CN104035956A - Time-series data storage method based on distributive column storage - Google Patents
Time-series data storage method based on distributive column storage Download PDFInfo
- Publication number
- CN104035956A CN104035956A CN201410143604.1A CN201410143604A CN104035956A CN 104035956 A CN104035956 A CN 104035956A CN 201410143604 A CN201410143604 A CN 201410143604A CN 104035956 A CN104035956 A CN 104035956A
- Authority
- CN
- China
- Prior art keywords
- value
- title
- storage
- measuring point
- data
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/221—Column-oriented storage; Management thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a time-series data storage method based on distributive column storage, and belongs to the technical field of database. The method is characterized in that a spare table composed of row keys and rows is used to store data stored in measuring points, wherein each of the row keys contains as much retrieval information; data that are dispersedly acquired at some time slots are merged in the record at each row; a name mapping table is created to store name information. With the adoption of the time-series data storage method based on the distributive column storage, various advantages of the distributive column storage technology can be fully utilized, thus the efficiency of storing and retrieving time-series data in the distributive column storage can be ensured, and the safety of the data storage is improved due to the distributive storage manner.
Description
Technical field
The invention belongs to database technical field, more precisely, the present invention relates to a kind of time series data storage means based on distributed column storage.
Background technology
Along with deepening continuously that industrialization, informationization " two change fusion " are built, industry-by-industry is all faced with the explosive growth of data, and data processing pressure is increasing.From the visual angle of national grid, along with the high speed development of national grid, the extensive and deep application of informationization technology in national grid miscellaneous service, the popularization of intelligent grid and universal.There is the growth of explosion type in the data of each operation system of electric system, wherein seasonal effect in time series growth rate is particularly evident.No matter traditional time series databases, in the data volume of processing or in response speed, more and more cannot meet existing business needs, this point in the power information of each net province company gathers, embodiment particularly evident.Meanwhile, along with the fast development of domestic and international Internet service, development and the widespread use of large data and cloud computing, distributed row memory technology becomes the primary importance technology of processing in mass data, and is much enlightened with us.
In electrical network business, time series data, as a kind of important non-structured data type, has extensive and deep application in several scenes such as marketing, scheduling, fortune prison, productions.In tradition application, time series data is often established as structurized data model by abstract, and deposits relevant database in.And the line pattern of relevant database, when the time series data of storage Different sampling period, can cause the significant wastage of storage space.On the other hand, along with the increase of mass data, traditional relevant database often cannot be tackled, and makes search efficiency sharp fall, thereby cannot meet the demand of service application scene.There are in addition a lot of producers to adopt the form of four-tuple, use the data structure of B+ tree, directly deposit file system in.This mode has saving storage space, and inquiry is direct, the feature that efficiency is higher.But there is later stage expansion difficulty simultaneously, do not possess the database common functions such as polymerization calculating, and in processing magnanimity time series data, exist a lot of not enough.In general, no matter these two kinds of modes are in theory or practical application, all exist significant limitation, restricted further developing for the service application of time series data.Adopt distributed column storage database, time series data is carried out to modeling, use this memory model at distributed column, to store the storage of enterprising line time sequence data, can solve to a great extent these restrictions and break through these limitations.
Along with the propelling of emerging development of Mobile Internet technology, the traditional forms of enterprises is also in constantly development and progress, and common is exactly a bit that data, as a kind of important resource and wealth, are more and more subject to the attention of each side enterprise.And along with monitoring and control object constantly complicated in each enterprise, the explosive growth of data volume, different application is deepening continuously to the use of various data and understanding that can value, to mass data efficient storage how, how to access easily, and the efficiency of access and the requirement of response speed are also being improved constantly.So the efficient storage to data, quick-searching, and various analyzing and processing ability has proposed requirements at the higher level.The present invention just for reach efficient storage and fast retrieval time sequence data and develop realization.
Summary of the invention
The object of the invention is: a kind of time series data storage means based on distributed column storage is provided, makes time series data can adapt to the storage mode of distributed column storage, bring into play the advantage that distributed column is stored.The method can guarantee efficient storage and the quick-searching of time series data in distributed column storage.
Specifically, the present invention adopts following technical scheme to realize, and comprises the following steps:
1) use measuring point data table storage measuring point record data, the sparse table that described measuring point data table is comprised of line unit and row forms, wherein, line unit is comprised of unique point, timestamp, three parts of label, unique point is for identifying the title of measuring point, timestamp is used for identifying value reference time, label is for the attribute of representation feature point, number of labels in each line unit is one or more, each label is comprised of one or more key-value pairs, key in key-value pair is for identifying the attribute of measuring point, and the value in key-value pair is for identifying the property value of measuring point; Row are comprised of son row, and every height row are for storing the record value of the side-play amount of the reference time of recording corresponding to line unit, and the quantity of son row is determined by the maximum offset setting in advance;
Use the name map table storage title of unique point and the title of label key-value pair, at name map table, use the corresponding corresponding title of unique point of name map value and the title of label key-value pair of regular length;
The information of using measuring point information table to store each measuring point, storage mode adopts the Distributed Storage mode of standard;
2), during data writing, in distributed column storage database environment, first according to each measuring point record data, obtain the title of corresponding unique point and the title of label key-value pair, and by the name storage of the title of unique point and label key-value pair in name map table;
Then, according to the name map value of the name map value of the unique point title of each measuring point record data, reference time, label key-value pair title and analog value, generate corresponding line unit value, and according to the side-play amount of each measuring point record data, the line unit value of each measuring point record data and record value are stored in respectively in measuring point data table to corresponding line unit with in son row relative with this line unit respective offsets amount.
Technique scheme is further characterized in that, the storage of described row employing variable length.
Technique scheme is further characterized in that, described name map table adopts two-way mapping, can, by the title of unique point or the title of label key-value pair retrieval respective name mapping value, also can retrieve the title of corresponding unique point or the title of label key-value pair by name map value.
Technique scheme is further characterized in that, described step 2) in, when a certain line item in measuring point data table is stored completely, then open new a line.
Beneficial effect of the present invention is as follows: the line unit in measuring point data table of the present invention has comprised retrieving information as much as possible, each line item combines the data of the distributed collection of certain time period, reduce the number of whole form line unit, thereby improved the speed of retrieval.And the mode essence that the present invention stores data according to the extension of time is a kind of stateless storage scheme.Stateless storage refers to that two records that deposit in continuously do not understand mutually, is incoherent.If two records that deposit in are continuously correlated with, after last record made mistakes, probably cause a rear record to be made mistakes so.Therefore, the stateless storage scheme of the inventive method can also improve system survivability.The present invention is simultaneously provided with name map table store name information, thereby when storing concrete time series data, can reduce the use of storage space.The beneficial effect of the inventive method is in sum, makes full use of every advantage of distributed column memory technology, guarantees storage and the effectiveness of retrieval of time series data in distributed column storage, and utilizes distributed storage to strengthen the security of data storage.
Accompanying drawing explanation
Fig. 1 is measuring point data list structure schematic diagram in the present invention.
Fig. 2 is row storage organization schematic diagram in the present invention.
Fig. 3 is name map list structure schematic diagram in the present invention.
Embodiment
With reference to the accompanying drawings and in conjunction with example the present invention is described in further detail.
An example of the present invention, the tables of data of whole time series databases is divided into measuring point information table METRIC_INFO, name map table METRICTAG_UID and measuring point data table METRIC_DATA tri-parts form.The information of the main storage of collected measuring point of measuring point information table, name map table is mainly stored title and the corresponding name map value of measuring point or label, and measuring point data table is stored concrete time series data.
As shown in Figure 1, the sparse table that measuring point data table METRIC_DATA is comprised of line unit and row forms, and row can be realized by one or more son row.As shown in Figure 3, line unit is comprised of unique point, timestamp, label the structure of line unit.In the present embodiment, feature is called the roll and is used 3 byte representations, timestamp value 4 byte representations, label is used for the attribute of representation feature point, as shown, this unique point comes from which information gathering point or other attributive character, a label is comprised of one or more key-value pairs, 6 bytes for each key-value pair, and each line unit can be carried out the attribute of Expressive Features point simultaneously by a plurality of labels.
Son row storage be the side-play amount corresponding to line unit reference time time record value, the quantity of son row is determined by the maximum offset setting in advance.Maximum offset in shown in Fig. 1 is 3600 seconds, can be according to real needs adjustment in practical application.As shown in Figure 2, the memory model of row adopts 2 bytes statements, and wherein first 12 are used for representing reference time, the 13rd data type (integer or floating type) for the storage of type code bit representation, last 3 length that are used for representing the record value stored.Row memory model adopts variable length storage rather than fixed-length value for the storage of record value, can make full use of storage space, and this storage mode is consistent with the object of the sparse feature of measuring point memory model.
Visible, measuring point data table METRIC_DATA has realized measuring point storage, simultaneously by carrying out the continuous data point of sparse storage measuring point (time series data) corresponding to the line unit side-play amount of reference time (son row), can significantly reduce the line number of storage, reach the effect of compression storage information.
The information of using measuring point information table METRIC_INFO to store each measuring point, storage mode adopts the Distributed Storage mode of standard.The information of measuring point comprises type, unit, address, description, display Name, remarks of measuring point etc.
Use the key-value pair title of name map table METRICTAG_UID storage unique point title and label.The corresponding corresponding title of unique point of name map value (UID) and the title of label key-value pair of in name map table METRICTAG_UID, using regular length (as 3 bytes), the concrete corresponding relation of UID and title can adopt the mode of two-way mapping.As shown in Figure 3, name map table METRICTAG_UID is comprised of line unit, row name and tri-parts of row id.Name map table METRICTAG_UID line unit be used for storing line unit in measuring point data table METRIC_DATA (unique point, label. key, label. value) corresponding UID and title, in row name and row id, store corresponding title or UID simultaneously.The UID of regular length, makes, when storing concrete time series data, to have reduced the use of storage space, has accelerated retrieval rate simultaneously.By the mode of two-way mapping, can also by title, retrieve corresponding UID at a high speed, also can retrieve at a high speed title by UID.
In the time series data memory model in distributed column storage, the row of row can not unconfined continuation expansion.Therefore, when data point is set up for the first time in this model, first set up a new row, after storage one given data, (after storage is full) opens another row again.Owing to there being concurrent situation about writing in distributed environment, measuring point data table has guaranteed that from UID name map different line units (being comprised of unique point, timestamp, label) do not exist conflict, and for same line unit, regular time side-play amount is set and decides line feed, different clients, for same line unit different time skew storage respectively, has been avoided the write conflict under distributed environment.In concrete application deployment, can by configuration file, specify according to demand the time offset of line feed.Under distributed environment each independently this model can judge according to the time row at current data place.
Concrete ablation process is: during data writing, in distributed column storage database environment, first according to each measuring point record data, obtain the title of corresponding unique point and the title of label key-value pair, and by the name storage of the title of unique point and label key-value pair in name map table METRICTAG_UID.Then, according to the name map value of the title of the name map value of the title of the unique point of each measuring point record data, reference time, label key-value pair and analog value, generate corresponding line unit value, and according to the side-play amount of each measuring point record data, the line unit value of each measuring point record data and record value are stored in respectively in measuring point data table METRIC_DATA to corresponding line unit with in son row relative with this line unit respective offsets amount.Side-play amount is according in measuring point data table METRIC_DATA, this records corresponding calculating reference time.For a certain line item in measuring point data table METRIC_DATA, if the side-play amount in this row storage full (sometime after section) or new record data has surpassed the maximum offset of this row, the new line item of opening in measuring point data table METRIC_DATA.Repeat above process, complete the storage of all measuring point information datas.
The query script of data is as follows: when the phase of history data of certain measuring point of inquiry, according to the name query name map table METRICTAG_UID of the unique point of inquiry, obtain the UID of unique point, according to the time range of appointment, jointly determine that this unique point is mapped to the line unit in measuring point data table METRIC_DATA, is then determined the record value of the unique point of returning to inquiry by the temporal information of line unit and inquiry simultaneously.
With concrete example, describe below.Be provided with two measuring point recorded informations in certain measuring point time series to be stored, be respectively:
Measuring point recorded information 1, its content is: Crd.tsd.test 1,386,304,551,191 5400 Host=hostA Range=Suzhou.Wherein, Crd.tsd.test is that feature is called the roll, and Host=hostA, Range=Suzhou are label.
Measuring point recorded information 2, its content is: Crd.tsd.test 1,386,304,550,002 8500 Host=hostA Range=Suzhou.Wherein, Crd.tsd.test is that feature is called the roll, and Host=hostA, Range=Suzhou are label.
According to measuring point recorded information 1, generating corresponding line unit value is stored in the line unit of measuring point data table METRIC_DATA, according to its temporal information, be 1386304551191 to draw reference time (1386304550000) and side-play amount (1191), record value (5400) is stored in the son row (1191) that this line unit is corresponding, the storage of the value of completing, its result as shown in Figure 1.
Measuring point recorded information 2 generates corresponding line unit value and is stored in the line unit of measuring point data table METRIC_DATA, according to its temporal information, be 1386304550002 to draw reference time (1386304550000) and side-play amount (0002), record value (8500) is stored in the son row (2) that this line unit is corresponding to the storage of the value of completing.Measuring point recorded information 1 and measuring point recorded information 2 are stored in same line unit, for sequence data storage time, can reduce the line number of storage, have reached the effect of compression storing data.
Although the present invention with preferred embodiment openly as above, embodiment is not of the present invention for limiting.Without departing from the spirit and scope of the invention, any equivalence of doing changes or retouching, belongs to equally the present invention's protection domain.Therefore should to take the application's the content that claim was defined be standard to protection scope of the present invention.
Claims (4)
1. the time series data storage means based on distributed column storage, is characterized in that, comprises the steps:
1) use measuring point data table storage measuring point record data, the sparse table that described measuring point data table is comprised of line unit and row forms, wherein, line unit is comprised of unique point, timestamp, three parts of label, unique point is for identifying the title of measuring point, timestamp is used for identifying value reference time, label is for the attribute of representation feature point, number of labels in each line unit is one or more, each label is comprised of one or more key-value pairs, key in key-value pair is for identifying the attribute of measuring point, and the value in key-value pair is for identifying the property value of measuring point; Row are comprised of son row, and every height row are for storing the record value of the side-play amount of the reference time of recording corresponding to line unit, and the quantity of son row is determined by the maximum offset setting in advance;
Use the name map table storage title of unique point and the title of label key-value pair, at name map table, use the corresponding corresponding title of unique point of name map value and the title of label key-value pair of regular length;
The information of using measuring point information table to store each measuring point, storage mode adopts the Distributed Storage mode of standard;
2), during data writing, in distributed column storage database environment, first according to each measuring point record data, obtain the title of corresponding unique point and the title of label key-value pair, and by the name storage of the title of unique point and label key-value pair in name map table;
Then, according to the name map value of the name map value of the unique point title of each measuring point record data, reference time, label key-value pair title and analog value, generate corresponding line unit value, and according to the side-play amount of each measuring point record data, the line unit value of each measuring point record data and record value are stored in respectively in measuring point data table to corresponding line unit with in son row relative with this line unit respective offsets amount.
2. the time series data storage means based on distributed column storage according to claim 1, is characterized in that, described row adopt variable length storage.
3. the time series data storage means based on distributed column storage according to claim 1, it is characterized in that, described name map table adopts two-way mapping, can, by the title of unique point or the title of label key-value pair retrieval respective name mapping value, also can retrieve the title of corresponding unique point or the title of label key-value pair by name map value.
4. the time series data storage means based on distributed column storage according to claim 1, is characterized in that described step 2) in, when a certain line item in measuring point data table is stored completely, then open new a line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410143604.1A CN104035956A (en) | 2014-04-11 | 2014-04-11 | Time-series data storage method based on distributive column storage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410143604.1A CN104035956A (en) | 2014-04-11 | 2014-04-11 | Time-series data storage method based on distributive column storage |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104035956A true CN104035956A (en) | 2014-09-10 |
Family
ID=51466726
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410143604.1A Pending CN104035956A (en) | 2014-04-11 | 2014-04-11 | Time-series data storage method based on distributive column storage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104035956A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331432A (en) * | 2014-10-22 | 2015-02-04 | 江苏瑞中数据股份有限公司 | Power grid mass time series data access method applicable to section access mode |
| CN106331075A (en) * | 2016-08-18 | 2017-01-11 | 华为技术有限公司 | Method for storing files, metadata server and manager |
| CN106383844A (en) * | 2016-08-31 | 2017-02-08 | 天津南大通用数据技术股份有限公司 | Storage method and device applied to special data |
| CN108062378A (en) * | 2017-12-12 | 2018-05-22 | 清华大学 | The Connection inquiring method and system of more time serieses under a kind of column storage |
| CN108595553A (en) * | 2018-04-10 | 2018-09-28 | 红云红河烟草(集团)有限责任公司 | A kind of industrial number based on relevant database adopts time series data compression storage and decompression querying method |
| CN108647243A (en) * | 2018-04-13 | 2018-10-12 | 中国神华能源股份有限公司 | Industrial big data storage method based on time series |
| CN109597588A (en) * | 2018-12-11 | 2019-04-09 | 浙江中智达科技有限公司 | A kind of date storage method, data restoration method and device |
| CN109960693A (en) * | 2018-11-22 | 2019-07-02 | 成都长城开发科技有限公司 | One kind being based on relevant database load curve storage method |
| CN110968585A (en) * | 2019-12-20 | 2020-04-07 | 深圳前海微众银行股份有限公司 | Method, device and equipment for storing orientation column and computer readable storage medium |
| CN113177090A (en) * | 2021-04-30 | 2021-07-27 | 中国邮政储蓄银行股份有限公司 | Data processing method and device |
| CN116204684A (en) * | 2023-02-01 | 2023-06-02 | 浙江正泰仪器仪表有限责任公司 | Storage method, device, equipment and medium of electric energy meter |
| CN116719822A (en) * | 2023-08-10 | 2023-09-08 | 深圳市连用科技有限公司 | Method and system for storing massive structured data |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216813A1 (en) * | 2008-02-27 | 2009-08-27 | Olivieri Ricardo N | Method and system for generating a transaction-bound sequence of records in a relational database table |
| CN101727465A (en) * | 2008-11-03 | 2010-06-09 | 中国移动通信集团公司 | Methods for establishing and inquiring index of distributed column storage database, device and system thereof |
| CN102043726A (en) * | 2010-12-29 | 2011-05-04 | 北京播思软件技术有限公司 | Storage management method of large-scale timing sequence data |
| CN102685221A (en) * | 2012-04-29 | 2012-09-19 | 华北电力大学（保定） | Distributed storage and parallel mining method for state monitoring data |
| CN103051707A (en) * | 2012-12-20 | 2013-04-17 | 浪潮集团有限公司 | Dynamic user behavior-based cloud forensics method and dynamic user behavior-based cloud forensics system |
| CN103390045A (en) * | 2013-07-19 | 2013-11-13 | 北京京东尚科信息技术有限公司 | Time sequence storage method and time sequence storage device for monitoring system |
| CN103605805A (en) * | 2013-12-09 | 2014-02-26 | 冶金自动化研究设计院 | Storage method of massive time series data |
| CN103678519A (en) * | 2013-11-29 | 2014-03-26 | 中国科学院计算技术研究所 | Mixed storage system and mixed storage method for supporting Hive DML (data manipulation language) enhancement |
-
2014
- 2014-04-11 CN CN201410143604.1A patent/CN104035956A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090216813A1 (en) * | 2008-02-27 | 2009-08-27 | Olivieri Ricardo N | Method and system for generating a transaction-bound sequence of records in a relational database table |
| CN101727465A (en) * | 2008-11-03 | 2010-06-09 | 中国移动通信集团公司 | Methods for establishing and inquiring index of distributed column storage database, device and system thereof |
| CN102043726A (en) * | 2010-12-29 | 2011-05-04 | 北京播思软件技术有限公司 | Storage management method of large-scale timing sequence data |
| CN102685221A (en) * | 2012-04-29 | 2012-09-19 | 华北电力大学（保定） | Distributed storage and parallel mining method for state monitoring data |
| CN103051707A (en) * | 2012-12-20 | 2013-04-17 | 浪潮集团有限公司 | Dynamic user behavior-based cloud forensics method and dynamic user behavior-based cloud forensics system |
| CN103390045A (en) * | 2013-07-19 | 2013-11-13 | 北京京东尚科信息技术有限公司 | Time sequence storage method and time sequence storage device for monitoring system |
| CN103678519A (en) * | 2013-11-29 | 2014-03-26 | 中国科学院计算技术研究所 | Mixed storage system and mixed storage method for supporting Hive DML (data manipulation language) enhancement |
| CN103605805A (en) * | 2013-12-09 | 2014-02-26 | 冶金自动化研究设计院 | Storage method of massive time series data |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331432A (en) * | 2014-10-22 | 2015-02-04 | 江苏瑞中数据股份有限公司 | Power grid mass time series data access method applicable to section access mode |
| CN106331075A (en) * | 2016-08-18 | 2017-01-11 | 华为技术有限公司 | Method for storing files, metadata server and manager |
| CN106331075B (en) * | 2016-08-18 | 2020-01-17 | 华为技术有限公司 | Method for storing file, metadata server and manager |
| CN106383844A (en) * | 2016-08-31 | 2017-02-08 | 天津南大通用数据技术股份有限公司 | Storage method and device applied to special data |
| WO2019114754A1 (en) * | 2017-12-12 | 2019-06-20 | 清华大学 | Join query method and system for multiple time sequences under columnar storage |
| CN108062378A (en) * | 2017-12-12 | 2018-05-22 | 清华大学 | The Connection inquiring method and system of more time serieses under a kind of column storage |
| CN108062378B (en) * | 2017-12-12 | 2018-12-11 | 清华大学 | The Connection inquiring method and system of more time serieses under a kind of storage of column |
| CN108595553B (en) * | 2018-04-10 | 2022-02-08 | 红云红河烟草(集团)有限责任公司 | Industrial data acquisition time sequence data compression storage and decompression query method based on relational database |
| CN108595553A (en) * | 2018-04-10 | 2018-09-28 | 红云红河烟草(集团)有限责任公司 | A kind of industrial number based on relevant database adopts time series data compression storage and decompression querying method |
| CN108647243B (en) * | 2018-04-13 | 2021-11-23 | 中国神华能源股份有限公司 | Industrial big data storage method based on time series |
| CN108647243A (en) * | 2018-04-13 | 2018-10-12 | 中国神华能源股份有限公司 | Industrial big data storage method based on time series |
| CN109960693A (en) * | 2018-11-22 | 2019-07-02 | 成都长城开发科技有限公司 | One kind being based on relevant database load curve storage method |
| CN109597588A (en) * | 2018-12-11 | 2019-04-09 | 浙江中智达科技有限公司 | A kind of date storage method, data restoration method and device |
| CN109597588B (en) * | 2018-12-11 | 2020-09-04 | 浙江中智达科技有限公司 | Data storage method, data restoration method and device |
| CN110968585A (en) * | 2019-12-20 | 2020-04-07 | 深圳前海微众银行股份有限公司 | Method, device and equipment for storing orientation column and computer readable storage medium |
| CN110968585B (en) * | 2019-12-20 | 2023-11-03 | 深圳前海微众银行股份有限公司 | Storage method, device, equipment and computer readable storage medium for alignment |
| CN113177090A (en) * | 2021-04-30 | 2021-07-27 | 中国邮政储蓄银行股份有限公司 | Data processing method and device |
| CN116204684A (en) * | 2023-02-01 | 2023-06-02 | 浙江正泰仪器仪表有限责任公司 | Storage method, device, equipment and medium of electric energy meter |
| CN116204684B (en) * | 2023-02-01 | 2024-06-04 | 浙江正泰仪器仪表有限责任公司 | Storage method, device, equipment and medium of electric energy meter |
| CN116719822A (en) * | 2023-08-10 | 2023-09-08 | 深圳市连用科技有限公司 | Method and system for storing massive structured data |
| CN116719822B (en) * | 2023-08-10 | 2023-12-22 | 深圳市连用科技有限公司 | Method and system for storing massive structured data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104035956A (en) | Time-series data storage method based on distributive column storage | |
| CN107423422B (en) | Spatial data distributed storage and search method and system based on grid | |
| CN108052551B (en) | Method for storing large amount of time sequence data realized on REDIS | |
| CN106682077B (en) | Mass time sequence data storage implementation method based on Hadoop technology | |
| CN106611046A (en) | Big data technology-based space data storage processing middleware framework | |
| CN102929936B (en) | Log recording method, log inquiring method and system | |
| CN110275920A (en) | Data query method, apparatus, electronic equipment and computer readable storage medium | |
| CN102629269B (en) | Searching and storing method for embedded database | |
| CN102332030A (en) | Data storing, managing and inquiring method and system for distributed key-value storage system | |
| CN106503276A (en) | A kind of method and apparatus of the time series databases for real-time monitoring system | |
| CN105677826A (en) | Resource management method for massive unstructured data | |
| CN105631003A (en) | Intelligent index establishing, inquiring and maintaining method supporting mass data classification and counting | |
| CN103488704A (en) | Method and device for storing data | |
| CN104008207B (en) | The data external storage system and date storage method of the database based on CD | |
| CN102779138B (en) | The hard disk access method of real time data | |
| CN103336849A (en) | Method and device for increasing retrieval speed in database retrieval system | |
| CN103488710B (en) | The non-fixed-length data method of efficient storage in big data page | |
| CN103885887A (en) | User data storage method, reading method and system | |
| CN109783441A (en) | Mass data inquiry method based on Bloom Filter | |
| CN104834648A (en) | Log query method and device | |
| CN104636349A (en) | Method and equipment for compression and searching of index data | |
| CN105787090A (en) | Index building method and system of OLAP system of electric data | |
| CN106780157B (en) | Ceph-based power grid multi-temporal model storage and management system and method | |
| CN101963993B (en) | Method for fast searching database sheet table record | |
| CN104156395A (en) | Data storage system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140910 |
|
| RJ01 | Rejection of invention patent application after publication |