicting the next business to be transacted of the user based on the current user behavior data, the historical user behavior data and the trained user behavior prediction model;
step 3, if the image data corresponding to the next business to be processed is stored in the cold data area, the image data is called to the temporary preheating area, and then the step 4 is carried out, if the image data is stored in the hot data area, the processing is not carried out;
and 4, when the next business to be transacted is processed, the image data in the temporary preheating area is called, the image data in the temporary preheating area is moved to the hot data area for storage, and if the image data is not called, the image data is moved back to the cold data area for storage.
Further, the method also comprises cold and hot updating of the image data, namely detecting the image data stored in the hot data area every day, judging whether the image data exceeds a given time threshold and is not called, if yes, judging the image data as cold image data, moving the image data to the cold data area for storage, and if not, still storing the image data as hot image data in the hot data area, wherein the given time threshold is 7-30 days.
Further, the specific steps of the step 1 are as follows:
step 1.1, in an application scene of calling image data, when a user logs in or browses an entry page, generating user operation behavior data, namely user behavior data, in real time based on a buried point technology, recording the user behavior data in real time after the generation, and acquiring the image data submitted based on the user behavior data;
step 1.2, judging whether the acquired image data is submitted for the first time, namely judging whether the image data submitted currently is in a temporary preheating area, a hot data area or a cold data area according to a label comparison method, if not, namely, in the temporary preheating area, the hot data area or the cold data area, turning to step 2, if so, namely, not in the temporary preheating area, the hot data area and the cold data area, storing the image data in the hot data area, and turning to step 2; the method for comparing the labels is to compare the labels of the image data with the labels of the image data stored in the temporary preheating area, the hot data area and the cold data area.
Further, in the step 2, the specific steps of obtaining the trained user behavior prediction model are as follows:
step 2-1, acquiring historical user behavior data, namely acquiring historical geographic position information of a user, historical current business handling and historical next business to be handled and business flow sequence rules, wherein the business flow sequence rules refer to the sequence of each business in the business handling process and the sequence of calling image data in each business;
step 2-2, performing one-hot coding conversion treatment on historical geographic position information and business process sequence rules, and mapping the treated historical geographic position information and business process sequence rules into a vector form, wherein the missing value is filled with a median or an average value during mapping; mapping the URL of the history access page into a vector space, and forming feature vectors by all mapped results, wherein the URL of the history access page refers to the current business and the next business to be processed in the history;
and 2-3, training a user behavior prediction model based on the feature vector to obtain a trained user behavior model, wherein the user behavior prediction model is an LSTM long-short-term memory neural network model.
Further, in the step 2-1, the service flow sequence rule is included in the trust scene, and the service sequence is that the user registers to real name authentication, applies for trust and pays; the method comprises the steps of registering a user to real-name authentication, wherein the step of calling the sequence of uploading identity card image data, OCR (optical character recognition) of the identity card image data, face comparison and lip language recognition; calling the sequence of identity image data check and credit investigation image data check in the application credit investigation; and calling the sequence of examination of the contract image data during the paying.
Further, the specific steps of the step 2 are as follows: and inputting the current geographic position information, the current business handling, the business process sequence rule and the historical user behavior data of the user into a trained user behavior model, and predicting the next business to be handled by the user.
Further, the temporary preheating area in the step 3 is one or more of NAS server, memory storage medium, cache storage medium, or buffer storage medium.
Further, the cold data area is one or more of a large data platform HDFS, HBASE storage medium, or FileNet storage medium, and the hot data area is a high performance storage medium including one or more of NAS or SSD storage medium.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention predicts the user behavior in real time, prepares the needed image data to the temporary preheating area in advance, and fundamentally solves the problems of overlong service response time, too slow processing efficiency and poor user experience.
2. According to the invention, cold and hot data are judged according to the time threshold value, the cold and hot state is updated in time, and the problem of low accuracy of the cold and hot data is solved.
3. The invention solves the dependence of online business on using high-capacity high-performance storage by predicting the user behavior in real time, and only needs to use a small amount of high-performance storage by predicting the user behavior in real time; the system construction cost is greatly saved. Such as: in order to meet the response requirement, the 100TB image data needs to be stored by a high-performance storage medium, and only 1TB high-performance storage medium and 99TB common storage medium are needed to be stored.
4. The invention can save half of the time compared with the prior art when the invention calls the cold data under the condition that the next business is required to be handled successfully.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a flow chart of a trained user behavior prediction model according to the present invention.
Detailed Description
The foregoing of the invention is further described in detail below in conjunction with the detailed description of the embodiments, in order to facilitate understanding and practice of various aspects of the invention by those of ordinary skill in the art. It should not be understood that the scope of the above subject matter of the present invention is limited to the following examples only. Various substitutions and alterations are also possible, without departing from the spirit of the invention, and are intended to be within the scope of the invention.
A method for realizing cold and hot separation storage of data based on calling business scenes of image data comprises the following steps:
step 1, acquiring user behavior data in real time according to a point burying technology in an application scene, acquiring image data submitted based on the user behavior data, judging whether the acquired image data is submitted for the first time, if not, turning to step 2, if so, storing in a hot data area, and turning to step 2, wherein the image data is unstructured data, including pictures, videos and PDF;
the method comprises the following specific steps:
step 1.1, in an application scene of calling image data, when a user logs in or browses an entry page, generating user operation behavior data, namely user behavior data, in real time based on a buried point technology, recording the user behavior data in real time after the generation, and acquiring the image data submitted based on the user behavior data;
step 1.2, judging whether the acquired image data is submitted for the first time, namely judging whether the image data submitted currently is in a temporary preheating area, a hot data area or a cold data area according to a label comparison method, if not, namely, in the temporary preheating area, the hot data area or the cold data area, turning to step 2, if so, namely, not in the temporary preheating area, the hot data area and the cold data area, storing the image data in the hot data area, and turning to step 2; the method for comparing the labels is to compare the labels of the image data with the labels of the image data stored in the temporary preheating area, the hot data area and the cold data area.
Step 2, predicting the next business to be transacted of the user based on the current user behavior data, the historical user behavior data and the trained user behavior prediction model; the current geographic position information, the current business, the business process sequence rule and the historical user behavior data of the user are input into a trained user behavior model, and the next business to be processed of the user is predicted.
The specific steps for obtaining the trained user behavior prediction model are as follows:
step 2-1, acquiring historical user behavior data, namely acquiring historical geographic position information of a user, historical current business handling and historical next business to be handled and business flow sequence rules, wherein the business flow sequence rules refer to the sequence of each business in the business handling process and the sequence of calling image data in each business;
the service flow sequence rule comprises that in a credit granting scene, the service sequence is that a user registers to real name authentication, applies credit granting and paying in sequence; the method comprises the steps of registering a user to real-name authentication, wherein the step of calling the sequence of uploading identity card image data, OCR (optical character recognition) of the identity card image data, face comparison and lip language recognition; calling the sequence of identity image data check and credit investigation image data check in the application credit investigation; and calling the sequence of examination of the contract image data during the paying.
Step 2-2, performing one-hot coding conversion treatment on historical geographic position information and business process sequence rules, and mapping the treated historical geographic position information and business process sequence rules into a vector form, wherein the missing value is filled with a median or an average value during mapping; mapping the URL of the history access page into a vector space, and forming feature vectors by all mapped results, wherein the URL of the history access page refers to the current business and the next business to be processed in the history;
and 2-3, training a user behavior prediction model based on the feature vector to obtain a trained user behavior model, wherein the user behavior prediction model is an LSTM long-short-term memory neural network model.
Step 3, if the image data corresponding to the next business to be processed is stored in the cold data area, the image data is called to the temporary preheating area, and then the step 4 is carried out, if the image data is stored in the hot data area, the processing is not carried out: the temporary preheating area is one or more of a NAS server, a memory storage medium, a cache storage medium or a buffer zone storage medium.
And 4, when the next business to be transacted is processed, the image data in the temporary preheating area is called, the image data in the temporary preheating area is moved to the hot data area for storage, and if the image data is not called, the image data is moved back to the cold data area for storage.
Further, the method also comprises cold and hot updating of the image data, namely detecting the image data stored in the hot data area every day, judging whether the image data exceeds a given time threshold and is not called, if yes, judging the image data as cold image data, moving the image data to the cold data area for storage, and if not, still storing the image data as hot image data in the hot data area, wherein the given time threshold is 7-30 days.
Further, the cold data area is one or more of a large data platform HDFS, HBASE storage medium, or FileNet storage medium, and the hot data area is a high performance storage medium including one or more of NAS or SSD storage medium.
Examples
As shown in FIG. 1, a method for realizing cold and hot separation storage of data based on calling business scenes of image data comprises the following steps:
A. initializing and storing image data:
firstly, when a user logs in, browses pages and transacts business, the user can access operation behavior data in real time, and a series of operation behavior data, namely user behavior data, when the user browses pages is recorded in real time through a real-time data acquisition system.
Specifically, a real-time data acquisition system built by a front page embedded point generation log+kafka cluster (a high throughput distributed publishing and subscribing message system, which can process all action stream data of a consumer in a website) records current page Token information (current session ID) corresponding to user behavior in real time according to the sequence of browsing pages and operating services of the user, and the user browses page information and service information corresponding to the current page.
And judging whether the image data related to the operation flow of the user exists in the storage medium or not in real time.
Specifically, the real-time data acquisition system compares the image data labels (the labels are selected when the user uploads the image data labels and the content are consistent, otherwise, the current step is returned) and the image data labels which need to be uploaded are compared according to the label comparison method, and whether the image data which need to be uploaded currently exist in the current storage medium is determined.
If the image data currently required to be uploaded exists in the current storage medium, the next step is directly carried out. Otherwise, the currently uploaded image data is initialized and stored in the hot data area in real time.
If the next business to be transacted needs to be predicted, sequentially executing the steps B and C, wherein the steps are as follows:
B. predicting user access behaviors, including determining a business process sequence rule and an LSTM long and short time memory neural network model:
firstly, determining a sequence rule of a business process according to a business handling process on a banking line, wherein the sequence rule is determined before the business is on line.
Specifically, for example, the next step after user registration is a real-name authentication process, and real-name authentication needs to upload the image data of the user identification card, and perform a series of sequential operations of invoking the image data of the identification card for authentication, such as OCR recognition, face comparison, networking verification, and the like. For example, the next step after authentication of the real name is a credit process, and a credit limit is obtained. For example, in the credit granting process, risk control operation is performed, and whether the current user is trusted and the credit limit is determined by checking the credit investigation image data.
And then, predicting the next business to be transacted of the current user in real time according to the trained LSTM long and short-term memory neural network model, and determining whether the processing of the image data is involved.
Specifically, as shown in fig. 2, the LSTM long-term memory neural network model includes:
b_1, acquiring historical access behavior data of the user, namely acquiring historical geographic position information, historical currently transacted business and historical next business to be transacted and business flow sequence rules.
B_2, preprocessing data, namely performing one-hot coding conversion processing on historical geographic position information and business flow sequence rules, and mapping the processed information into a vector form; wherein missing values in the map are filled in with a median or mean, or otherwise processed; mapping the URL of the historical access page into a vector space, and forming feature vectors by all mapping results; .
B_3, training a model, namely training the preprocessed feature vector, wherein the target loss function rule is the minimum network error of the current service to be processed and the next service to be processed in the history during model training; and updating the LSTM long-term memory neural network model parameters along the gradient descent direction of the target loss function until the network error is smaller than a given experience value, thereby obtaining a trained LSTM long-term memory neural network model.
B_4, based on the current user behavior data and the historical user behavior data, the trained LSTM long-term memory neural network model is combined, the service which the user needs to transact next time is predicted, the service is output as a model, and the probability of the result and the corresponding service are expressed as (p_i, flow_i), such as (p_1, flow_1); (p_2, flow_2).
C. Preheating image data and updating data state:
and preheating the required image data in advance according to the prediction result, and updating the data state according to the result.
Specifically, according to the real-time behavior data of the user at the buried point, predicting the next service to be processed by the user, processing whether the service needs to be called, if the predicted result is needed, pre-heating the service corresponding to the maximum probability max (p_i) of the user in advance, and storing the related image data in a temporary pre-heating area in a copy form in high-efficiency storage media such as a NAS server, a memory, a cache and the like as an intermediate processing step.
If the predicted result is that the image data needs to be called, and the predicted result is that the image data also needs to be called, the target preheating data can be directly called from the temporary preheating area, and meanwhile, the preheating data of the user is transferred to the hot data area and changed into a hot data state. If the predicted result is that the image data needs to be called, but the image data does not need to be called actually, deleting the duplicate image data stored in the temporary preheating area, and keeping the original state of the original image data stored in the cold data area unchanged.
D. Updating the cold and hot states of the image data:
the full data cold and hot status is detected and updated once at daily timing (i.e., during execution A, B, C, or at other times).
Specifically, if the hot data exceeds the set time threshold and is not called, the hot data state is updated to be a cold data state, and the hot data storage is transferred to a cold data area, at this time, whether the hot data is converted to the cold data is determined according to the service experience time threshold, and the time interval range of the cold and hot judgment threshold of the image data is usually set to be 7-30 days. In addition, the original state is kept unchanged.
Claims (5)
1. A method for realizing cold and hot separation storage of data based on calling business scenes of image data is characterized by comprising the following steps:
step 1, acquiring user behavior data in real time according to a point burying technology in an application scene, acquiring image data submitted based on the user behavior data, judging whether the acquired image data is submitted for the first time, if not, turning to step 2, if so, storing in a hot data area, and turning to step 2, wherein the image data is unstructured data, including pictures, videos and PDF;
step 1.1, in an application scene of calling image data, when a user logs in or browses an entry page, generating user operation behavior data, namely user behavior data, in real time based on a buried point technology, recording the user behavior data in real time after the generation, and acquiring the image data submitted based on the user behavior data;
step 1.2, judging whether the acquired image data is submitted for the first time, namely judging whether the image data submitted currently is in a temporary preheating area, a hot data area or a cold data area according to a label comparison method, if not, namely, in the temporary preheating area, the hot data area or the cold data area, turning to step 2, if so, namely, not in the temporary preheating area, the hot data area and the cold data area, storing the image data in the hot data area, and turning to step 2; the label comparison method is to compare the label of the image data with the labels of the image data stored in the temporary preheating area, the hot data area and the cold data area;
step 2, predicting the next business to be transacted of the user based on the current user behavior data, the historical user behavior data and the trained user behavior prediction model;
the specific steps for obtaining the trained user behavior prediction model are as follows:
step 2-1, acquiring historical user behavior data, namely acquiring historical geographic position information of a user, historical current business handling and historical next business to be handled and business flow sequence rules, wherein the business flow sequence rules refer to the sequence of each business in the business handling process and the sequence of calling image data in each business;
step 2-2, performing one-hot coding conversion treatment on historical geographic position information and business process sequence rules, and mapping the treated historical geographic position information and business process sequence rules into a vector form, wherein the missing value is filled with a median or an average value during mapping; mapping the URL of the history access page into a vector space, and forming feature vectors by all mapped results, wherein the URL of the history access page refers to the current business and the next business to be processed in the history;
step 2-3, training a user behavior prediction model based on the feature vector to obtain a trained user behavior model, wherein the user behavior prediction model is an LSTM long-short-term memory neural network model;
step 3, if the image data corresponding to the next business to be processed is stored in the cold data area, the image data is called to the temporary preheating area, and then the step 4 is carried out, if the image data is stored in the hot data area, the processing is not carried out;
step 4, when the next business to be transacted is processed, the image data in the temporary preheating area is called, the image data in the temporary preheating area is moved to the hot data area for storage, and if the image data is not called, the image data is moved back to the cold data area for storage;
the method also comprises the steps of detecting the image data stored in the hot data area every day, judging whether the image data exceeds a given time threshold and is not called, if yes, judging the image data as cold image data, moving the image data to the cold data area for storage, and if not, still storing the image data as hot image data in the hot data area, wherein the given time threshold is 7-30 days.
2. The method for realizing cold and hot separation and storage of data based on image data calling service scene according to claim 1, wherein in step 2-1, the service flow sequencing rule is included in the trust scene, and the service sequencing is that the user is registered to real name authentication, applies trust and pays; the method comprises the steps of registering a user to real-name authentication, wherein the step of calling the sequence of uploading identity card image data, OCR (optical character recognition) of the identity card image data, face comparison and lip language recognition; calling the sequence of identity image data check and credit investigation image data check in the application credit investigation; and calling the sequence of examination of the contract image data during the paying.
3. The method for realizing cold and hot separation storage of data based on the image data calling service scene according to claim 2, wherein the specific steps of the step 2 are as follows: and inputting the current geographic position information, the current business handling, the business process sequence rule and the historical user behavior data of the user into a trained user behavior model, and predicting the next business to be handled by the user.
4. The method for realizing cold and hot separation storage of data based on image data calling service scene according to claim 1, wherein the temporary preheating area in the step 3 is one or more of NAS server, memory storage medium, cache storage medium or buffer storage medium.
5. The method for realizing cold and hot separation storage of data based on image data calling service scenes according to claim 1 or 4, wherein the cold data area is one or more of a large data platform HDFS, HBASE storage medium or FileNet storage medium, and the hot data area is a high-performance storage medium including one or more of NAS or SSD storage medium.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010004175.5A CN111209414B (en) | 2020-01-03 | 2020-01-03 | Method for realizing cold-hot separation storage of data based on image data calling business scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010004175.5A CN111209414B (en) | 2020-01-03 | 2020-01-03 | Method for realizing cold-hot separation storage of data based on image data calling business scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111209414A CN111209414A (en) | 2020-05-29 |
| CN111209414B true CN111209414B (en) | 2023-08-11 |
Family
ID=70787436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010004175.5A Active CN111209414B (en) | 2020-01-03 | 2020-01-03 | Method for realizing cold-hot separation storage of data based on image data calling business scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111209414B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014209234A1 (en) * | 2013-06-26 | 2014-12-31 | Agency For Science, Technology And Research | Method and apparatus for hot data region optimized dynamic management |
| CN104793901A (en) * | 2015-04-09 | 2015-07-22 | 北京鲸鲨软件科技有限公司 | Storage device and storage method |
| CN107943825A (en) * | 2017-10-19 | 2018-04-20 | 阿里巴巴集团控股有限公司 | Data processing method, device and the electronic equipment of page access |
| CN108563730A (en) * | 2018-04-04 | 2018-09-21 | 北京蓝杞数据科技有限公司天津分公司 | A kind of cold and hot data automatic switching method, device, electronic equipment and storage medium |
| CN109344092A (en) * | 2018-09-11 | 2019-02-15 | 天津易华录信息技术有限公司 | A kind of method and system improving cold storing data reading speed |
| CN109447320A (en) * | 2018-09-27 | 2019-03-08 | 中国联合网络通信集团有限公司 | The recognition methods of customer service, apparatus and system |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107179878B (en) * | 2016-03-11 | 2021-03-19 | 伊姆西Ip控股有限责任公司 | Data storage method and device based on application optimization |
| US10802733B2 (en) * | 2018-04-27 | 2020-10-13 | Western Digital Technologies, Inc. | Methods and apparatus for configuring storage tiers within SSDs |
-
2020
- 2020-01-03 CN CN202010004175.5A patent/CN111209414B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014209234A1 (en) * | 2013-06-26 | 2014-12-31 | Agency For Science, Technology And Research | Method and apparatus for hot data region optimized dynamic management |
| CN104793901A (en) * | 2015-04-09 | 2015-07-22 | 北京鲸鲨软件科技有限公司 | Storage device and storage method |
| CN107943825A (en) * | 2017-10-19 | 2018-04-20 | 阿里巴巴集团控股有限公司 | Data processing method, device and the electronic equipment of page access |
| CN108563730A (en) * | 2018-04-04 | 2018-09-21 | 北京蓝杞数据科技有限公司天津分公司 | A kind of cold and hot data automatic switching method, device, electronic equipment and storage medium |
| CN109344092A (en) * | 2018-09-11 | 2019-02-15 | 天津易华录信息技术有限公司 | A kind of method and system improving cold storing data reading speed |
| CN109447320A (en) * | 2018-09-27 | 2019-03-08 | 中国联合网络通信集团有限公司 | The recognition methods of customer service, apparatus and system |
Non-Patent Citations (1)
| Title |
|---|
| 适应冷热数据存储的多编码架构的设计与实证;魏学才 等;《计算机应用与软件》(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111209414A (en) | 2020-05-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190392258A1 (en) | Method and apparatus for generating information | |
| US20240020270A1 (en) | Efficient similarity detection | |
| CN113901320A (en) | Scene service recommendation method, device, equipment and storage medium | |
| CN113378710A (en) | Layout analysis method and device for image file, computer equipment and storage medium | |
| WO2023178930A1 (en) | Image recognition method and apparatus, training method and apparatus, system, and storage medium | |
| WO2021026034A1 (en) | Artificial intelligence job recommendation neural network machine learning training based on embedding technologies and actual and synthetic job transition latent information | |
| CN110162689B (en) | Information pushing method, device, computer equipment and storage medium | |
| CN111209414B (en) | Method for realizing cold-hot separation storage of data based on image data calling business scene | |
| CN117009873A (en) | Method for generating payment risk identification model, and method and device for payment risk identification | |
| CN110688515A (en) | Text image semantic conversion method and device, computing equipment and storage medium | |
| CN113255824B (en) | Method and apparatus for training classification model and data classification | |
| CN111091198B (en) | Data processing method and device | |
| CN112633285A (en) | Domain adaptation method, domain adaptation device, electronic equipment and storage medium | |
| CN113569094A (en) | Video recommendation method and device, electronic equipment and storage medium | |
| US11804057B1 (en) | Computer systems and computer-implemented methods utilizing a digital asset generation platform for classifying data structures | |
| CN112579865B (en) | Price adjustment method and device for search keywords, storage medium and electronic equipment | |
| CN117216803B (en) | Intelligent finance-oriented user information protection method and system | |
| CN113434676B (en) | Text relation extraction model training, text relation extraction method, device and equipment | |
| US20230419098A1 (en) | Utilizing selective transformation and replacement with high-dimensionality projection layers to implement neural networks in tabular data environments | |
| CN113011646B (en) | Data processing method, device and readable storage medium | |
| CN112766407B (en) | Image recognition method, device and storage medium | |
| Yang et al. | Copyright storage method of dance short video based on blockchain | |
| US20240171400A1 (en) | Feedback mining with domain-specific modeling and token-based transactions | |
| Manikandan et al. | A Document Imaging Technique for Implementing Electronic Loan Approval Process | |
| Pan et al. | Convolutional Self-attention Guided Graph Neural Network for Few-Shot Action Recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |