Intermediate-state timing collaborative computing method and apparatus, and device and storage medium

Abstract

An intermediate-state timing collaborative computing method and apparatus, and a device and a storage medium, which relate to the field of data analysis. The method comprises: by means of a timing scheduling node, generating an ordered node set corresponding to computing nodes, and determining a time period corresponding to each computing node in the ordered node set; determining a target time period corresponding to each piece of timing data to be processed, so as to send each piece of said timing data to a target computing node corresponding to the target time period; by means of the target computing nodes, merging said timing data on the basis of dimension information, so as to obtain several timing aggregation-state objects, and respectively sending the several timing aggregation-state objects to corresponding summarizing nodes on the basis of the dimension information; and merging all the timing aggregation-state objects in the nodes by means of the summarizing nodes, so as to obtain a target aggregation result. Therefore, computing resources can be utilized to the maximum extent, and the performance and stability of aggregation computing are improved.

Description

Intermediate state time sequence coordination calculation method, device and equipment and storage medium

[0001] The present application claims priority to the Chinese patent application No. 202410658563.3, filed on May 27, 2024, and entitled "Intermediate state time sequence coordination calculation method, device and equipment and storage medium", the whole content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of data analysis, in particular to an intermediate state time sequence coordination calculation method, device and equipment and storage medium. BACKGROUND

[0003] The feature of the intermediate state time sequence is that it is an intermediate state data structure with time axis, dimension and result, which can complete the merging and time sequence calculation of summation, average, increment and decrement. Its characteristic is that multiple time sequence intermediate state results can be formed for average, increment and decrement, and multiple time sequence intermediate state results can be segmented and merged. The calculation of a single time sequence intermediate state must ensure the time sequence of the data, and the merging of multiple time sequence intermediate state objects must also ensure the time sequence. On this basis, the accuracy of the calculation can be ensured through the data structure.

[0004] However, in the prior art, when performing aggregation index calculation, grouping calculation is performed according to the dimension, and the calculation of each dimension must be completed by a single node and thread. In addition, for complex time sequence calculation such as increment and decrement, the data needs to be sorted and then linearly calculated in a single dimension. Therefore, when the data of a certain dimension is too much and data skew occurs, the calculation performance and stability will be greatly reduced. SUMMARY

[0005] Therefore, the purpose of the present application is to provide an intermediate state time sequence coordination calculation method, device and equipment and storage medium, which can maximize the use of computing resources based on the innovation of real-time intelligent technology system, and improve the performance and stability of aggregation calculation. The specific scheme is as follows:

[0006] In a first aspect, the present application discloses an intermediate state time sequence coordination calculation method, comprising:

[0007] Generating an ordered node set corresponding to the calculation nodes through the time sequence scheduling node, and determining the time segment corresponding to each calculation node in the ordered node set;

[0008] Determining the target time segment corresponding to each piece of to-be-processed time sequence data, so as to send each piece of to-be-processed time sequence data to the target calculation node corresponding to the target time segment;

[0009] merge, by the target computing node, the to-be-processed time-series data based on dimension information to obtain a plurality of time-series aggregate objects, and send the plurality of time-series aggregate objects to corresponding summary nodes based on the dimension information;

[0010] merge, by the summary node, all time-series aggregate objects within the node to obtain a target aggregation result.

[0011] Optionally, the generating, by the time-series scheduling node, an ordered node set corresponding to the computing nodes comprises:

[0012] receiving, by the time-series scheduling node, node information of the computing nodes to register the computing nodes based on the node information and generate an ordered node set corresponding to the computing nodes based on the node information.

[0013] Optionally, the determining the time segment corresponding to each computing node in the ordered node set comprises:

[0014] based on the node order corresponding to the ordered node set and a preset time range, marking the computing nodes in the ordered node set to determine the time segment responsible by each computing node, and obtaining a first correspondence relationship between each computing node and the time segment.

[0015] Optionally, the determining the target time segment corresponding to each to-be-processed time-series data to send each to-be-processed time-series data to a target computing node corresponding to the target time segment comprises:

[0016] traversing, by the time-series scheduling node, to-be-processed time-series data to determine the target time segment corresponding to the to-be-processed time-series data according to a timestamp corresponding to the time-series data;

[0017] determining, based on the first correspondence relationship, a target computing node corresponding to the target time segment, and sending the to-be-processed time-series data to the target computing node.

[0018] Optionally, the merging, by the target computing node, the to-be-processed time-series data based on dimension information to obtain a plurality of time-series aggregate objects, and sending the plurality of time-series aggregate objects to corresponding summary nodes based on the dimension information comprises:

[0019] sorting, by the target computing node, the to-be-processed time-series data based on time sequence to obtain sorted time-series data;

[0020] determining dimension information corresponding to the sorted time-series data to merge the to-be-processed time-series data based on the dimension information to obtain a plurality of time-series aggregate objects;

[0021] send the several time-series aggregate objects to corresponding summary nodes respectively based on the dimension information.

[0022] Optionally, the sending the several time-series aggregate objects to corresponding summary nodes respectively based on the dimension information comprises:

[0023] performing hash calculation on the dimension information to determine a hash value corresponding to the dimension information;

[0024] determining a second correspondence between the dimension information and the summary nodes through the hash value, to send the several time-series aggregate objects to corresponding summary nodes respectively according to the second correspondence.

[0025] Optionally, the merging, by the summary node, all time-series aggregate objects within the node to obtain a target aggregation result comprises:

[0026] performing sorting on all time-series aggregate objects received by the summary node based on time sequence, and performing merging on the sorted time-series aggregate objects to obtain a target aggregation result; the time-series aggregate objects in the summary node are time-series aggregate objects of the same latitude.

[0027] In a second aspect, the present application discloses an intermediate state time-series coordination computing device, comprising:

[0028] a time segment determination module configured to generate an ordered node set corresponding to a computing node through a time-series scheduling node, and determine a time segment corresponding to each computing node in the ordered node set;

[0029] a computing node determination module configured to determine a target time segment corresponding to each piece of to-be-processed time-series data, to send each piece of the to-be-processed time-series data to a target computing node corresponding to the target time segment;

[0030] an object sending module configured to perform merging on the to-be-processed time-series data based on dimension information through the target computing node, to obtain several time-series aggregate objects, and send the several time-series aggregate objects to corresponding summary nodes respectively based on the dimension information;

[0031] an object merging module configured to perform merging on all time-series aggregate objects within the summary node through the summary node, to obtain a target aggregation result.

[0032] In a third aspect, the present application discloses an electronic device, comprising:

[0033] a memory configured to save a computer program;

[0034] A processor is configured to execute the computer program to implement the intermediate state time sequence coordination calculation method.

[0035] In a fourth aspect, the present application discloses a computer readable storage medium for storing a computer program, which, when executed by a processor, implements the intermediate state time sequence coordination calculation method.

[0036] In the present application, first, an ordered node set corresponding to a computing node is generated by a time sequence scheduling node, and a time segment corresponding to each computing node in the ordered node set is determined, then a target time segment corresponding to each piece of to-be-processed time sequence data is determined, so as to send each piece of to-be-processed time sequence data to a target computing node corresponding to the target time segment, and the target computing node is used to merge the to-be-processed time sequence data based on dimension information, so as to obtain a plurality of time sequence aggregation state objects, and the plurality of time sequence aggregation state objects are respectively sent to a corresponding summary node based on the dimension information, and finally, all time sequence aggregation state objects in the node are merged by the summary node to obtain a target aggregation result. As can be seen, by the method of the present application, an ordered node set of a computing node can be generated by a time sequence scheduling node, and a time segment corresponding to each computing node is determined, then after determining a target time segment corresponding to each piece of to-be-processed time sequence data, each piece of to-be-processed time sequence data can be sent to a corresponding target computing node, so that the target computing node merges the to-be-processed time sequence data based on dimension information, and a plurality of time sequence aggregation state objects obtained based on the dimension information are sent to a corresponding summary node, and then all time sequence aggregation state objects in the node are merged by the summary node to obtain a final target aggregation result. In this way, by performing time sequence coordination processing on the intermediate state result, the concept of time is introduced in addition to the dimension in the calculation process, so that data can be uniformly dispersed to each node for calculation, and the accuracy of calculation is guaranteed. By the dual concepts of dimension and time, the computing resources can be maximally utilized, and the performance and stability of the aggregation calculation are improved. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on the provided drawings.

[0038] Fig. 1 is a flow chart of an intermediate state time sequence coordination calculation method disclosed by the present application;

[0039] Fig. 2 is a time sequence diagram of an intermediate state time sequence coordination calculation method disclosed by the present application;

[0040] Fig. 3 is a schematic diagram of a structure of an intermediate state time sequence coordination calculation device disclosed in the application;

[0041] Fig. 4 is a structure diagram of an electronic device disclosed in the application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the application will be apparently and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0043] In the prior art, when performing aggregation index calculation, grouping calculation is performed according to dimensions, and the calculation of each dimension must be completed by a single node and thread. In addition, for complex time sequence calculation such as increment and decrement, data needs to be sorted, and linear calculation is performed on a single dimension. Therefore, when the data of a certain dimension is too much and data skew occurs, the calculation performance and stability will be greatly reduced.

[0044] In order to overcome the above technical problems, the application provides an intermediate state time sequence coordination calculation method, device, equipment and storage medium. Based on the data structure of the intermediate state of the time sequence, two concepts of time and dimension are introduced for data calculation, so that the data is evenly distributed on each node for parallel calculation, the calculation resources can be maximally utilized, and the performance and stability of the aggregation calculation are improved.

[0045] Referring to Fig. 1, the embodiment of the application discloses an intermediate state time sequence coordination calculation method, comprising:

[0046] In step S11, an ordered node set corresponding to a calculation node is generated by a time sequence scheduling node, and a time segment corresponding to each calculation node in the ordered node set is determined.

[0047] In the embodiment, the ordered node set corresponding to the calculation node needs to be generated by the time sequence scheduling node. Specifically, a time sequence scheduling node can be defined, and all calculation nodes need to complete information registration to the node after starting. Therefore, the node information of the calculation node can be received by the time sequence scheduling node. After all node information is received, all calculation nodes can be registered according to the received node information. After registration is completed, the ordered node set P={P1, P2,...Pn} corresponding to the calculation node can be generated according to the node information. In this way, the time sequence scheduling node can maintain a list storing all calculation node information.

[0048] It needs to be explained that after the ordered node set is generated, the nodes need to be marked in order to determine the time segment corresponding to each computing node in the ordered node set. Specifically, the computing nodes in the ordered node set can be marked based on the node order corresponding to the ordered node set and the preset time range, for example, the nodes are marked in order in a specified time window, such as 1 hour, within the time range of the data, P = {(P1, t1), (P2, t2)... (Pn, tn)}, starting from the first hour, all nodes are marked, node 1 corresponds to 0 point, node 2 corresponds to 1 point, and when the node marking is completed, the corresponding time is marked from the first node again to determine the time segment responsible for each computing node and the first correspondence between each computing node and the time segment.

[0049] Step S12, determining the target time segment corresponding to each piece of to-be-processed time series data to send each piece of to-be-processed time series data to a target computing node corresponding to the target time segment.

[0050] In this embodiment, when all nodes are marked, the time series scheduling node can traverse the time series data to determine the target time segment corresponding to each piece of to-be-processed time series data, and then send each piece of to-be-processed time series data to the corresponding node according to the determined target time segment. Specifically, as shown in FIG. 2, the time series scheduling node can traverse the to-be-processed time series data to determine the target time segment corresponding to the to-be-processed time series data according to the timestamp corresponding to the time series data, and then send the to-be-processed data to the corresponding target computing node according to the first correspondence between each computing node and the time segment. For example, by scanning, it is determined that the computing node corresponding to the timestamp of data 1 and data 2 is a, and then data 1 and data 2 are sent to computing node a.

[0051] Step S13, merging the to-be-processed time series data based on the dimension information by the target computing node to obtain a plurality of time series aggregation state objects, and sending the plurality of time series aggregation state objects to corresponding summary nodes based on the dimension information.

[0052] In this embodiment, each target computing node can merge the time sequence data received by itself to obtain a plurality of time sequence intermediate states. Specifically, as shown in FIG. 2, after receiving all the time sequence data, the four target computing nodes, i.e., the target computing node a, the target computing node b, the target computing node c and the target computing node d, can sort the time sequence data according to the time sequence to obtain sorted time sequence data, and then determine the dimension information corresponding to the sorted time sequence data. It should be noted that the dimension information here is the node that sends the data, and the data sent by the same node can be regarded as being at the same latitude. Then, the time sequence data is merged according to the dimension information to obtain a plurality of time sequence aggregation state objects. It should be noted that since one node can process data of multiple time segments, data of the same dimension and different time segments cannot be merged at present, and only data of the same segment and the same dimension are merged to form a time sequence intermediate state object. In this way, each target computing node contains one or more time sequence aggregation state objects sorted by time. Then, the plurality of time sequence aggregation state objects of each target computing node are respectively sent to the corresponding summary nodes according to the determined dimension information.

[0053] It should be noted that the specific process of sending the plurality of time sequence aggregation state objects to the corresponding summary nodes based on the dimension information is as follows: the dimension information is hashed to determine the hash value corresponding to the dimension information, and then the second correspondence between the dimension information and the summary nodes is determined through the hash value. Each target computing node sends the plurality of time sequence aggregation state objects thereof to the corresponding summary nodes according to the second correspondence. As shown in FIG. 2, the target computing node a and the target computing node d send the plurality of time sequence aggregation state objects thereof to the same summary node, and the target computing node b and the target computing node c send the plurality of time sequence aggregation state objects thereof to the same summary node.

[0054] In step S14, the summary node merges all the time sequence aggregation state objects in the node to obtain a target aggregation result.

[0055] In this embodiment, the summary node can merge all the time sequence aggregation state objects in the node itself to obtain the target aggregation result. Specifically, the summary node needs to sort all the received time sequence aggregation state objects based on time sequence, and then merge the sorted time sequence aggregation state objects to obtain the target aggregation result. After distributing all the time sequence intermediate state objects according to the dimensions to the corresponding nodes, the transmission is the time sequence intermediate state objects rather than the original data. The merging of the time sequence intermediate state of the current node must also ensure the sequential merging of the same dimension. As shown in FIG. 2, the target aggregation results obtained after merging are ad and bc, respectively. In this way, the time sequence intermediate state objects can be used to complete the segmented calculation and secondary merging operation of complex operators such as summation, average, increment, and decrement, effectively reducing the computing pressure of the node and improving the computing efficiency.

[0056] As can be seen, through the method of the present application, an ordered node set of the computing node can be generated by the time sequence scheduling node, and the time segment corresponding to each computing node is determined. Then, after determining the target time segment corresponding to each piece of to-be-processed time sequence data, each piece of to-be-processed time sequence data can be sent to the corresponding target computing node, so that the target computing node merges the to-be-processed time sequence data in the node based on the dimension information, and sends the obtained several time sequence aggregation state objects to the corresponding summary node. Then, the summary node merges all the time sequence aggregation state objects in the node to obtain the final target aggregation result. In this way, on the one hand, by performing time sequence coordination processing on the intermediate state result, the concept of time is introduced in addition to the dimension, so that data can be uniformly dispersed to each node for calculation, and the accuracy of calculation is guaranteed. Through the dual concepts of dimension and time, the computing resources can be maximally utilized, and the performance and stability of the aggregation calculation are improved. On the other hand, the problem of serial calculation of time sequence data of the same dimension can be effectively solved. Through the design of time sequence scheduling, the computing resources can be fully utilized, and the time sequence data can be calculated in parallel while ensuring the accuracy of the calculation. On the other hand, the dependence of time sequence data on dimension data for calculation can be eliminated. Whether the multiple pieces of data are of the same dimension or not, they can be calculated in parallel, maximizing the utilization of program resources while reducing the computing pressure of the single node in the case of data skew.

[0057] Referring to FIG. 3, the present embodiment discloses an intermediate state time sequence coordination calculation device, which comprises:

[0058] The time segment determination module 11 is configured to generate an ordered node set corresponding to the computing node through the time sequence scheduling node, and determine the time segment corresponding to each computing node in the ordered node set.

[0059] The computing node determination module 12 is configured to determine a target time segment corresponding to each piece of the to-be-processed time-series data, so as to send each piece of the to-be-processed time-series data to a target computing node corresponding to the target time segment;

[0060] The object sending module 13 is configured to merge, by the target computing node, the to-be-processed time-series data based on dimension information, to obtain a plurality of time-series aggregate objects, and send the plurality of time-series aggregate objects to corresponding summary nodes based on the dimension information respectively.

[0061] The object merging module 14 is configured to merge, by the summary node, all time-series aggregate objects in the node to obtain a target aggregation result.

[0062] In the present application, firstly, an ordered node set corresponding to the computing nodes is generated by a time-series scheduling node, and a time segment corresponding to each of the computing nodes in the ordered node set is determined, then a target time segment corresponding to each piece of to-be-processed time-series data is determined, so as to send each piece of the to-be-processed time-series data to a target computing node corresponding to the target time segment, the to-be-processed time-series data is merged by the target computing node based on dimension information, to obtain a plurality of time-series aggregate objects, and the plurality of time-series aggregate objects are sent to corresponding summary nodes based on the dimension information respectively, finally, all time-series aggregate objects in the node are merged by the summary node to obtain a target aggregation result. As can be seen, by the method of the present application, an ordered node set of the computing nodes can be generated by a time-series scheduling node, and a time segment corresponding to each of the computing nodes is determined, then after determining a target time segment corresponding to each piece of to-be-processed time-series data, each piece of to-be-processed time-series data can be sent to a corresponding target computing node, so that the target computing node merges the to-be-processed time-series data in the node based on dimension information, and sends a plurality of time-series aggregate objects obtained based on the dimension information to corresponding summary nodes, then all time-series aggregate objects in the node are merged by the summary node to obtain a final target aggregation result. In this way, by performing time-series coordination processing on the intermediate state result, the concept of time is introduced in addition to the dimension in the computing process, so that data can be uniformly dispersed to each node for calculation, and the accuracy of the calculation is guaranteed. By the dual concepts of dimension and time, the computing resources can be maximally utilized, and the performance and stability of the aggregation calculation are improved.

[0063] In some embodiments, the time segment determination module 11 can specifically include:

[0064] The node set generation unit is configured to receive node information of the computing nodes by the time-series scheduling node, register the computing nodes based on the node information, and generate an ordered node set corresponding to the computing nodes based on the node information.

[0065] In some embodiments, the time segment determination module 11 can specifically include:

[0066] a time segment determination unit, configured to mark the computing nodes in the ordered node set based on the node order corresponding to the ordered node set and a preset time range, to determine a time segment responsible for by each of the computing nodes, and obtain a first correspondence between each of the computing nodes and the time segment.

[0067] In some embodiments, the computing node determination module 12 can specifically include:

[0068] a time segment determination unit, configured to traverse the to-be-processed time series data by the timing scheduling node, to determine a target time segment corresponding to the to-be-processed time series data according to a timestamp corresponding to the time series data;

[0069] a data sending unit, configured to determine a target computing node corresponding to the target time segment based on the first correspondence, and send the to-be-processed time series data to the target computing node.

[0070] In some embodiments, the object sending module 13 can specifically include:

[0071] a data sorting submodule, configured to sort the to-be-processed time series data based on time sequence by the target computing node, to obtain sorted time series data;

[0072] a data merging submodule, configured to determine dimension information corresponding to the sorted time series data, to merge the to-be-processed time series data based on the dimension information, to obtain a plurality of time series aggregate objects;

[0073] a data sending submodule, configured to send the plurality of time series aggregate objects to corresponding summary nodes respectively based on the dimension information.

[0074] In some embodiments, the data sending submodule can specifically include:

[0075] a hash calculation unit, configured to perform hash calculation on the dimension information, to determine a hash value corresponding to the dimension information;

[0076] a data sending unit, configured to determine a second correspondence between the dimension information and the summary nodes by the hash value, to send the plurality of time series aggregate objects to the corresponding summary nodes respectively according to the second correspondence.

[0077] In some embodiments, the object merging module 14 can specifically include:

[0078] The object merging unit is configured to sort all the received time-aggregated state objects based on time sequence by the summary node, and merge the sorted time-aggregated state objects to obtain a target aggregation result; the time-aggregated state objects in the summary node are time-aggregated state objects of the same latitude.

[0079] Further, the embodiment of the present application further discloses an electronic device, and FIG. 4 is a structural diagram of the electronic device 20 according to an exemplary embodiment, and the content in the figure should not be considered as any limitation on the use range of the present application.

[0080] FIG. 4 is a structural schematic diagram of an electronic device 20 provided by the embodiment of the present application. The electronic device 20 can specifically include at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 is configured to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the related steps in the intermediate state time sequence coordination calculation method disclosed in any of the preceding embodiments. In addition, the electronic device 20 in the embodiment can be specifically an electronic computer.

[0081] In the embodiment, the power supply 23 is configured to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol followed by the communication interface 24 can be any communication protocol applicable to the technical solution of the present application, which is not specifically limited here; the input / output interface 25 is configured to obtain external input data or output data to the outside, and the specific interface type can be selected according to the specific application needs, which is not specifically limited here.

[0082] In addition, the memory 22 as a carrier for resource storage can be a read-only memory, a random access memory, a magnetic disk or an optical disk, etc., and the resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage mode can be temporary storage or permanent storage.

[0083] The operating system 221 is configured to manage and control each hardware device on the electronic device 20 and the computer program 222, and can be Windows Server, Netware, Unix, Linux, etc. In addition to the computer program capable of completing the intermediate state time sequence coordination calculation method executed by the electronic device 20 disclosed in any of the preceding embodiments, the computer program 222 can further include a computer program capable of completing other specific work.

[0084] Further, the application also discloses a computer readable storage medium for storing a computer program, wherein the computer program is executed by a processor to implement the intermediate state timing coordination calculation method disclosed above. For the specific steps of the method, refer to the corresponding content disclosed in the foregoing embodiments, which will not be repeated here.

[0085] The various embodiments are described in a progressive manner in the specification, and each embodiment focuses on the difference from other embodiments. For the same or similar parts between the various embodiments, refer to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and refer to the method part for the relevant content.

[0086] The skilled person can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in the above description in general terms. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the application.

[0087] The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly implemented by hardware, a software module executed by a processor, or a combination of both. The software module can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0088] Finally, it should be noted that, in this document, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0089] The technical solutions provided by the present application are described in detail above, and the principles and implementation manners of the present application are described by using specific examples. The above description of the examples is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed, and the above description of the content of the specification should not be understood as a limitation on the present application.

Claims

1. A method for intermediate-state timing coordination computation, characterized in that, The method comprises the following steps: generating, by a time sequence scheduling node, an ordered node set corresponding to a computing node, and determining a time segment corresponding to each computing node in the ordered node set; determining a target time segment corresponding to each piece of to-be-processed time sequence data, so as to send each piece of to-be-processed time sequence data to a target computing node corresponding to the target time segment; merging, by the target computing node, the to-be-processed time sequence data based on dimension information, so as to obtain a plurality of time sequence aggregation state objects, and sending the plurality of time sequence aggregation state objects to corresponding summary nodes based on the dimension information; merging, by the summary node, all time sequence aggregation state objects in the node, so as to obtain a target aggregation result.

2. The intermediate state timing coordination calculation method of claim 1, wherein, The generating, by a time sequence scheduling node, an ordered node set corresponding to a computing node comprises: receiving, by a time sequence scheduling node, node information of a computing node, so as to register the computing node based on the node information, and generating an ordered node set corresponding to the computing node based on the node information.

3. The intermediate state timing coordination calculation method of claim 1, wherein, The determining a time segment corresponding to each computing node in the ordered node set comprises: based on the node order corresponding to the ordered node set and a preset time range, marking the computing nodes in the ordered node set, so as to determine a time segment responsible by each computing node, and obtaining a first correspondence relationship between each computing node and the time segment.

4. The intermediate state timing coordination calculation method of claim 3, wherein, The determining a target time segment corresponding to each piece of to-be-processed time sequence data, so as to send each piece of to-be-processed time sequence data to a target computing node corresponding to the target time segment comprises: traversing, by the time sequence scheduling node, to-be-processed time sequence data, so as to determine a target time segment corresponding to the to-be-processed time sequence data according to a time stamp corresponding to the time sequence data; determining a target computing node corresponding to the target time segment based on the first correspondence relationship, and sending the to-be-processed time sequence data to the target computing node.

5. The meso-state timing coordination computation method of claim 1, wherein, The merging, by the target computing node, the to-be-processed time sequence data based on dimension information, so as to obtain a plurality of time sequence aggregation state objects, and sending the plurality of time sequence aggregation state objects to corresponding summary nodes based on the dimension information comprises: sorting, by the target computing node, the to-be-processed time sequence data based on time sequence, so as to obtain sorted time sequence data; determining dimension information corresponding to the sorted time sequence data, so as to merge the to-be-processed time sequence data based on the dimension information, so as to obtain a plurality of time sequence aggregation state objects; sending the plurality of time sequence aggregation state objects to corresponding summary nodes based on the dimension information.

6. The intermediate state timing coordination calculation method of claim 5, wherein, The sending the plurality of time sequence aggregation state objects to corresponding summary nodes based on the dimension information comprises: performing hash calculation on the dimension information, so as to determine a hash value corresponding to the dimension information; determining a second correspondence relationship between the dimension information and the summary node based on the hash value, so as to send the plurality of time sequence aggregation state objects to corresponding summary nodes according to the second correspondence relationship.

7. The meso-state timing coordination calculation method according to any one of claims 1 to 6, characterized in that, The merging, by the summary node, of all the in-node time-series aggregate objects to obtain a target aggregation result comprises: The summary node sorts all the received time-series aggregate objects in time sequence and merges the sorted time-series aggregate objects to obtain a target aggregation result; the time-series aggregate objects in the summary node are time-series aggregate objects of the same latitude.

8. An intermediate state timing coordination computing device, characterized by, The method comprises: a time segment determination module configured to generate, by a time-series scheduling node, an ordered node set corresponding to a computing node and determine a time segment corresponding to each computing node in the ordered node set; a computing node determination module configured to determine a target time segment corresponding to each piece of to-be-processed time-series data so as to send each piece of the to-be-processed time-series data to a target computing node corresponding to the target time segment; an object sending module configured to merge, by the target computing node, the to-be-processed time-series data based on dimension information to obtain a plurality of time-series aggregate objects and send the plurality of time-series aggregate objects to corresponding summary nodes based on the dimension information; an object merging module configured to merge, by the summary node, all the in-node time-series aggregate objects to obtain a target aggregation result.

9. An electronic device, comprising: The method comprises: a memory configured to save a computer program; a processor configured to execute the computer program to implement the intermediate-state time-series coordination calculation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is saved and, when executed by a processor, implements the intermediate-state time-series coordination calculation method according to any one of claims 1 to 7.

Images