WO2021027137A1 - 时序数据存储方法、装置、计算机设备和存储介质 - Google Patents

时序数据存储方法、装置、计算机设备和存储介质 Download PDF

Info

Publication number
WO2021027137A1
WO2021027137A1 PCT/CN2019/117280 CN2019117280W WO2021027137A1 WO 2021027137 A1 WO2021027137 A1 WO 2021027137A1 CN 2019117280 W CN2019117280 W CN 2019117280W WO 2021027137 A1 WO2021027137 A1 WO 2021027137A1
Authority
WO
WIPO (PCT)
Prior art keywords
data point
time series
series data
field
current time
Prior art date
Application number
PCT/CN2019/117280
Other languages
English (en)
French (fr)
Inventor
林泽全
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2021027137A1 publication Critical patent/WO2021027137A1/zh

Links

Images

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

Definitions

  • This application relates to a time series data storage method, device, computer equipment and storage medium.
  • Time series data is a series of data with a time stamp.
  • the attribute monitoring data of the unmanned vehicle at the first moment of time is one piece of time series data; the attribute monitoring data of the unmanned vehicle at the second moment is another piece of time series data.
  • time series data is stored based on a dedicated time series database. Due to the continuous generation of time series data, the data volume of time series data is usually very large. It is necessary to deploy multiple time series databases in a distributed manner to store massive time series data, which increases the storage cost of time series data.
  • a time-series data storage method is provided.
  • a method for storing time series data executed by a computer device, the method comprising: obtaining a current time series data point; the time series data point includes a timestamp field; according to the timestamp field, it is determined that the current time series data point corresponds to Calculate the relative change value of the current time series data point and the reference data point; add the relative change value to the compressed data point corresponding to the reference data point in the time series database according to the time sequence; and When receiving the next time series data point, mark the next time series data point as the current time series data point, and return to the step of determining the reference data point corresponding to the current time series data point.
  • the obtaining the current time series data points includes: monitoring the running status of the business product to obtain multiple time series data points; performing slicing processing on the multiple time series data points to obtain multiple time series fragments; Determine the traversal sequence of different time series data points in each time sequence segment according to the timestamp field; call multiple threads to synchronously traverse the time sequence data points of multiple time sequence segments; and obtain the time sequence of the current traversal sequence in each time sequence segment data point.
  • the determining the reference data point corresponding to the current time series data point according to the timestamp field includes: determining the compression period to which the timestamp field belongs; identifying whether there is a corresponding compression period Reference data point; if not, generate the reference data point corresponding to the compression period based on the initial time in the compression period; or mark the first time sequence data point in the compression period as the reference period corresponding to the compression period Data points; and marking the reference data point corresponding to the compression period as the reference data point corresponding to the current time series data point.
  • the relative change value includes a first difference value and a first difference value; the time series data point further includes a monitoring attribute field; the calculation of the relative change between the current time series data point and the reference data point The value includes: identifying whether the current time series data point is the first time series data point corresponding to the compression period; if so, calculating the current time series data point and the corresponding reference data point based on the first time stamp field Difference and the second difference based on the monitored attribute field; and otherwise, calculate the first difference between the current time series data point and the previous time sequence data point based on the timestamp field and the second difference based on the monitored attribute field Difference.
  • the reference data point includes a reference time stamp field and a reference attribute field; the calculation of the current time series data point and the corresponding reference data point is based on the first difference between the time stamp field and the monitoring attribute field.
  • the second difference of includes: performing time conversion on the current timestamp field and the reference timestamp field according to a preset compression accuracy, and calculating and converting the first difference between the current timestamp field and the reference timestamp field; And performing a target system conversion on the monitoring attribute field and the reference attribute field, and performing a preset logical operation on the converted monitoring attribute field and the reference attribute field to obtain a second difference.
  • the adding the relative change value to the compressed data point corresponding to the reference data point in the time series database according to the time sequence includes: using a first preset identifier and a second preset character to add The first difference value and the second difference value are spliced to obtain the target character string corresponding to the current time series data point; if the current time series data point is the time series data point in the first time sequence corresponding to the compression period, use all The target character string replaces the reference monitoring field in the reference data point to obtain the compressed data point corresponding to the compression period; and if the current time series data point is not the first time sequence data point corresponding to the compression period, use The first preset identifier splices the first difference value to the compressed data point corresponding to the compression field, and uses the second preset identifier to splice the second difference value into the first difference After the value.
  • the method further includes: receiving a data query request sent by the terminal; the data query request carries a query timestamp; determining the compression period to which the query timestamp belongs; obtaining the compressed data corresponding to the compression period Point; the compressed data point includes a plurality of target character strings arranged in chronological order; the plurality of target character strings in the compressed data point are decoded in chronological order; and the query timestamp corresponds to the target character in the chronological order
  • the decoded result of the string is returned to the terminal.
  • a time series data storage device comprising: a data receiving module for obtaining current time series data points; the time series data points comprising a time stamp field; and a compression processing module for determining all data points according to the time stamp field
  • a computer device includes a memory and one or more processors.
  • the memory stores computer-readable instructions.
  • the steps of the sequential data storage method provided in any embodiment of the present application are implemented.
  • One or more non-volatile computer-readable storage media storing computer-readable instructions.
  • the one or more processors implement any one of the embodiments of the present application. Provide the steps of the time series data storage method.
  • Fig. 1 is an application scenario diagram of a time-series data storage method according to one or more embodiments.
  • Fig. 2 is a schematic flowchart of a method for storing time series data according to one or more embodiments.
  • Fig. 3 is a schematic flowchart of a calculation step of a relative change value according to one or more embodiments.
  • Fig. 4 is a structural block diagram of a sequential data storage device according to one or more embodiments.
  • Figure 5 is a block diagram of a computer device according to one or more embodiments.
  • the time series data storage method provided in this application can be applied to the application environment as shown in FIG. 1.
  • the terminal 102 and the server 104 communicate through the network.
  • the terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices.
  • the server 104 may be implemented by an independent server or a server cluster composed of multiple servers.
  • the server 104 itself or other servers monitor the operating status of the service platform, and generate multiple time series data points. Time series data points include timestamp fields and monitoring attribute fields.
  • the server 104 deploys a corresponding time series database.
  • Each time the server 104 receives a time series data point it is compressed according to the following logic and then stored in the time series database, or multiple received time series data points are compressed according to a preset time frequency and then stored in the time series database. Specifically, the server 104 determines the reference data point corresponding to each time series data point according to the timestamp field, and calculates the relative change value between the current time series data point and the reference data point. The server 104 adds the relative change value after the compressed data point corresponding to the reference data point in the time series database according to the time sequence. The user can query the compressed data points in the time series database based on the terminal 102.
  • the above-mentioned time series data storage process compresses multiple time series data points according to the difference between the time series data points and the reference data points, which can greatly reduce the time series data storage cost while retaining the information contained in the massive time series data points.
  • a time-series data storage method is provided. Taking the method applied to the server in FIG. 1 as an example for description, the method includes the following steps:
  • Step 202 Obtain the current time series data point; the time series data point includes a timestamp field.
  • the server monitors the running status of the business product and generates multiple time series data points. Or, the server receives monitoring data containing multiple time series data points sent by other monitoring platforms.
  • Business products can be software products, automation products, or industrial machinery products.
  • the storage format of time series data points in the time series database is [timestamp field, monitoring attribute field].
  • the time series data point includes a time stamp field and a monitoring attribute field. Among them, the timestamp field is used to identify the time when the time series data point is generated.
  • Time series data points can include multiple monitoring attribute fields at the same time.
  • the monitoring attribute field includes a metric attribute field that changes with the time stamp.
  • three of the continuously generated time series data points can be: [2018-12-20 13:01:02,12], [2018-12-20 13:02:02,12] , [2018-12-20 13:03:02,24].
  • the monitored attribute field may also include a label attribute field that does not change with the time stamp.
  • the server can compress and store the received time series data points one by one. In other words, every time a time series data point is received, the server compresses the time series data point and stores the compression result in the time series database. The server can also compress and store the received time series data points in batches. The following describes the compressed storage one by one as an example.
  • Step 204 Determine the reference data point corresponding to the current time series data point according to the timestamp field.
  • determining the reference data point corresponding to the current time series data point according to the time stamp field includes: determining the compression period to which the time stamp field belongs; identifying whether there is a corresponding reference data point in the compression period; The initial time in the compression period generates the reference data point corresponding to the compression period; or the first time sequence data point in the compression period is marked as the reference data point corresponding to the compression period; the reference data point corresponding to the compression period is marked as the current The reference data point corresponding to the time series data point.
  • the server initializes the logic of compressed storage one by one or compressed storage in batches once every preset time. In other words, the server re-determines the reference data point of the corresponding compression period every preset duration. For example, assuming that the preset duration is 1 hour and the initial monitoring time is 8:00, the first compression period is 8:00 ⁇ 9:00, the next compression period is 9:00 ⁇ 10:00, and so on. All time series data points received in the same compression period have the same reference data point.
  • the benchmark data point includes a benchmark timestamp field and a benchmark attribute field.
  • the reference data point may be the first time series data point received in the corresponding compression period, that is, the time series data point with the earliest time stamp is determined as the reference data point.
  • the three time series data points in the above example belong to the simultaneous compression period, so the corresponding reference data points can all be [2018-12-20 13:01:02,12].
  • the reference data point may also be a time series data point customized according to the initial time in the compression period.
  • the reference time stamp field of the custom reference data point record can be different from the time stamp field of any time series data point record in the corresponding shard, and the reference attribute field of the record can also be the same as that of any time series data point record in the corresponding shard.
  • the monitoring attribute fields are different.
  • the reference data points corresponding to the three time series data points in the above example can all be [2018-12-20 13:00:00,00].
  • Step 206 Calculate the relative change value between the current time series data point and the reference data point.
  • the relative change value includes the difference value based on the timestamp field, which is recorded as the first difference value.
  • the relative change value also includes the difference value based on the monitored attribute field, which is recorded as the second difference value.
  • calculating the relative change value between the current time series data point and the reference data point includes: identifying whether the current time series data point is the first time sequence data point corresponding to the compression period; if so, calculating the current time series data point; The first difference between the time series data point and the corresponding reference data point is based on the time stamp field and the second difference based on the monitoring attribute field; otherwise, the current time series data point and the time series data point of the previous time sequence are calculated based on the time stamp field The first difference of and the second difference based on the monitored attribute field.
  • the server identifies whether the current time series data point is the first time series data point received in the current compression period. If the current time series data point is the first time series data point received in the current compression period, the server calculates the first difference and the second difference between the current time series data point and the corresponding reference data point. If the current time series data point is not the first time series data point received in the current compression period, the server calculates the first difference and the second difference between the current time series data point and the previous time series time series data point.
  • time series data point [2018-12-20 13:00:00,00] is used as the reference data point
  • the time series data point [2018-12-20 13:01:02,12] is relative to the reference data point
  • the first difference can be 62
  • the second difference can be 12
  • time series data point [2018-12-20 13:02:02,12] relative to the previous time sequence data point [2018-12-20 13: 01:02,12] may be 2 and the second difference may be 0.
  • the relative change value of all the time series data points is the change value of the reference data point instead of the change value of the time series data point relative to the previous time sequence, and there is no restriction on this.
  • Step 208 Add the relative change value to the compressed data point corresponding to the reference data point in the time series database according to the time sequence.
  • adding the relative change value to the compressed data point corresponding to the reference data point in the time series database according to the time sequence includes: using a first preset identifier and a second preset character to compare the first difference with The second difference is spliced to obtain the target character string corresponding to the current time series data point; if the current time series data point is the first time sequence data point corresponding to the compression period, the target character string is used to replace the benchmark in the benchmark data point Monitor the field to obtain the compressed data point corresponding to the compression period; if the current time series data point is not the first time sequence data point corresponding to the compression period, use the first preset identifier to splice the first difference to the compression field corresponding Using the second preset identifier to splice the second difference after the first difference.
  • the first preset identifier and the second identifier can be "->”, “
  • the server sequentially splices the first preset identifier, the first difference value, the second preset character, and the second difference value to obtain the target character string corresponding to the current time series data point.
  • the time series data point [2018-12-20 13:01:02,12] corresponds to the target string obtained by splicing can be "->62
  • the target string corresponding to the splicing can be "->-2
  • the server obtains the reference data point corresponding to the current compression period, and replaces the reference attribute field in the reference data point with that corresponding to the current time series data point The target string. If the compression period is divided into multiple compression moments according to the time when each time series data point is received, it is easy to understand that in the compression storage mode one by one, the data content of the compressed data point corresponding to each compression period is different at different compression moments. of. After the above-mentioned target character replacement processing, the compressed data point of the current compression period at the first compression moment can be obtained.
  • the compressed data point at the first compression moment can be obtained, and the compressed data point is ["2018- 12-20 13:00:00"->62
  • the target character string corresponding to the current time series data point of the server is spliced after the compressed data point corresponding to the previous compression time in the compression field. For example, in the above example, if the current time series data point is [2018-12-20 13:02:02,12], the compressed data point at the second compression moment can be obtained, and the compressed data point is [2018- 12-20 13:00:00->62
  • Step 210 When the next time series data point is received, mark the next time series data point as the current time series data point, and return to the step of determining the reference data point corresponding to the current time series data point.
  • the time series data point will be compressed according to the above method, and the target character string obtained by the compression process will be spliced to the compression field after the corresponding compressed data point at the previous compression time . This is repeated until the last time series data point received in the current compression period, and finally all the time series data points received in the current compression period are compressed into one compressed data point.
  • the reference data point corresponding to the current time series data point can be determined according to the time stamp field of the time series data point record; by calculating the current time series data point and the reference data point The relative change value, and the relative change value is appended to the compressed data point corresponding to the reference data point in the time series database in chronological order; each time series data point received afterwards is processed according to the above method, and a large number of time series data points can be realized Compression. Since multiple time series data points are compressed according to the difference between time series data points and reference data points, the compression logic is simple, which not only ensures the efficiency of time series data compression and storage, but also reduces the information contained in massive time series data points. Time series data storage cost.
  • obtaining the current time series data point includes: monitoring the running status of the business product to obtain multiple time series data points; performing slicing processing on the multiple time series data points to obtain multiple time series fragments; The timestamp field determines the traversal sequence of different time series data points in each time sequence segment; calls multiple threads to synchronously traverse the time series data points of multiple time series segments; obtains the time series data points of the current traversal sequence in each time sequence segment.
  • the server compresses and stores the received multiple time series data points in batches. Specifically, the server counts the number of time-series data points received within a preset time period, and determines whether the number reaches a preset compression threshold. If so, the server fragments the received multiple time series data points according to the preset fragmentation threshold to obtain multiple time series fragments. For example, if the fragmentation threshold is 60, 60 consecutively generated time series data points are regarded as one fragment.
  • the setting of the compression threshold and the fragmentation threshold should comprehensively consider the amount of data compression and subsequent decoding efficiency.
  • Each time sequence segment includes multiple time sequence data points.
  • the server can call multiple threads to synchronously traverse the time series data points of multiple time series segments, and compress the time series data points in different traversal sequences in each time series segment according to the above-mentioned method to obtain the corresponding time series segment. Of one or more compressed data points.
  • multiple time series data points are fragmented to achieve batch compression storage; the time series data points in multiple fragments are compressed synchronously to improve the efficiency of time series data compression.
  • the first difference between the current time series data point and the corresponding reference data point is calculated based on the timestamp field and the second difference based on the monitored attribute field, that is, the calculation of the relative change value
  • Step 302 Perform time conversion on the current timestamp field and the reference timestamp field according to the preset compression accuracy, and calculate and convert the first difference between the current timestamp field and the reference timestamp field.
  • the server calculates the first difference between the current time series data point and the corresponding reference data point based on the timestamp field based on the preset first operation logic.
  • the first operation logic may be to first perform time conversion on the current timestamp field and the reference timestamp field according to a preset compression accuracy, and then calculate the conversion to obtain the first difference between the current timestamp field and the reference timestamp field.
  • Compression accuracy refers to the time unit for time conversion.
  • the reference data point [2018-12-20 13:00:00,00] corresponds to the converted reference timestamp field, which can be 2018-12-2013:0
  • the time series data point is [2018-12-20 13:01:02,12]
  • the timestamp field after the corresponding time conversion can be 2018-12-20 13:62, so that the time series data point is the first of the reference data point
  • the difference can be 62.
  • Step 304 Perform a target hexadecimal conversion on the monitored attribute field and the reference attribute field, and perform a preset logical operation on the converted monitoring attribute field and the reference attribute field to obtain a second difference.
  • the server calculates the second difference between the current time series data point and the corresponding reference data point based on the monitored attribute field based on the preset second operation logic.
  • the second operation logic may be to perform target system conversion on the monitored attribute field and the reference attribute field, and then perform a preset logical operation on the converted monitoring attribute field and the reference attribute field.
  • the target base conversion can be binary conversion, hexadecimal conversion, etc.
  • the preset logical operation can be logical AND operation, logical OR operation, XOR operation, etc.
  • the reference data point [2018-12-20 13:00:00,00] corresponding to the converted reference attribute field can be 0X0000 0000
  • the time series data point is [2018-12-20 13:01:02,12]
  • the monitored attribute field after the corresponding time conversion can be 0X0000 1100.
  • the preset logic operation is an exclusive OR operation
  • the second difference between the time series data point and the reference data point The value can be 0X0000 1100.
  • the preset logical operation result is converted in a preset representation manner, and the converted preset logical operation result is used as the second difference.
  • the ratio A:B ratio can be used to represent the result of the preset logic operation.
  • A represents the preset value in the preset logic operation result
  • B is the preset value
  • the preset value B can be "1" in binary.
  • the preset logical operation result contains multiple consecutive preset values, it can be characterized in the form of A1-A2: B, etc., for example, 0X0001 1100 can be recorded as 4-6:1.
  • the preset logic operation result contains multiple preset values of intervals, it can be represented in the form of A1.A2:B, etc., for example, 0X0010 0110 can be represented as 3.6-7:1.
  • both the first arithmetic logic and the second arithmetic logic can adopt other arithmetic logics, and there is no limitation on this.
  • the calculation of the first difference and the second difference with simple arithmetic logic can not only improve the efficiency of data compression, but also lower the decoding threshold for compressed data points, thereby improving time series data query based on compressed data points. effectiveness.
  • the method further includes: receiving a data query request sent by the terminal; the data query request carries a query timestamp; determining the compression period to which the query timestamp belongs; obtaining the compressed data point corresponding to the compression period; compressing the data point It includes multiple target character strings arranged in chronological order; decodes multiple target character strings in the compressed data point in chronological order; and returns the decoding result of the target character string corresponding to the time sequence of the query timestamp to the terminal.
  • the user can enter the query time stamp through the terminal, and the terminal generates a data query request based on the query time stamp, and sends the data query request to the server.
  • users are also supported to query the operating status of the business product in a certain period of time.
  • the server determines the compression period to which the query timestamp belongs, and generates a query statement corresponding to the compression period, and queries the corresponding compressed data point in the time series database based on the query statement.
  • the compressed data point includes multiple target strings arranged in chronological order.
  • the server calculates the time sequence corresponding to the query timestamp in the compression period and records it as the target sequence.
  • the server traverses multiple target strings in the compressed data points in chronological order. Specifically, the server performs a reverse operation on the first difference in the target character string in the current traversal sequence according to the operation logic (denoted as the first reverse logic) opposite to the above first operation logic to obtain the initial timestamp field. The server performs a reverse operation on the second difference in the target character string in the current traversal sequence according to the operation logic (denoted as the second reverse logic) opposite to the above second operation logic to obtain the initial monitoring attribute field. The server uses the decoding result of the target character string in the current traversal sequence to decode the target character string in the next traversal sequence in the above-mentioned manner until the decoding result corresponding to the target character string in the target sequence is obtained. The server returns the decoding result corresponding to the target character string in the target order to the terminal.
  • the server uses the decoding result of the target character string in the current traversal sequence to decode the target character string in the next traversal sequence in the above-mentioned manner until the decoding
  • data decoding can be realized by reversely calculating the relative change value of the current time series data point and the reference data point.
  • the decoding logic is simple, and the time for decoding compressed data points during data query can be reduced, thereby ensuring Data query efficiency.
  • a time series data storage device which includes: a data receiving module 402, a compression processing module 404, and a compression storage module 406, wherein:
  • the data receiving module 402 is used to obtain the current time series data point; the time series data point includes a time stamp field;
  • the compression processing module 404 is configured to determine the reference data point corresponding to the current time series data point according to the timestamp field; calculate the relative change value between the current time series data point and the reference data point;
  • the compression storage module 406 is used to add the relative change value to the compressed data point corresponding to the reference data point in the time series database according to the time sequence; the data receiving module is also used to add the next time series data point when the next time series data point is received The point is marked as the current time series data point.
  • the data receiving module 402 is also used to monitor the running status of the business product to obtain multiple time series data points; perform slicing processing on the multiple time series data points to obtain multiple time series fragments;
  • the field determines the traversal order of different time series data points in each time sequence segment; calls multi-threads to synchronously traverse the time series data points of multiple time sequence segments; obtains the time series data points of the current traversal sequence in each time sequence segment.
  • the compression processing module 404 is also used to determine the compression period to which the timestamp field belongs; identify whether there is a corresponding reference data point in the compression period; if not, generate the compression period corresponding to the initial time in the compression period.
  • Reference data point or mark the first time sequence data point in the compression period as the reference data point corresponding to the compression period; mark the reference data point corresponding to the compression period as the reference data point corresponding to the current time sequence data point.
  • the relative change value includes a first difference value and a first difference value
  • the time series data point also includes a monitoring attribute field
  • the compression processing module 404 is also used to identify whether the current time series data point corresponds to the compression period.
  • Time series data points in the first time sequence if yes, calculate the first difference between the current time series data point and the corresponding reference data point based on the timestamp field and the second difference based on the monitored attribute field; otherwise, calculate the current time series data point
  • the time series data point in the previous time sequence is based on the first difference of the timestamp field and the second difference based on the monitored attribute field.
  • the reference data point includes a reference time stamp field and a reference attribute field; the compression processing module 404 is further configured to perform time conversion on the current time stamp field and the reference time stamp field according to a preset compression precision, and calculate the conversion Obtain the first difference between the current timestamp field and the reference timestamp field; perform target hexadecimal conversion on the monitored attribute field and the reference attribute field, and perform preset logical operations on the converted monitoring attribute field and reference attribute field to obtain The second difference.
  • the compression storage module 406 is configured to use the first preset identifier and the second preset character to splice the first difference value and the second difference value to obtain the target character string corresponding to the current time series data point ; If the current time series data point is the first time sequence data point corresponding to the compression period, replace the reference monitoring field in the reference data point with the target string to obtain the compressed data point corresponding to the compression period; if the current time series data If the point is not a time series data point in the first time sequence corresponding to the compression period, the first difference value is spliced to the compressed data point corresponding to the compressed field using the first preset identifier, and the second difference value is converted using the second preset identifier The splicing is after the first difference.
  • the device further includes a data query module 408 for receiving a data query request sent by the terminal; the data query request carries a query time stamp; determines the compression period to which the query time stamp belongs; and obtains the compression corresponding to the compression period Data points; compressed data points include multiple target strings arranged in chronological order; decode multiple target strings in the compressed data points in chronological order; return the decoded result of the target string corresponding to the time sequence of the query timestamp to terminal.
  • a data query module 408 for receiving a data query request sent by the terminal; the data query request carries a query time stamp; determines the compression period to which the query time stamp belongs; and obtains the compression corresponding to the compression period Data points; compressed data points include multiple target strings arranged in chronological order; decode multiple target strings in the compressed data points in chronological order; return the decoded result of the target string corresponding to the time sequence of the query timestamp to terminal.
  • Each module in the above-mentioned sequential data storage device may be implemented in whole or in part by software, hardware, and a combination thereof.
  • the foregoing modules may be embedded in the form of hardware or independent of the processor in the computer device, or may be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the foregoing modules.
  • a computer device is provided.
  • the computer device may be a server, and its internal structure diagram may be as shown in FIG. 5.
  • the computer equipment includes a processor, a memory, a network interface and a time series database connected through a system bus.
  • the processor of the computer device is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system, computer readable instructions, and a database.
  • the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium.
  • the time series database of the computer equipment is used to store compressed data points.
  • the network interface of the computer device is used to communicate with an external terminal through a network connection.
  • the computer-readable instructions are executed by the processor to realize a time-series data storage method.
  • FIG. 5 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied.
  • the specific computer device may Including more or fewer parts than shown in the figure, or combining some parts, or having a different arrangement of parts.
  • One or more non-volatile storage media storing computer-readable instructions.
  • the computer-readable instructions When executed by one or more processors, the one or more processors implement the timing sequence provided in any one of the embodiments of the present application. The steps of the data storage method.
  • Non-volatile memory may include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Channel (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

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)
  • Debugging And Monitoring (AREA)

Abstract

一种基于大数据处理技术的时序数据存储方法，包括：获取当前的时序数据点；所述时序数据点包括时间戳字段；根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；计算所述当前的时序数据点与基准数据点的相对变化值；根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点，返回至所述确定所述当前的时序数据点对应的基准数据点的步骤。

Description

时序数据存储方法、装置、计算机设备和存储介质
本申请要求于2019年08月13日提交中国专利局，申请号为2019107453842，申请名称为“时序数据存储方法、装置、计算机设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及一种时序数据存储方法、装置、计算机设备和存储介质。
背景技术
在运维场景中，需要对业务产品在多个时间节点的运行状态进行监控，从而有越来越多的时序数据产生。时序数据是带有时间戳的一系列数据。例如，无人车在第一时刻的坐标、速度、方向、温度、湿度等属性监控数据即为一条时序数据；无人车在第二时刻的属性监控数据即为另一条时序数据。目前基于专用的时序数据库存储时序数据。由于时序数据持续不间断产生，通常时序数据的数据量非常大，需要分布式部署多个时序数据库来存储海量的时序数据，使得时序数据存储成本升高。
发明内容
根据本申请公开的各种实施例，提供一种时序数据存储方法、装置、计算机设备和存储介质。
一种时序数据存储方法，由计算机设备执行，所述方法包括：获取当前的时序数据点；所述时序数据点包括时间戳字段；根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；计算所述当前的时序数据点与基准数据点的相对变化值；根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；及当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点，返回至所述确定所述当前的时序数据点对应的基准数据点的步骤。
在一个实施例中，所述获取当前的时序数据点，包括：对业务产品的运行状态进行监控，得到多个时序数据点；对多个时序数据点进行分片处理，得到多个时序片段；根据所述时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；调用多线程对多个所述时序片段的时序数据点进行同步遍历；及获取每个时序片段中当前遍历顺序的时序数据点。
在一个实施例中，所述根据所述时间戳字段确定所述当前的时序数据点对应的基准数据点，包括：确定所述时间戳字段所属的压缩时段；识别所述 压缩时段是否存在对应的基准数据点；若否，基于所属压缩时段中的初始时间生成所述压缩时段对应的基准数据点；或将所述压缩时段中第一时间顺序的时序数据点标记为所述压缩时段对应的基准数据点；及将所述压缩时段对应的基准数据点标记为所述当前的时序数据点对应的基准数据点。
在一个实施例中，所述相对变化值包括第一差值和第一差值；所述时序数据点还包括监控属性字段；所述计算所述当前的时序数据点与基准数据点的相对变化值，包括：识别所述当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；若是，计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；及否则，计算所述当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
在一个实施例中，所述基准数据点包括基准时间戳字段和基准属性字段；所述计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值，包括：按照预设的压缩精度对当前的时间戳字段及所述基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值；及对所述监控属性字段及所述基准属性字段进行目标进制转换，并对转换后的所述监控属性字段及所述基准属性字段进行预设逻辑运算，得到第二差值。
在一个实施例中，所述根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点，包括：利用第一预设标识符和第二预设字符将所述第一差值与所述第二差值拼接，得到当前的时序数据点对应的目标字符串；若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据点，利用所述目标字符串替换所述基准数据点中的基准监控字段，得到所述压缩时段对应的压缩数据点；及若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用所述第一预设标识符将所述第一差值拼接至所述压缩字段对应的压缩数据点，利用所述第二预设标识符将所述第二差值拼接在所述第一差值之后。
在一个实施例中，该方法还包括：接收终端发送的数据查询请求；所述数据查询请求携带了查询时间戳；确定所述查询时间戳所属的压缩时段；获取所述压缩时段对应的压缩数据点；所述压缩数据点包括按照时间顺序排列的多个目标字符串；按照时间顺序对所述压缩数据点中多个目标字符串进行解码；及将所述查询时间戳对应时间顺序的目标字符串的解码结果返回至终端。
一种时序数据存储装置，所述装置包括：数据接收模块，用于获取当前的时序数据点；所述时序数据点包括时间戳字段；压缩处理模块，用于根据 所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；计算所述当前的时序数据点与基准数据点的相对变化值；压缩存储模块，用于根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；及所述数据接收模块还用于当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点。
一种计算机设备，包括存储器和一个或多个处理器，存储器中存储有计算机可读指令，计算机可读指令被处理器执行时实现本申请任意一个实施例中提供的时序数据存储方法的步骤。
一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器实现本申请任意一个实施例中提供的时序数据存储方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1为根据一个或多个实施例中时序数据存储方法的应用场景图。
图2为根据一个或多个实施例中时序数据存储方法的流程示意图。
图3为根据一个或多个实施例中相对变化值的计算步骤的流程示意图。
图4为根据一个或多个实施例中时序数据存储装置的结构框图。
图5为根据一个或多个实施例中计算机设备的框图。
具体实施方式
为了使本申请的技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供的时序数据存储方法，可以应用于如图1所示的应用环境中。终端102与服务器104通过网络进行通信。终端102可以但不限于是各种个人计算机、笔记本电脑、智能手机、平板电脑和便携式可穿戴设备，服务器104可以用独立的服务器或者是多个服务器组成的服务器集群来实现。服务器104自身或者其他服务器对业务平台的运行状态进行监控，产生多个时序数据点。时序数据点包括时间戳字段和监控属性字段。服务器104部署了对 应的时序数据库。服务器104每接收到一个时序数据点按照以下逻辑进行压缩处理后将其存储至时序数据库，或者按照预设时间频率对接收到的多个时序数据点进行压缩后进行存储至时序数据库。具体的，服务器104根据时间戳字段确定每个时序数据点对应的基准数据点，并计算当前的时序数据点与基准数据点的相对变化值。服务器104根据时间顺序将相对变化值追加在时序数据库中与基准数据点对应的压缩数据点之后。用户可以基于终端102对时序数据库中的压缩数据点进行查询。上述时序数据存储过程，根据时序数据点与基准数据点之间的差异对多个时序数据点进行压缩，在保留海量时序数据点蕴含的信息的同时，可以大大减少时序数据存储成本。
在其中一个实施例中，如图2所示，提供了一种时序数据存储方法，以该方法应用于图1中的服务器为例进行说明，包括以下步骤：
步骤202，获取当前的时序数据点；时序数据点包括时间戳字段。
服务器对业务产品的运行状态进行监控，产生多个时序数据点。或者，服务器接收其他监控平台发送的包含多个时序数据点的监控数据。业务产品可以是软件产品、也可以是自动化产品，还可以是工业机械产品。时序数据点在时序数据库中的存储格式为[时间戳字段，监控属性字段]。换言之，时序数据点包括时间戳字段和监控属性字段。其中，时间戳字段用于标识该时序数据点产生的时间。时序数据点可以同时包括多种监控属性字段。监控属性字段包括随着时间戳变化而变化的度量属性字段。例如，对业务产品A进行监控，连续产生的其中三个时序数据点可以是：[2018-12-20 13:01:02,12]，[2018-12-20 13:02:02,12]，[2018-12-20 13:03:02,24]。在另一个实施例中，监控属性字段还可以包括不随着时间戳变化而变化的标签属性字段。
服务器可以对接收到的时序数据点逐个压缩存储。换言之，每接收到一个时序数据点，服务器对时序数据点进行压缩处理，将压缩结果存储至时序数据库。服务器也可以对接收到的时序数据点进行批量压缩存储。以下以逐个压缩存储为例进行描述。
步骤204，根据时间戳字段，确定当前的时序数据点对应的基准数据点。
在其中一个实施例中，根据时间戳字段确定当前的时序数据点对应的基准数据点，包括：确定时间戳字段所属的压缩时段；识别压缩时段是否存在对应的基准数据点；若否，基于所属压缩时段中的初始时间生成压缩时段对应的基准数据点；或将压缩时段中第一时间顺序的时序数据点标记为压缩时段对应的基准数据点；将压缩时段对应的基准数据点标记为当前的时序数据点对应的基准数据点。
服务器每隔预设时长初始化一次上述逐个压缩存储或者批量压缩存储的 逻辑。换言之，服务器每隔预设时长重新确定一次相应压缩时段的基准数据点。例如，假设预设时长为1小时，起始监控时间为8:00，则第一个压缩时段为8:00～9:00，下一个压缩时段为9:00～10:00，如此类推。同一压缩时段接收到的全部时序数据点具有相同的基准数据点。
基准数据点包括基准时间戳字段和基准属性字段。基准数据点可以为相应压缩时段接收到的第一个时序数据点，即将时间戳最早的一个时序数据点确定为基准数据点。例如，上述举例中三个时序数据点属于同时压缩时段，因而对应的基准数据点均可以是[2018-12-20 13:01:02,12]。
基准数据点也可以是根据所属压缩时段中的初始时间自定义的一个时序数据点。换言之，自定义基准数据点记录的基准时间戳字段可以与相应分片中任意一个时序数据点记录的时间戳字段不同，记录的基准属性字段也可以与相应分片中任意一个时序数据点记录的监控属性字段不同。例如，上述举例中三个时序数据点对应的基准数据点均可以是[2018-12-20 13:00:00,00]。
步骤206，计算当前的时序数据点与基准数据点的相对变化值。
相对变化值包括基于时间戳字段的差值，记作第一差值。相对变化值还包括基于监控属性字段的差值，记作第二差值。
在其中一个实施例中，计算当前的时序数据点与基准数据点的相对变化值，包括：识别当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；若是，计算当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；否则，计算当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
服务器识别当前的时序数据点是否为当前压缩时段接收到的第一个时序数据点。若当前的时序数据点为当前压缩时段接收到的第一个时序数据点，服务器计算当前的时序数据点与相应基准数据点的第一差值和第二差值。若当前的时序数据点并非当前压缩时段接收到的第一个时序数据点，则服务器计算当前的时序数据点与前一时间顺序的时序数据点的第一差值和第二差值。例如，在上述举例中，若将[2018-12-20 13:00:00,00]作为基准数据点，则时序数据点[2018-12-20 13:01:02,12]相对基准数据点的第一差值可以是62，第二差值可以是12；时序数据点[2018-12-20 13:02:02,12]相对前一时间顺序时序数据点[2018-12-20 13:01:02,12]的第一差值可以是2，第二差值可以是0。
容易理解，也可以采用其他的运算逻辑，比如全部时序数据点的相对变化值是其相对基准数据点的变化值，而非相对前一时间顺序的时序数据点的变化值，对此不作限制。
步骤208，根据时间顺序将相对变化值追加至时序数据库中与基准数据 点对应的压缩数据点。
在其中一个实施例中，根据时间顺序将相对变化值追加至时序数据库中与基准数据点对应的压缩数据点，包括：利用第一预设标识符和第二预设字符将第一差值与第二差值拼接，得到当前的时序数据点对应的目标字符串；若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据点，利用目标字符串替换基准数据点中的基准监控字段，得到压缩时段对应的压缩数据点；若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用第一预设标识符将第一差值拼接至压缩字段对应的压缩数据点，利用第二预设标识符将第二差值拼接在第一差值之后。
第一预设标识符与第二标识分别可以是“->”，“|”，“/”，“_”，“#”，“@”或“*”等。服务器将第一预设标识符、第一差值、第二预设字符及第二差值依次拼接，可以得到当前的时序数据点对应的目标字符串。例如，在上述举例中，时序数据点[2018-12-20 13:01:02,12]对应拼接得到的目标字符串可以是“->62|12”；时序数据点[2018-12-20 13:02:02,12]对应拼接得到的目标字符串可以是“->-2|0”。
若当前的时序数据点为当前压缩时段接收到的第一个时序数据点，服务器获取当前压缩时段对应的基准数据点，将该基准数据点中的基准属性字段替换为当前的时序数据点对应的目标字符串。若根据接收到每个时序数据点的时间将压缩时段区分为多个压缩时刻，则容易理解，在逐个压缩存储方式中，每个压缩时段对应的压缩数据点在不同压缩时刻的数据内容是不同的。经过上述目标字符替换处理，可以得到当前压缩时段在第一压缩时刻的压缩数据点。例如，在上述举例中，若当前的时序数据点为[2018-12-2013:01:02,12]，则可以得到第一压缩时刻的压缩数据点，且该压缩数据点为["2018-12-20 13:00:00"->62|12]。
若当前的时序数据点并非当前压缩时段接收到的第一个时序数据点，则服务器当前的时序数据点对应的目标字符串拼接在所属压缩字段在前一压缩时刻对应的压缩数据点之后。例如，在上述举例中，若当前的时序数据点为[2018-12-20 13:02:02,12]，则可以得到第二压缩时刻的压缩数据点，且该压缩数据点为[2018-12-20 13:00:00->62|12->-2|0]。
步骤210，当接收到下一个时序数据点时，将下一个时序数据点标记为当前的时序数据点，返回至确定当前的时序数据点对应的基准数据点的步骤。
若当前压缩时段接收到另一个时序数据点，则按照上述方式对该时序数据点进行压缩处理，并将压缩处理得到的目标字符串拼接至所属压缩字段在前一压缩时刻对应的压缩数据点之后。如此重复，直至当前压缩时段接收到的最后一个时序数据点，最终将当前压缩时段接收到的全部时序数据点压缩 为一个压缩数据点。
本实施例中，当接收到当前的时序数据点时，根据时序数据点记录的时间戳字段，可以确定当前的时序数据点对应的基准数据点；通过计算当前的时序数据点与基准数据点的相对变化值，并将相对变化值按照时间顺序追加至时序数据库中与基准数据点对应的压缩数据点；按照上述方式对后接收到的每个时序数据点进行处理，可以实现对大量时序数据点的压缩。由于根据时序数据点与基准数据点之间的差异对多个时序数据点进行压缩，压缩逻辑简单，不仅可以保证时序数据压缩存储效率，且在保留海量时序数据点蕴含的信息的同时，可以减少时序数据存储成本。
在其中一个实施例中，获取当前的时序数据点，包括：对业务产品的运行状态进行监控，得到多个时序数据点；对多个时序数据点进行分片处理，得到多个时序片段；根据时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；调用多线程对多个时序片段的时序数据点进行同步遍历；获取每个时序片段中当前遍历顺序的时序数据点。
服务器对接收到的多个时序数据点进行批量压缩存储。具体的，服务器统计在预设时长内接收到的时序数据点的数量，判断该数量是否达到预设的压缩阈值。若是，服务器根据预设的分片阈值对接收到的多个时序数据点进行分片，得到多个时序片段。例如，若分片阈值为60，则将连续产生的60个时序数据点作为一个分片。压缩阈值及分片阈值的设定应当综合考虑数据压缩量及后续解码效率。
每个时序片段包括多个时序数据点。为了提高压缩存储效率，服务器可以调用多线程对多个时序片段的时序数据点进行同步遍历，按照上述方式对每个时序片段中不同遍历顺序的时序数据点进行压缩处理，得到每个时序片段对应的一个或多个压缩数据点。
本实施例中，对多个时序数据点进行分片，可以实现批量压缩存储；同步对多个分片中的时序数据点进行压缩处理，提高时序数据压缩效率。
在其中一个实施例中，如图3所示，计算当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值，即相对变化值计算的步骤，包括：
步骤302，按照预设的压缩精度对当前的时间戳字段及基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值。
服务器基于预设的第一运算逻辑计算当前的时序数据点与相应基准数据点基于时间戳字段的第一差值。其中，第一运算逻辑可以是首先按照预设的压缩精度对当前的时间戳字段及基准时间戳字段进行时间换算，再计算换算 得到当前的时间戳字段相对基准时间戳字段的第一差值。压缩精度是指进行时间换算的时间单位。例如，在上述举例中，若压缩精度为“分”，则基准数据点[2018-12-20 13:00:00,00]对应转换后的基准时间戳字段可以是2018-12-2013:0，时序数据点为[2018-12-20 13:01:02,12]对应时间转换后的时间戳字段可以是2018-12-20 13:62，从而该时序数据点相对基准数据点的第一差值可以是62。
步骤304，对监控属性字段及基准属性字段进行目标进制转换，并对转换后的监控属性字段及基准属性字段进行预设逻辑运算，得到第二差值。
服务器基于预设的第二运算逻辑计算当前的时序数据点与相应基准数据点基于监控属性字段的第二差值。其中，第二运算逻辑可以是对监控属性字段及基准属性字段进行目标进制转换，再对转换后的监控属性字段及基准属性字段进行预设逻辑运算。目标进制转换可以是二进制转换、十六进制转换等，预设逻辑运算可以是逻辑与运算、逻辑或运算、异或运算等。例如，在上述举例中，若目标进制转换为二进制转换，则基准数据点[2018-12-20 13:00:00,00]对应转换后的基准属性字段可以是0X0000 0000，时序数据点为[2018-12-20 13:01:02,12]对应时间转换后的监控属性字段可以是0X0000 1100，若预设逻辑运算为异或运算，则该时序数据点相对基准数据点的第二差值可以是0X0000 1100。
在另一个实施例中，以预设的表征方式对预设逻辑运算结果进行转换，将转换后的预设逻辑运算结果作为第二差值。例如，可以采用比值A：B比值的形式表征预设逻辑运算结果。其中，A为代表预设逻辑运算结果中第几位为预设值；B为预设值，预设值B可以是二进制中的“1”。若预设逻辑运算结果包含连续的多个预设值，则可以表征为A1-A2：B等的形式，比如，0X0001 1100可以记录为4-6:1。若预设逻辑运算结果包含间隔的多个预设值，则可以表征为A1.A2:B等的形式，比如，0X0010 0110可以表征为3.6-7:1。
需要说明的是，第一运算逻辑及第二运算逻辑均可以采用其他的运算逻辑，对此不作限制。
本实施例中，以简单的运算逻辑计算第一差值及第二差值，不仅可以提高数据压缩效率，还可降低对压缩数据点的解码门槛，进而可以提高基于压缩数据点的时序数据查询效率。
在其中一个实施例中，该方法还包括：接收终端发送的数据查询请求；数据查询请求携带了查询时间戳；确定查询时间戳所属的压缩时段；获取压缩时段对应的压缩数据点；压缩数据点包括按照时间顺序排列的多个目标字符串；按照时间顺序对压缩数据点中多个目标字符串进行解码；将查询时间戳对应时间顺序的目标字符串的解码结果返回至终端。
当对业务产品进行问题排查时，可以对业务产品在某个时间点的运行状态进行查询。比如要查询业务产品在2018-12-20 17:01:05的运行状态。用户可以通过终端录入查询时间戳，终端基于查询时间戳生成数据查询请求，将数据查询请求发送至服务器。在另一个实施例中，也支持用户对业务产品在某个时间段的运行状态进行查询。
服务器确定查询时间戳所属的压缩时段，并生成该压缩时段对应的查询语句，基于查询语句在时序数据库中查询对应的压缩数据点。压缩数据点包括按照时间顺序排列的多个目标字符串。服务器计算查询时间戳在所属压缩时段对应的时间顺序，记作目标顺序。
服务器按照时间顺序对压缩数据点中多个目标字符串进行遍历。具体的，服务器按照与上述第一运算逻辑相反的运算逻辑(记作第一反向逻辑)对当前遍历顺序的目标字符串中的第一差值进行反向运算，得到初始的时间戳字段。服务器按照与上述第二运算逻辑相反的运算逻辑(记作第二反向逻辑)对当前遍历顺序的目标字符串中的第二差值进行反向运算，得到初始的监控属性字段。服务器按照上述方式利用当前遍历顺序的目标字符串的解码结果对下一遍历顺序的目标字符串进行解码，直至得到目标顺序的目标字符串对应的解码结果。服务器将目标顺序的目标字符串对应的解码结果返回至终端。
本实施例中，只需反向计算当前的时序数据点与基准数据点的相对变化值即可实现数据解码，解码逻辑简单，可以减少在数据查询时对压缩数据点解码的时间，从而可以保证数据查询效率。
应该理解的是，虽然图2和图3的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图2和图3中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些子步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在其中一个实施例中，如图4所示，提供了一种时序数据存储装置，包括：数据接收模块402、压缩处理模块404和压缩存储模块406，其中：
数据接收模块402，用于获取当前的时序数据点；时序数据点包括时间戳字段；
压缩处理模块404，用于根据时间戳字段，确定当前的时序数据点对应的基准数据点；计算当前的时序数据点与基准数据点的相对变化值；
压缩存储模块406，用于根据时间顺序将相对变化值追加至时序数据库中与基准数据点对应的压缩数据点；数据接收模块还用于当接收到下一个时序数据点时，将下一个时序数据点标记为当前的时序数据点。
在其中一个实施例中，数据接收模块402还用于对业务产品的运行状态进行监控，得到多个时序数据点；对多个时序数据点进行分片处理，得到多个时序片段；根据时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；调用多线程对多个时序片段的时序数据点进行同步遍历；获取每个时序片段中当前遍历顺序的时序数据点。
在其中一个实施例中，压缩处理模块404还用于确定时间戳字段所属的压缩时段；识别压缩时段是否存在对应的基准数据点；若否，基于所属压缩时段中的初始时间生成压缩时段对应的基准数据点；或将压缩时段中第一时间顺序的时序数据点标记为压缩时段对应的基准数据点；将压缩时段对应的基准数据点标记为当前的时序数据点对应的基准数据点。
在其中一个实施例中，相对变化值包括第一差值和第一差值；时序数据点还包括监控属性字段；压缩处理模块404还用于识别当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；若是，计算当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；否则，计算当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
在其中一个实施例中，基准数据点包括基准时间戳字段和基准属性字段；压缩处理模块404还用于按照预设的压缩精度对当前的时间戳字段及基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值；对监控属性字段及基准属性字段进行目标进制转换，并对转换后的监控属性字段及基准属性字段进行预设逻辑运算，得到第二差值。
在其中一个实施例中，压缩存储模块406，用于利用第一预设标识符和第二预设字符将第一差值与第二差值拼接，得到当前的时序数据点对应的目标字符串；若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据点，利用目标字符串替换基准数据点中的基准监控字段，得到压缩时段对应的压缩数据点；若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用第一预设标识符将第一差值拼接至压缩字段对应的压缩数据点，利用第二预设标识符将第二差值拼接在第一差值之后。
在其中一个实施例中，该装置还包括数据查询模块408，用于接收终端发送的数据查询请求；数据查询请求携带了查询时间戳；确定查询时间戳所属的压缩时段；获取压缩时段对应的压缩数据点；压缩数据点包括按照时间顺 序排列的多个目标字符串；按照时间顺序对压缩数据点中多个目标字符串进行解码；将查询时间戳对应时间顺序的目标字符串的解码结果返回至终端。
关于时序数据存储装置的具体限定可以参见上文中对于时序数据存储方法的限定，在此不再赘述。上述时序数据存储装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在其中一个实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图5所示。该计算机设备包括通过系统总线连接的处理器、存储器、网络接口和时序数据库。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机可读指令和数据库。该内存储器为非易失性存储介质中的操作系统和计算机可读指令的运行提供环境。该计算机设备的时序数据库用于存储压缩数据点。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机可读指令被处理器执行时以实现一种时序数据存储方法。
本领域技术人员可以理解，图5中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
一个或多个存储有计算机可读指令的非易失性存储介质，计算机可读指令被一个或多个处理器执行时，使得一个或多个处理器实现本申请任意一个实施例中提供的时序数据存储方法的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机可读指令来指令相关的硬件来完成，的计算机可读指令可存储于一非易失性计算机可读取存储介质中，该计算机可读指令在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限，RAM以多种形式可得，诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink) DRAM(SLDRAM)、存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种时序数据存储方法，由计算机设备执行，所述方法包括：
    获取当前的时序数据点；所述时序数据点包括时间戳字段；
    根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；
    计算所述当前的时序数据点与基准数据点的相对变化值；
    根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；及
    当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点，返回至所述确定所述当前的时序数据点对应的基准数据点的步骤。
  2. 根据权利要求1所述的方法，其特征在于，所述获取当前的时序数据点，包括：
    对业务产品的运行状态进行监控，得到多个时序数据点；
    对多个时序数据点进行分片处理，得到多个时序片段；
    根据所述时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；
    调用多线程对多个所述时序片段的时序数据点进行同步遍历；及
    获取每个时序片段中当前遍历顺序的时序数据点。
  3. 根据权利要求1所述的方法，其特征在于，所述根据所述时间戳字段确定所述当前的时序数据点对应的基准数据点，包括：
    确定所述时间戳字段所属的压缩时段；
    识别所述压缩时段是否存在对应的基准数据点；
    若否，基于所属压缩时段中的初始时间生成所述压缩时段对应的基准数据点；或将所述压缩时段中第一时间顺序的时序数据点标记为所述压缩时段对应的基准数据点；及
    将所述压缩时段对应的基准数据点标记为所述当前的时序数据点对应的基准数据点。
  4. 根据权利要求3所述的方法，其特征在于，所述相对变化值包括第一差值和第一差值；所述时序数据点还包括监控属性字段；所述计算所述当前的时序数据点与基准数据点的相对变化值，包括：
    识别所述当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；
    若是，计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；及
    否则，计算所述当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
  5. 根据权利要求4所述的方法，其特征在于，所述基准数据点包括基准时间戳字段和基准属性字段；所述计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值，包括：
    按照预设的压缩精度对当前的时间戳字段及所述基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值；及
    对所述监控属性字段及所述基准属性字段进行目标进制转换，并对转换后的所述监控属性字段及所述基准属性字段进行预设逻辑运算，得到第二差值。
  6. 根据权利要求4所述的方法，其特征在于，所述根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点，包括：
    利用第一预设标识符和第二预设字符将所述第一差值与所述第二差值拼接，得到当前的时序数据点对应的目标字符串；
    若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据点，利用所述目标字符串替换所述基准数据点中的基准监控字段，得到所述压缩时段对应的压缩数据点；及
    若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用所述第一预设标识符将所述第一差值拼接至所述压缩字段对应的压缩数据点，利用所述第二预设标识符将所述第二差值拼接在所述第一差值之后。
  7. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    接收终端发送的数据查询请求；所述数据查询请求携带了查询时间戳；
    确定所述查询时间戳所属的压缩时段；
    获取所述压缩时段对应的压缩数据点；所述压缩数据点包括按照时间顺序排列的多个目标字符串；
    按照时间顺序对所述压缩数据点中多个目标字符串进行解码；及
    将所述查询时间戳对应时间顺序的目标字符串的解码结果返回至终端。
  8. 一种时序数据存储装置，所述装置包括：数据接收模块，用于获取当前的时序数据点；所述时序数据点包括时间戳字段；压缩处理模块，用于根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；计算所述当前的时序数据点与基准数据点的相对变化值；压缩存储模块，用于根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；所述数据接收模块还用于当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点。
  9. 一种计算机设备，包括存储器及一个或多个处理器，所述存储器中储 存有计算机可读指令，所述计算机可读指令被所述一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取当前的时序数据点；所述时序数据点包括时间戳字段；
    根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；
    计算所述当前的时序数据点与基准数据点的相对变化值；
    根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；及
    当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点，返回至所述确定所述当前的时序数据点对应的基准数据点的步骤。
  10. 根据权利要求9所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    对业务产品的运行状态进行监控，得到多个时序数据点；
    对多个时序数据点进行分片处理，得到多个时序片段；
    根据所述时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；
    调用多线程对多个所述时序片段的时序数据点进行同步遍历；及
    获取每个时序片段中当前遍历顺序的时序数据点。
  11. 根据权利要求9所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    确定所述时间戳字段所属的压缩时段；
    识别所述压缩时段是否存在对应的基准数据点；
    若否，基于所属压缩时段中的初始时间生成所述压缩时段对应的基准数据点；或将所述压缩时段中第一时间顺序的时序数据点标记为所述压缩时段对应的基准数据点；及
    将所述压缩时段对应的基准数据点标记为所述当前的时序数据点对应的基准数据点。
  12. 根据权利要求11所述的计算机设备，其特征在于，所述相对变化值包括第一差值和第一差值；所述时序数据点还包括监控属性字段；所述处理器执行所述计算机可读指令时还执行以下步骤：
    识别所述当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；
    若是，计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；及
    否则，计算所述当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
  13. 根据权利要求12所述的计算机设备，其特征在于，所述基准数据点包括基准时间戳字段和基准属性字段；所述处理器执行所述计算机可读指令时还执行以下步骤：
    按照预设的压缩精度对当前的时间戳字段及所述基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值；及
    对所述监控属性字段及所述基准属性字段进行目标进制转换，并对转换后的所述监控属性字段及所述基准属性字段进行预设逻辑运算，得到第二差值。
  14. 根据权利要求12所述的计算机设备，其特征在于，所述处理器执行所述计算机可读指令时还执行以下步骤：
    利用第一预设标识符和第二预设字符将所述第一差值与所述第二差值拼接，得到当前的时序数据点对应的目标字符串；
    若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据点，利用所述目标字符串替换所述基准数据点中的基准监控字段，得到所述压缩时段对应的压缩数据点；及
    若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用所述第一预设标识符将所述第一差值拼接至所述压缩字段对应的压缩数据点，利用所述第二预设标识符将所述第二差值拼接在所述第一差值之后。
  15. 一个或多个存储有计算机可读指令的非易失性计算机可读存储介质，所述计算机可读指令被一个或多个处理器执行时，使得所述一个或多个处理器执行以下步骤：
    获取当前的时序数据点；所述时序数据点包括时间戳字段；
    根据所述时间戳字段，确定所述当前的时序数据点对应的基准数据点；
    计算所述当前的时序数据点与基准数据点的相对变化值；
    根据时间顺序将所述相对变化值追加至时序数据库中与所述基准数据点对应的压缩数据点；及
    当接收到下一个时序数据点时，将所述下一个时序数据点标记为当前的时序数据点，返回至所述确定所述当前的时序数据点对应的基准数据点的步骤。
  16. 根据权利要求15所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    对业务产品的运行状态进行监控，得到多个时序数据点；
    对多个时序数据点进行分片处理，得到多个时序片段；
    根据所述时间戳字段确定每个时序片段中不同时序数据点的遍历顺序；
    调用多线程对多个所述时序片段的时序数据点进行同步遍历；及
    获取每个时序片段中当前遍历顺序的时序数据点。
  17. 根据权利要求15所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    确定所述时间戳字段所属的压缩时段；
    识别所述压缩时段是否存在对应的基准数据点；
    若否，基于所属压缩时段中的初始时间生成所述压缩时段对应的基准数据点；或将所述压缩时段中第一时间顺序的时序数据点标记为所述压缩时段对应的基准数据点；及
    将所述压缩时段对应的基准数据点标记为所述当前的时序数据点对应的基准数据点。
  18. 根据权利要求17所述的存储介质，其特征在于，所述相对变化值包括第一差值和第一差值；所述时序数据点还包括监控属性字段；所述计算机可读指令被所述处理器执行时还执行以下步骤：
    识别所述当前的时序数据点是否为所属压缩时段对应的第一时间顺序的时序数据点；
    若是，计算所述当前的时序数据点与相应基准数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值；及
    否则，计算所述当前的时序数据点与前一时间顺序的时序数据点基于时间戳字段的第一差值以及基于监控属性字段的第二差值。
  19. 根据权利要求17所述的存储介质，其特征在于，所述基准数据点包括基准时间戳字段和基准属性字段；所述计算机可读指令被所述处理器执行时还执行以下步骤：
    按照预设的压缩精度对当前的时间戳字段及所述基准时间戳字段进行时间换算，计算换算得到当前的时间戳字段相对基准时间戳字段的第一差值；及
    对所述监控属性字段及所述基准属性字段进行目标进制转换，并对转换后的所述监控属性字段及所述基准属性字段进行预设逻辑运算，得到第二差值。
  20. 根据权利要求15所述的存储介质，其特征在于，所述计算机可读指令被所述处理器执行时还执行以下步骤：
    利用第一预设标识符和第二预设字符将所述第一差值与所述第二差值拼接，得到当前的时序数据点对应的目标字符串；
    若当前的时序数据点为所属压缩时段对应的第一时间顺序的时序数据 点，利用所述目标字符串替换所述基准数据点中的基准监控字段，得到所述压缩时段对应的压缩数据点；及
    若当前的时序数据点并非所属压缩时段对应的第一时间顺序的时序数据点，利用所述第一预设标识符将所述第一差值拼接至所述压缩字段对应的压缩数据点，利用所述第二预设标识符将所述第二差值拼接在所述第一差值之后。
PCT/CN2019/117280 2019-08-13 2019-11-12 时序数据存储方法、装置、计算机设备和存储介质 WO2021027137A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910745384.2A CN110633277B (zh) 2019-08-13 2019-08-13 时序数据存储方法、装置、计算机设备和存储介质
CN201910745384.2 2019-08-13

Publications (1)

Publication Number Publication Date
WO2021027137A1 true WO2021027137A1 (zh) 2021-02-18

Family

ID=68970389

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/117280 WO2021027137A1 (zh) 2019-08-13 2019-11-12 时序数据存储方法、装置、计算机设备和存储介质

Country Status (2)

Country Link
CN (1) CN110633277B (zh)
WO (1) WO2021027137A1 (zh)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111639072A (zh) * 2020-04-14 2020-09-08 北京中交兴路信息科技有限公司 一种车联网场景下的数据存储方法及系统、可读存储介质
CN111651338B (zh) * 2020-05-28 2023-06-27 杭州迪普科技股份有限公司 日志格式化时间的获取系统及其方法
CN112000886B (zh) * 2020-08-14 2024-02-02 北京百度网讯科技有限公司 用户行为记录方法、装置、设备和可读存储介质
CN112286969B (zh) * 2020-10-29 2024-03-15 广州汽车集团股份有限公司 低频数据连续性确定方法和装置
CN112614519A (zh) * 2021-01-05 2021-04-06 中国人民解放军国防科技大学 基于时序逻辑的大数据压缩方法及复盘播放方法
CN113792036A (zh) * 2021-01-15 2021-12-14 北京沃东天骏信息技术有限公司 业务数据处理方法、装置、介质及电子设备
CN112835518B (zh) * 2021-01-27 2023-04-25 歌尔科技有限公司 数据存储方法、装置、可穿戴设备及存储介质
CN113032453B (zh) * 2021-02-25 2024-03-01 广州虎牙科技有限公司 数据的存储和解压方法、装置、电子设备及存储介质
CN113078908B (zh) * 2021-03-10 2022-03-25 杭州又拍云科技有限公司 一种适用于时间序列数据库的简易编解码方法
CN113190176B (zh) * 2021-05-11 2023-01-10 上海华东汽车信息技术有限公司 一种数据存储方法、装置、电子设备及存储介质
CN113342817A (zh) * 2021-06-23 2021-09-03 蘑菇物联技术（深圳）有限公司 数据降采样方法、装置、系统与计算机可读存储介质
CN113596624A (zh) * 2021-07-21 2021-11-02 青岛海尔科技有限公司 热水器数据的处理方法、装置、存储介质和电子装置
CN113821178A (zh) * 2021-11-23 2021-12-21 中航金网(北京)电子商务有限公司 一种数据处理方法及系统
CN117278046A (zh) * 2023-09-18 2023-12-22 武汉含秋数据科技有限公司 时序数据压缩存储方法、装置、电子设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180121516A1 (en) * 2016-11-02 2018-05-03 Fujitsu Limited Information processing device, computer-readable recording medium having stored therein information processing program, and information processing method
CN109074669A (zh) * 2018-07-06 2018-12-21 深圳市汇顶科技股份有限公司 数据压缩方法和装置
CN109739819A (zh) * 2019-01-15 2019-05-10 北京智融时代信息技术有限公司 可回溯的快照无损压缩方法、装置、设备及可读存储介质
CN109787638A (zh) * 2019-01-10 2019-05-21 杭州幻方科技有限公司 一种数据压缩存储处理装置及方法
CN109981111A (zh) * 2019-03-28 2019-07-05 杭州即得科技有限公司 一种数据压缩编码的方法、装置及存储电子设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4868570A (en) * 1988-01-15 1989-09-19 Arthur D. Little, Inc. Method and system for storing and retrieving compressed data
CN103577456B (zh) * 2012-07-31 2016-12-21 国际商业机器公司 用于处理时序数据的方法和装置
EP2698679A1 (en) * 2012-08-16 2014-02-19 Siemens Aktiengesellschaft System and method for compressing production data stream and filtering compressed data with different criteria.
IN2014CH00896A (zh) * 2014-02-24 2015-08-28 Inmobi Pte Ltd

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180121516A1 (en) * 2016-11-02 2018-05-03 Fujitsu Limited Information processing device, computer-readable recording medium having stored therein information processing program, and information processing method
CN109074669A (zh) * 2018-07-06 2018-12-21 深圳市汇顶科技股份有限公司 数据压缩方法和装置
CN109787638A (zh) * 2019-01-10 2019-05-21 杭州幻方科技有限公司 一种数据压缩存储处理装置及方法
CN109739819A (zh) * 2019-01-15 2019-05-10 北京智融时代信息技术有限公司 可回溯的快照无损压缩方法、装置、设备及可读存储介质
CN109981111A (zh) * 2019-03-28 2019-07-05 杭州即得科技有限公司 一种数据压缩编码的方法、装置及存储电子设备

Also Published As

Publication number Publication date
CN110633277B (zh) 2024-05-03
CN110633277A (zh) 2019-12-31

Similar Documents

Publication Publication Date Title
WO2021027137A1 (zh) 时序数据存储方法、装置、计算机设备和存储介质
CN110209652B (zh) 数据表迁移方法、装置、计算机设备和存储介质
US20180365264A1 (en) Telemetry system for a cloud synchronization system
CN109542889B (zh) 流式数据列存储方法、装置、设备和存储介质
CN110795499B (zh) 基于大数据的集群数据同步方法、装置、设备及存储介质
CN112910945A (zh) 请求链路跟踪方法和业务请求处理方法
TW201800967A (zh) 分布式流式資料處理的方法和裝置
CN110908778B (zh) 任务部署方法、系统和存储介质
CN110569222B (zh) 链路追踪方法、装置、计算机设备和可读存储介质
CN110825820A (zh) 实时数据标签获取方法、装置、计算机设备和存储介质
CN110634052A (zh) 分布式架构生成订单号的方法及装置
CN112311902B (zh) 基于微服务的文件发送方法及装置
CN111475324A (zh) 日志信息的分析方法、装置、计算机设备和存储介质
WO2017097159A1 (zh) 一种随机字符串生成方法及装置
CN110647531A (zh) 数据同步方法、装置、设备及计算机可读存储介质
WO2021068891A1 (en) Method, system, electronic device, and storage medium for storing and collecting temperature data
CN110275703B (zh) 键值对数据的赋值方法、装置、计算机设备和存储介质
CN110716916A (zh) 资源在线管理方法、装置、计算机设备和存储介质
CN110442439B (zh) 任务进程处理方法、装置和计算机设备
WO2021114613A1 (zh) 基于人工智能的故障节点识别方法、装置、设备和介质
CN111046240B (zh) 网关流量统计方法、装置、计算机设备和存储介质
CN108829534B (zh) 数据问题修复方法、装置、计算机设备和存储介质
CN104506636A (zh) 一种数据同步方法及装置
CN116910820A (zh) 数据报表处理方法、装置、计算机设备以及存储介质
CN111198916A (zh) 一种数据传输方法、装置，电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19941011

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19941011

Country of ref document: EP

Kind code of ref document: A1