Node task processing method, device and electronic device based on bloodline graph

Abstract

The embodiments of the present disclosure disclose a node task processing method, device and electronic device based on a bloodline map. A specific implementation of the method includes: determining a second node group sequence; for the second node group, performing the following processing steps: determining the second task information corresponding to each second node in the second node group; determining the real-time path status information corresponding to the second node group; performing dynamic task decomposition on the node task corresponding to the first node according to the obtained real-time path status information set and the first task information corresponding to the first node to generate a third task information set; for the third task information, in response to at least one second sub-node group data corresponding to the third task information, processing the sub-node task corresponding to the third task information; generating the node task result corresponding to the first node according to the obtained sub-node task result set and the third task information set. This implementation greatly improves the efficiency of computing resource utilization.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of computer technology, and in particular to a node task processing method, device and electronic device based on a bloodline graph. Background Art

[0002] A lineage graph is a data structure that can describe the node relationships between lineage nodes. In practice, when processing complex computing tasks, there are often dependencies between computing tasks. Therefore, combining a lineage graph can well describe the dependencies between computing tasks.

[0003] However, when combining the bloodline graph to process computing tasks, the following technical problems were found:

[0004] Since there are dependencies between computing tasks, that is, the progress of the previous computing task will affect the task processing of the subsequent computing tasks with which there is a dependency, the task execution of the subsequent computing tasks cannot be effectively controlled, which may cause computing resources to be occupied for a long time due to the failure of computing tasks to be completed, affecting the efficiency of computing resource utilization.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose a node task processing method, device and electronic device based on a bloodline graph to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a node task processing method based on a bloodline graph, the method comprising: determining a second node group sequence, wherein the second node in the second node group sequence is a bloodline node having a reverse bloodline dependency with the first node, the first node is a bloodline node of a node task to be executed, and each second node in the second node group is located on the same bloodline path; for each second node group in the second node group sequence, performing the following processing steps: determining second task information corresponding to each second node in the second node group, wherein the second task information represents a real-time task description of the node task corresponding to the second node, and the second task information comprises: task description information, task processing status, task start time and task predicted end time; determining, according to the second task information group corresponding to the second node group, Determine the real-time path status information corresponding to the above-mentioned second node group; according to the obtained real-time path status information set and the first task information corresponding to the above-mentioned first node, dynamically decompose the node task corresponding to the above-mentioned first node to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above-mentioned first node, and the third task information in the above-mentioned third task information set corresponds to at least one second sub-node group; for each third task information in the above-mentioned third task information set, in response to at least one second sub-node group data corresponding to the above-mentioned third task information, process the sub-node task corresponding to the above-mentioned third task information to generate a sub-node task result; according to the obtained sub-node task result set and the above-mentioned third task information set, generate the node task result corresponding to the above-mentioned first node.

[0009] In a second aspect, some embodiments of the present disclosure provide a node task processing device based on a bloodline map, the device comprising: a determination unit, configured to determine a second node group sequence, wherein the second node in the above-mentioned second node group sequence is a bloodline node that has a reverse bloodline dependency with the first node, the above-mentioned first node is a bloodline node of the node task to be executed, and each second node in the second node group is located on the same bloodline path; an execution unit, configured to perform the following processing steps for each second node group in the above-mentioned second node group sequence: determine the second task information corresponding to each second node in the above-mentioned second node group, wherein the second task information represents a real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time; determine the above-mentioned second node according to the second task information group corresponding to the above-mentioned second node group. real-time path status information corresponding to the group; a dynamic task decomposition unit, configured to dynamically decompose the node task corresponding to the above-mentioned first node according to the obtained real-time path status information set and the first task information corresponding to the above-mentioned first node, so as to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above-mentioned first node, and the third task information in the above-mentioned third task information set corresponds to at least one second sub-node group; a processing unit, configured to, for each third task information in the above-mentioned third task information set, process the sub-node task corresponding to the above-mentioned third task information in response to at least one second sub-node group data corresponding to the above-mentioned third task information, so as to generate a sub-node task result; a generating unit, configured to generate a node task result corresponding to the above-mentioned first node according to the obtained sub-node task result set and the above-mentioned third task information set.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0012] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the node task processing method based on the bloodline graph of some embodiments of the present disclosure, the utilization efficiency of computing resources is improved. Specifically, the reason for the low utilization efficiency of computing resources is that: since there is a dependency relationship between computing tasks, that is, the progress of the previous computing task will affect the task processing of the subsequent computing tasks with which there is a dependency relationship, thus, the task execution of the subsequent computing tasks cannot be effectively controlled, which may cause the computing resources to be occupied for a long time due to the failure of the computing tasks to be completed, affecting the utilization efficiency of computing resources. Based on this, the node task processing method based on the bloodline graph of the present disclosure, first, determines the second node group sequence, wherein the second node in the above-mentioned second node group sequence is a bloodline node that has a reverse bloodline dependency with the first node, the above-mentioned first node is a bloodline node of the node task to be executed, and each second node in the second node group is located on the same bloodline path. In this way, the bloodline node that affects the node task corresponding to the first node is obtained. Secondly, for each second node group in the above second node group sequence, the following processing steps are performed: the first step is to determine the second task information corresponding to each second node in the above second node group, wherein the second task information represents the real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time. In this way, the task characterization of the node task corresponding to the second node from four dimensions of task description information, task processing status, task start time and task predicted end time is obtained. The second step is to determine the real-time path state information corresponding to the above second node group according to the second task information group corresponding to the above second node group. Since there is a blood relationship between the blood nodes, the overall processing progress of the node task corresponding to the blood path is characterized in combination with each blood node on the blood path. Further, according to the obtained real-time path state information set and the first task information corresponding to the above first node, the node task corresponding to the above first node is dynamically decomposed to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above first node, and the third task information in the above third task information set corresponds to at least one second sub-node group. Next, for each third task information in the third task information set, in response to at least one second sub-node group data corresponding to the third task information, the sub-node task corresponding to the third task information is processed to generate a sub-node task result. In practice, since there are multiple bloodline nodes on the bloodline path that affect the node task of the first node, in order to avoid long-term occupation of computing resources caused by waiting, the computing tasks are split and executed independently through task decomposition to improve the efficiency of computing resource utilization. Finally, based on the obtained sub-node task result set and the third task information set, the node task result corresponding to the first node is generated. The above method greatly improves the efficiency of computing resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the node task processing method based on the bloodline graph according to the present disclosure;

[0015] Figure 2 It is a schematic diagram of the graph structure of the bloodline graph;

[0016] Figure 3 is a schematic diagram of a first distribution map;

[0017] Figure 4 is a schematic diagram of a model structure of a real-time path state information generation model;

[0018] Figure 5 It is a structural schematic diagram of some embodiments of the node task processing device based on the blood relationship graph according to the present disclosure;

[0019] Figure 6 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0020] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0021] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0022] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0023] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0024] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0025] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0026] refer to Figure 1 , shows a process 100 of some embodiments of the node task processing method based on the bloodline map according to the present disclosure. The node task processing method based on the bloodline map includes the following steps:

[0027] Step 101: determine a second node group sequence.

[0028] In some embodiments, the execution subject (e.g., computing device) of the node task processing method based on the lineage graph may determine a second node group sequence, wherein the second node in the second node group sequence is a lineage node having a reverse lineage dependency with the first node, the first node is a lineage node of the node task to be executed, and each second node in the second node group is located on the same lineage path.

[0029] As an example, see Figure 2 The diagram of the structure of the blood relationship map is shown in FIG. Figure 2Including: bloodline node A1, bloodline node A2, bloodline node A3, bloodline node A4, bloodline node A5, bloodline node A6, bloodline node A7, bloodline node A8, bloodline node A9, bloodline node A10, bloodline node A11, bloodline node A12, bloodline node A13, bloodline node A14, bloodline node A15. Among them, the first node can be bloodline node A15. The second node group sequence can be [second node group G1, second node group G2, second node group G3, second node group G4, second node group G5, second node group G6, second node group G7, second node group G8, second node group G9]. The second node group G1 can be [bloodline node A1, bloodline node A8, bloodline node A12]. The second node group G2 can be [bloodline node A2, bloodline node A8, bloodline node A12]. The second node group G3 can be [bloodline node A3, bloodline node A7, bloodline node A11]. The second node group G4 can be [bloodline node A3, bloodline node A8, bloodline node A12]. The second node group G5 can be [bloodline node A4, bloodline node A9, bloodline node A12]. The second node group G6 can be [bloodline node A5, bloodline node A10, bloodline node A12]. The second node group G7 can be [bloodline node A5, bloodline node A10, bloodline node A13]. The second node group G8 can be [bloodline node A5, bloodline node A10, bloodline node A14]. The second node group G9 can be [bloodline node A6, bloodline node A10, bloodline node A14].

[0030] It should be noted that the above-mentioned computing device can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here. It should be understood that the number of computing devices can be any number according to the implementation needs.

[0031] In some optional implementations of some embodiments, the execution subject determining the second node group sequence may include the following steps:

[0032] The first step is to determine the first bloodline path set.

[0033] The number of first lineage paths in the first lineage path set is, in an initial state, the node in-degree value of the first node.

[0034] As an example, further reference Figure 2, when the first node is the bloodline node A15, the node indegree value of the first node is 4. Therefore, the number of first bloodline paths in the first bloodline path set is 4. The first bloodline path set can be [first bloodline path R1, first bloodline path R2, first bloodline path R3, first bloodline path R4]. Among them, the first bloodline path R1 can be the bloodline path between the bloodline node A11 and the bloodline node A15. The first bloodline path R2 can be the bloodline path between the bloodline node A12 and the bloodline node A15. The first bloodline path R3 can be the bloodline path between the bloodline node A13 and the bloodline node A15. The first bloodline path R4 can be the bloodline path between the bloodline node A14 and the bloodline node A15.

[0035] In the second step, according to the first bloodline path set, the following second node group determination steps are performed:

[0036] The first sub-step is to take out a bloodline path from the first bloodline path set as the target bloodline path.

[0037] In practice, the execution subject may randomly or sequentially select a lineage path from the first lineage path set as the target lineage path. For example, the target lineage path may be the first lineage path R2.

[0038] The second sub-step is to perform reverse path traversal on the target bloodline path.

[0039] In practice, the execution subject may traverse the target lineage path in reverse along the target lineage path. For example, taking the target lineage path as the first lineage path R2, the execution subject may take the lineage node A15 as the starting node and traverse the lineage path including the target lineage path in reverse along the target lineage path.

[0040] The third sub-step is, in response to traversing to a bloodline node whose node indegree value is greater than 1, creating a first bloodline path with the above bloodline node as the starting point and adding it to the first bloodline path set.

[0041] For example, when traversing the first bloodline path R2 in reverse, the node in-degree value of bloodline node A12 is 3, so the bloodline path between bloodline node A8 and bloodline node A12 can be used as the first bloodline path, and the bloodline path between bloodline node A12 and bloodline node A9 can be used as the first bloodline path. At this time, the execution subject further traverses in reverse along bloodline node A12 and bloodline node A10.

[0042] The fourth sub-step is to update the node counter corresponding to the target lineage path in response to traversing to a lineage node whose corresponding task processing status is a task completion status.

[0043] The node counter is initially 0. In practice, the above execution subject can update the node counter by accumulation.

[0044] In the fifth sub-step, in response to the count value of the node counter being greater than or equal to a pre-designed value or the node in-degree value of the traversed bloodline node being 0, the bloodline nodes traversed along the target bloodline path are taken as the second node group.

[0045] For example, since the execution subject traverses reversely along the lineage node A12, the lineage node A10, and the lineage node A6, and the node in-degree value of the lineage node A6 is 0, the above execution subject can use the lineage node A12, the lineage node A10, and the lineage node A6 as the second node group.

[0046] The sixth sub-step is, in response to the first lineage path set excluding the target lineage path being empty, ending the second node group determination step.

[0047] In the third step, in response to the first lineage path set excluding the target lineage path being non-empty, the first lineage path set excluding the target lineage path is used as the first lineage path set, and the second node group determination step is performed again.

[0048] Step 102: for each second node group in the second node group sequence, perform the following processing steps:

[0049] Step 1021: Determine the second task information corresponding to each second node in the second node group.

[0050] In some embodiments, the execution subject may determine the second task information corresponding to each second node in the second node group. The second task information represents the real-time task description of the node task corresponding to the second node. The second task information includes: task description information, task processing status, task start time and task predicted end time. In practice, the task description information is a task description for the node task corresponding to the second task information. For example, the task description information may include but is not limited to: the data source involved in the node task, the task calculation logic of the node task. The task processing status represents the current state of the task corresponding to the node task of the second node. For example, the task processing status may include but is not limited to: node task processing, node task not started, node task completed, node task abnormal. The task start time represents the actual task start time of the node task corresponding to the second node. The task predicted end time represents the predicted task end time of the node task corresponding to the second node. In practice, the execution subject may send a node task information acquisition request to the second node, wherein the node task information acquisition request may be a request for acquiring node task information from the second node. The execution subject may use the response to the node task information acquisition request returned by the second node as the second task information corresponding to the second node.

[0051] In some optional implementations of some embodiments, the execution subject determines the second task information corresponding to each second node in the second node group, which may include the following steps:

[0052] The first step is to determine a task description for the node task corresponding to the second node as task description information included in the second task information corresponding to the second node.

[0053] In practice, first, the execution subject may communicate with the second node to determine whether the second node is in an active state. Second, in response, the execution subject may pull the task description information included in the second task information corresponding to the second node from the second node.

[0054] The second step is to determine the actual task start time of the node task corresponding to the second node as the task start time included in the second task information corresponding to the second node.

[0055] In practice, the execution subject can determine the time when the second node switches from the inactive state to the active state as the task start time included in the second task information corresponding to the second node. Specifically, when the second node has not started to execute the corresponding node task, it will be in a sleep state. Therefore, the task start time when the second node starts to execute the node task can be determined by judging the time when the second node switches from the sleep state to the active state.

[0056] The third step is to determine the current task state of the node task corresponding to the second node as the task processing state included in the second task information corresponding to the second node.

[0057] In practice, when the second node starts to execute the node task, the current task status will be sent to the node task status list in real time for storage, wherein the node task status list is used to store the task status when different second nodes execute the node task. Therefore, the execution subject can determine the current task status of the node task corresponding to the second node from the node task status list through the node identifier corresponding to the second node, as the task processing status included in the second task information corresponding to the second node.

[0058] The fourth step is to obtain a set of historical task information of the second node at a preset time granularity.

[0059] Among them, the historical task information in the above historical task information set includes: task start time and task end time. In practice, the preset time granularity can be a day granularity or a month granularity. In practice, when the node task corresponding to the second node is executed, the corresponding execution process will be recorded in the task log in real time. Therefore, the above execution subject can obtain the historical task information corresponding to each historical node task executed by the second node at the preset time granularity by reading the task log, thereby obtaining the above historical task information set.

[0060] The fifth step is to determine the first distribution map based on the above historical task information set.

[0061] Among them, the above-mentioned first distribution graph represents the duration distribution of the task running time of the historical node tasks corresponding to the historical task information. In practice, task running time = task end time - task start time. Therefore, firstly, for each historical task information in the historical task information set, the above-mentioned execution subject can use the difference between the task end time and the task start time included in the historical task information as the task running time. Then, based on the obtained task running time set, the above-mentioned first distribution graph is obtained. In practice, the first distribution graph is a histogram. Specifically, the horizontal axis of the first distribution graph corresponds to the task running time, and the vertical axis of the first distribution graph corresponds to the statistical value (number of occurrences) of each running task time.

[0062] As an example, Figure 3 Schematic diagram of the first distribution graph shown, wherein the number of occurrences of a task runtime of 100 seconds in the first distribution graph is 6. The number of occurrences of a task runtime of 125 seconds in the first distribution graph is 14. The number of occurrences of a task runtime of 150 seconds in the first distribution graph is 11. The number of occurrences of a task runtime of 175 seconds in the first distribution graph is 4. The number of occurrences of a task runtime of 200 seconds in the first distribution graph is 2. The number of occurrences of a task runtime of 225 seconds in the first distribution graph is 1. The number of occurrences of a task runtime of 250 seconds in the first distribution graph is 4. The number of occurrences of a task runtime of 275 seconds in the first distribution graph is 2. The number of occurrences of a task runtime of 300 seconds in the first distribution graph is 2. The number of occurrences of a task runtime of 375 seconds in the first distribution graph is 1.

[0063] Step 6: Remove outliers from the first distribution graph to obtain the second distribution graph.

[0064] In practice, the execution subject may propose outliers to the first distribution graph. Figure 3 , the execution entity can remove the tasks with a running time of 375 seconds in the first distribution graph to obtain the second distribution graph.

[0065] The seventh step is to determine the graph boundary of the second distribution graph to obtain the graph boundary.

[0066] In practice, since the second distribution graph is a histogram, the execution entity may connect and smooth the midpoints of every two adjacent columns in the second distribution graph, and connect the smoothed column midpoints to obtain the graph boundary.

[0067] The eighth step is to extract the boundary features of the above-mentioned image boundary to obtain the boundary features.

[0068] In practice, the execution subject may extract the coordinate points on the boundary of the graph with a preset step length as the boundary feature. Specifically, 0≤preset step length≤the number of columns included in the second distribution graph. The preset step length is a positive integer.

[0069] The ninth step is to determine the distribution type corresponding to the first distribution graph according to the boundary features.

[0070] In practice, the distribution type can characterize the probability distribution type corresponding to the first distribution graph. Specifically, the distribution type may include, but is not limited to: normal distribution type, uniform distribution type, exponential distribution type, log-normal distribution type. The above-mentioned execution subject can determine the distribution type through a distribution type classifier. Among them, the distribution type classifier includes 3 serially connected convolutional layers and a multi-classifier. Since the feature dimension of the boundary feature is short, a shorter convolutional layer structure is used for feature extraction, which can reduce the amount of feature processing and avoid feature forgetting.

[0071] In the tenth step, interval estimation is performed based on the above distribution type and the above first distribution graph to determine the predicted execution time of the task.

[0072] In practice, the execution subject may perform interval estimation on the first distribution graph based on the probability density function corresponding to the distribution type under a preset confidence level to determine the predicted execution time of the task. Specifically, the preset confidence level may be 95%.

[0073] In the eleventh step, based on the predicted execution time of the task and the task start time included in the second task information corresponding to the second node, the predicted task end time included in the second task information corresponding to the second node is obtained.

[0074] In practice, the predicted end time of a task = the start time of the task + the predicted execution time of the task.

[0075] The above-mentioned "in some optional implementations of some embodiments" is an inventive point of the present disclosure. By combining the historical task information set to generate a first distribution map, the corresponding distribution type is automatically identified according to the boundary features corresponding to the first distribution map. And interval estimation is performed in combination with the distribution type. In this way, there is no need to determine the distribution type manually, which greatly improves the efficiency of determining the predicted end time of the task.

[0076] Step 1022: Determine the real-time path state information corresponding to the second node group according to the second task information group corresponding to the second node group.

[0077] In some embodiments, the execution subject may determine the real-time path status information corresponding to the second node group according to the second task information group corresponding to the second node group, wherein the real-time path status information is used to characterize the overall node task completion status of the bloodline path corresponding to the second node group.

[0078] In some optional implementations of some embodiments, the execution subject determines the real-time path state information corresponding to the second node group according to the second task information group corresponding to the second node group, including:

[0079] In the first step, for each second task information in the second task information group corresponding to the second node group, task information features are extracted from the second task information according to the task processing status included in the second task information and the node distance between the second node corresponding to the second task information and the first node to obtain the task information features.

[0080] The second step is to generate real-time path status information corresponding to the second node group according to the obtained task information feature sequence.

[0081] In practice, the execution subject can generate the real-time path status information through a real-time path status information generation model. For details, see Figure 4 The model structure diagram of the real-time path state information generation model shown in FIG. 4 is a schematic diagram of the model structure of the real-time path state information generation model, wherein the backbone network 401 of the real-time path state information generation model includes: M serially connected convolutional layers, wherein: Figure 4The network structure of the backbone network 401 when M=6 is shown, wherein the backbone network 401 includes: a convolutional layer 4011, a convolutional layer 4012, a convolutional layer 4013, a convolutional layer 4014, a convolutional layer 4015, and a convolutional layer 4016. The output of the convolutional layer 4011 is the input of the convolutional layer 4012. The output of the convolutional layer 4012 is the input of the convolutional layer 4013. The output of the convolutional layer 4013 is the input of the convolutional layer 4014. The output of the convolutional layer 4014 is the input of the convolutional layer 4015. The output of the convolutional layer 4015 is the input of the convolutional layer 4016. The outputs of the convolutional layers at even positions except the last convolutional layer in the backbone network 401 are also jumped to the next adjacent convolutional layer through the gating unit. For example, the input of the convolutional layer 4012 is input to the convolutional layer 4014 through the gating unit 4017. The input of the convolution layer 4014 is input to the convolution layer 4016 through the gating unit 4018. Among them, the gating unit 4017 is controlled to open and close by the task processing status included in the second task information corresponding to the convolution layer 4012 and the node distance between the second node corresponding to the second task information and the first node. The gating unit 4018 is controlled to open and close by the task processing status included in the second task information corresponding to the convolution layer 4014 and the node distance between the second node corresponding to the second task information and the first node. Taking the first task information in the second task information group as an example, when the task processing status included in the first task information is not task processing completed, and the node distance between the second node corresponding to the first second task information and the first node is less than or equal to 80% of the number of second information in the second information group, the gating unit is in a closed state, otherwise it is in an open state. In addition, since the second task information in the second task information group has a sequential relationship along the bloodline path, the first task information in the second task information group is the input of the convolution layer 4011. The second task information in the second task information group is the input of the convolution layer 4012. The third task information in the second task information group is the input of the convolution layer 4013. The fourth task information in the second task information group is the input of the convolution layer 4014. The fifth task information in the second task information group is the input of the convolution layer 4015. The sixth task information in the second task information group is the input of the convolution layer 4016. In addition, before the second task information is input into the corresponding convolution layer, it needs to be preprocessed by the corresponding information preprocessing model. Taking the convolution layer 4011 as an example, the first second task information corresponding to the convolution layer 4011 needs to be preprocessed by the information preprocessing model 402. The preprocessing model 4011 includes: a gating unit 4021, a preprocessing layer 4022, a preprocessing layer 4023 and a preprocessing layer 4024. The gating unit 4021 is controlled by the task processing state included in the second information.When the task processing status indicates that the task processing is completed, the gate control unit 4021 is in the open state, that is, the first second task information corresponding to the convolution layer 4011 does not need to be preprocessed by the information preprocessing model 4022, that is, the first second task information does not participate in the feature extraction of the backbone network 401. Specifically, the preprocessing layer 4022 includes an encoding layer and a decoding layer for extracting features from the task description information included in the first second task information. The preprocessing layer 4023 is used to map the task processing time included in the first second task information to a timestamp. The preprocessing layer 4024 is used to map the task prediction end time included in the first second task information to a timestamp. Then, the preprocessing model 4011 splices the outputs of the preprocessing layer 4022, the preprocessing layer 4023 and the preprocessing layer 4024 as the input of the convolution layer 4011. The convolution layer 4016 also includes a multi-classifier for classifying and obtaining the path state information and the corresponding confidence.

[0082] Step 103 : dynamically decompose the node task corresponding to the first node according to the obtained real-time path state information set and the first task information corresponding to the first node, so as to generate a third task information set.

[0083] In some embodiments, the execution subject may dynamically decompose the node task corresponding to the first node according to the obtained real-time path state information set and the first task information corresponding to the first node to generate a third task information set, wherein the third task information is a task description of the decomposed sub-node task.

[0084] In some optional implementations of some embodiments, the execution subject dynamically decomposes the node task corresponding to the first node according to the obtained real-time path state information set and the first task information corresponding to the first node to generate a third task information set, including:

[0085] The first step is to decompose the node task of the first node into atomic tasks according to the first task information to obtain an atomic task set.

[0086] In practice, the above-mentioned execution subject can combine the task description information included in the first task information to split the node task corresponding to the first node into at least one meta-subtask to obtain a set of meta-subtasks. Among them, the meta-subtask refers to a task unit that cannot be further split into tasks. For example, the task description information of the first node can be "Indicator A = (Indicator B + Indicator C) × Indicator D", so the meta-subtask can be "Indicator B + Indicator C", "Indicator D", "Indicator B × Indicator D", "Indicator C × Indicator D".

[0087] In the second step, based on the real-time path state information set, the atomic tasks in the atomic task set are dynamically combined to obtain a target task set, wherein the target task is an atomic task or an intermediate task composed of at least two atomic tasks.

[0088] In practice, the above execution entities can be dynamically combined according to the real-time path status information of the lineage path corresponding to the meta-task to obtain the target task set. Specifically, due to the lineage dependency, the execution of the node task of the first node depends on the execution progress of the node task corresponding to the second node, that is, when the node task corresponding to the second node is completed and the corresponding indicator value is calculated, the node task corresponding to the first node can be executed.

[0089] As an example, the real-time path status information corresponding to indicator C indicates that the overall node task completion status of the bloodline path corresponding to the second node group is about to be completed. The real-time path status information corresponding to indicator D indicates that the overall node task completion status of the bloodline path corresponding to the second node group is about to be completed. Therefore, "indicator B × indicator D", "indicator C × indicator D", and "intermediate indicator B + intermediate indicator C" can be used as the target task set.

[0090] In the third step, for each target task in the target task set, the predicted task start time and the predicted task end time of the target task are determined according to the real-time path status information set as the third task information corresponding to the target task.

[0091] In practice, since each target task corresponds to at least one lineage path, each target task corresponds to at least one real-time path status information. Therefore, the execution subject can predict the task prediction start time and task prediction end time of the target task through the at least one real-time path status information corresponding to the target task. Specifically, the time series prediction model can be used as the backbone network and two time predictors, and the at least one real-time path status information corresponding to the target task is used as input to predict the task prediction start time and task prediction end time of the target task. Among them, the time series prediction model can be implemented as the backbone network and the two time predictors using the LSTM (Long Short-Term Memory) model.

[0092] Step 104: for each third task information in the third task information set, in response to at least one second sub-node group data corresponding to the third task information, process the sub-node task corresponding to the third task information to generate a sub-node task result.

[0093] In some embodiments, for each third task information in the third task information set, in response to at least one second sub-node group data corresponding to the third task information being continued, the execution subject may process the sub-node task corresponding to the third task information to generate a sub-node task result. In practice, the execution subject may allocate computing resources related to the third task information to execute the sub-node task corresponding to the third task information to generate a sub-node task result.

[0094] As an example, taking the sub-node task (target node task) "indicator B×indicator D" as an example, the sub-node task result obtained is the product value of "indicator B×indicator D".

[0095] Step 105: Generate a node task result corresponding to the first node according to the obtained sub-node task result set and the third task information set.

[0096] In some embodiments, the execution entity may generate a node task result corresponding to the first node based on the obtained sub-node task result set and the third task information set.

[0097] In practice, since there is an association relationship between the sub-node tasks (target node tasks) corresponding to the third task information in the third task information set, such as the sum of the sub-node task results of "indicator B×indicator D" and the sub-node task results of "indicator C×indicator D" is the node task result "indicator A" of the first node, therefore, the task of summing the sub-node task results of "indicator B×indicator D" and the sub-node task results of "indicator C×indicator D" can be further executed to obtain the node task result corresponding to the first node.

[0098] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the node task processing method based on the bloodline graph of some embodiments of the present disclosure, the utilization efficiency of computing resources is improved. Specifically, the reason for the low utilization efficiency of computing resources is that: since there is a dependency relationship between computing tasks, that is, the progress of the previous computing task will affect the task processing of the subsequent computing tasks with which there is a dependency relationship, thus, the task execution of the subsequent computing tasks cannot be effectively controlled, which may cause the computing resources to be occupied for a long time due to the failure of the computing tasks to be completed, affecting the utilization efficiency of computing resources. Based on this, the node task processing method based on the bloodline graph of the present disclosure, first, determines the second node group sequence, wherein the second node in the above-mentioned second node group sequence is a bloodline node that has a reverse bloodline dependency with the first node, the above-mentioned first node is a bloodline node of the node task to be executed, and each second node in the second node group is located on the same bloodline path. In this way, the bloodline node that affects the node task corresponding to the first node is obtained. Secondly, for each second node group in the above second node group sequence, the following processing steps are performed: the first step is to determine the second task information corresponding to each second node in the above second node group, wherein the second task information represents the real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time. In this way, the task characterization of the node task corresponding to the second node from four dimensions of task description information, task processing status, task start time and task predicted end time is obtained. The second step is to determine the real-time path state information corresponding to the above second node group according to the second task information group corresponding to the above second node group. Since there is a blood relationship between the blood nodes, the overall processing progress of the node task corresponding to the blood path is characterized in combination with each blood node on the blood path. Further, according to the obtained real-time path state information set and the first task information corresponding to the above first node, the node task corresponding to the above first node is dynamically decomposed to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above first node, and the third task information in the above third task information set corresponds to at least one second sub-node group. Next, for each third task information in the third task information set, in response to at least one second sub-node group data corresponding to the third task information, the sub-node task corresponding to the third task information is processed to generate a sub-node task result. In practice, since there are multiple bloodline nodes on the bloodline path that affect the node task of the first node, in order to avoid long-term occupation of computing resources caused by waiting, the computing tasks are split and executed independently through task decomposition to improve the efficiency of computing resource utilization. Finally, based on the obtained sub-node task result set and the third task information set, the node task result corresponding to the first node is generated. The above method greatly improves the efficiency of computing resource utilization.

[0099] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a node task processing device based on a bloodline graph. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the node task processing device based on the bloodline graph can be specifically applied to various electronic devices.

[0100] like Figure 5 As shown, some embodiments of the node task processing device 500 based on the lineage graph include: a determination unit 501, an execution unit 502, a dynamic task decomposition unit 503, a processing unit 504 and a generation unit 505. The determination unit 501 is configured to determine a second node group sequence, wherein the second node in the second node group sequence is a lineage node with a reverse lineage dependency on the first node, the first node is a lineage node of the node task to be executed, and each second node in the second node group is located on the same lineage path; the execution unit 502 is configured to perform the following processing steps for each second node group in the second node group sequence: determine the second task information corresponding to each second node in the second node group, wherein the second task information represents the real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time; determine the real-time path status information corresponding to the second node group according to the second task information group corresponding to the second node group; the dynamic task decomposition unit 505 generates a task. Element 503 is configured to dynamically decompose the node task corresponding to the above-mentioned first node according to the obtained real-time path status information set and the first task information corresponding to the above-mentioned first node, so as to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above-mentioned first node, and the third task information in the above-mentioned third task information set corresponds to at least one second sub-node group; processing unit 504 is configured to, for each third task information in the above-mentioned third task information set, process the sub-node task corresponding to the above-mentioned third task information in response to at least one second sub-node group data corresponding to the above-mentioned third task information, so as to generate a sub-node task result; generation unit 505 is configured to generate the node task result corresponding to the above-mentioned first node according to the obtained sub-node task result set and the above-mentioned third task information set.

[0101] It can be understood that the units recorded in the node task processing device 500 based on the bloodline map are similar to the reference Figure 1Therefore, the operations, features and beneficial effects described above for the method are also applicable to the node task processing device 500 based on the bloodline map and the units contained therein, and will not be repeated here.

[0102] Reference below Figure 6 , which shows a structural schematic diagram of an electronic device (eg, a computing device) 600 suitable for implementing some embodiments of the present disclosure. Figure 6 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0103] like Figure 6 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory 602 or a program loaded from a storage device 608 to a random access memory 603. Various programs and data required for the operation of the electronic device 600 are also stored in the random access memory 603. The processing device 601, the read-only memory 602, and the random access memory 603 are connected to each other via a bus 604. An input / output interface 605 is also connected to the bus 604.

[0104] Typically, the following devices may be connected to the input / output interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 6 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0105] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a read-only memory 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0106] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0107] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0108] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being installed in the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: determines a second node group sequence, wherein the second node in the above-mentioned second node group sequence is a bloodline node that has a reverse bloodline dependency with the first node, the above-mentioned first node is a bloodline node of the node task to be executed, and each second node in the second node group is located on the same bloodline path; for each second node group in the above-mentioned second node group sequence, the following processing steps are performed: determine the second task information corresponding to each second node in the above-mentioned second node group, wherein the second task information represents the real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time; according to the second task corresponding to the above-mentioned second node group Information group, determine the real-time path status information corresponding to the above-mentioned second node group; according to the obtained real-time path status information set and the first task information corresponding to the above-mentioned first node, dynamically decompose the node task corresponding to the above-mentioned first node to generate a third task information set, wherein the third task information corresponds to the sub-node task decomposed by the node task corresponding to the above-mentioned first node, and the third task information in the above-mentioned third task information set corresponds to at least one second sub-node group; for each third task information in the above-mentioned third task information set, in response to at least one second sub-node group data corresponding to the above-mentioned third task information, process the sub-node task corresponding to the above-mentioned third task information to generate a sub-node task result; according to the obtained sub-node task result set and the above-mentioned third task information set, generate the node task result corresponding to the above-mentioned first node.

[0109] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0110] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0111] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The units described may also be set in a processor, for example, may be described as: a processor includes a determination unit, an execution unit, a dynamic task decomposition unit and a generation unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the dynamic task decomposition unit may also be described as "configured to dynamically decompose the node task corresponding to the above-mentioned first node according to the obtained real-time path state information set and the first task information corresponding to the above-mentioned first node, so as to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the above-mentioned first node, and the third task information in the above-mentioned third task information set corresponds to at least one unit of the second sub-node group".

[0112] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0113] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A node task processing method based on a bloodline graph, comprising: Determine a second node group sequence, wherein the second nodes in the second node group sequence are lineage nodes having a reverse lineage dependency with the first node, the first node is a lineage node of a node task to be executed, and each second node in the second node group is located on the same lineage path; For each second node group in the second node group sequence, the following processing steps are performed: Determine second task information corresponding to each second node in the second node group, wherein the second task information represents a real-time task description of a node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time, and task predicted end time; determining, according to the second task information group corresponding to the second node group, real-time path state information corresponding to the second node group; According to the obtained real-time path state information set and the first task information corresponding to the first node, dynamically decompose the node task corresponding to the first node to generate a third task information set, wherein the third task information corresponds to the sub-node task decomposed from the node task corresponding to the first node, and the third task information in the third task information set corresponds to at least one second sub-node group; For each third task information in the third task information set, in response to at least one second sub-node group data corresponding to the third task information being ready, processing the sub-node task corresponding to the third task information to generate a sub-node task result; Generate a node task result corresponding to the first node according to the obtained sub-node task result set and the third task information set, wherein: The determining the second task information corresponding to each second node in the second node group includes: Determine a task description for the node task corresponding to the second node as task description information included in the second task information corresponding to the second node; Determine an actual task start time of the node task corresponding to the second node as the task start time included in the second task information corresponding to the second node; Determine a current task state of the node task corresponding to the second node as the task processing state included in the second task information corresponding to the second node; Acquire a historical task information set of the second node at a preset time granularity, wherein the historical task information in the historical task information set includes: task start time and task end time; Determine a first distribution graph according to the historical task information set, wherein the first distribution graph represents the duration distribution of task running durations of historical node tasks corresponding to the historical task information; Eliminate outliers from the first distribution graph to obtain a second distribution graph; determining a graph boundary of the second distribution graph to obtain a graph boundary; Extracting boundary features of the boundary of the graph to obtain boundary features; Determining, according to the boundary feature, a distribution type corresponding to the first distribution graph; Perform interval estimation according to the distribution type and the first distribution graph to determine the predicted execution time of the task; The task predicted end time included in the second task information corresponding to the second node is obtained according to the task predicted execution time and the task start time included in the second task information corresponding to the second node.

2. The method according to claim 1, wherein: The determining of the second node group sequence comprises: Determine a first lineage path set, wherein the number of first lineage paths in the first lineage path set is the node in-degree value of the first node in an initial state; According to the first lineage path set, the following second node group determination steps are performed: Taking a bloodline path from the first bloodline path set as the target bloodline path; Performing reverse path traversal on the target bloodline path; In response to traversing to a lineage node whose node in-degree value is greater than 1, creating a first lineage path with the lineage node as a starting point, and adding the first lineage path set; In response to traversing to a lineage node whose corresponding task processing state is a task end state, updating a node counter corresponding to the target lineage path, wherein the node counter is initially 0; In response to the count value of the node counter being greater than or equal to a pre-designed value or the node in-degree value of the traversed bloodline node being 0, the bloodline nodes traversed along the target bloodline path are used as the second node group; In response to the first lineage path set excluding the target lineage path being empty, ending the second node group determination step; In response to the first lineage path set excluding the target lineage path being non-empty, the first lineage path set excluding the target lineage path is used as the first lineage path set, and the second node group determining step is performed again.

3. The method according to claim 2, wherein: The determining, according to the second task information group corresponding to the second node group, the real-time path state information corresponding to the second node group includes: For each second task information in the second task information group corresponding to the second node group, according to the task processing state included in the second task information and the node distance between the second node corresponding to the second task information and the first node, extract the task information feature of the second task information to obtain the task information feature; According to the obtained task information feature sequence, real-time path status information corresponding to the second node group is generated.

4. The method according to claim 3, wherein: The step of dynamically decomposing the node task corresponding to the first node according to the obtained real-time path state information set and the first task information corresponding to the first node to generate a third task information set includes: Decomposing the node task of the first node into atomic tasks according to the first task information to obtain an atomic task set; According to the real-time path state information set, dynamically combining the atomic tasks in the atomic task set to obtain a target task set, wherein the target task is an atomic task or an intermediate task composed of at least two atomic tasks; For each target task in the target task set, a task prediction start time and a task prediction end time of the target task are determined according to the real-time path status information set as third task information corresponding to the target task.

5. A node task processing device based on a bloodline graph, comprising: a determining unit configured to determine a second node group sequence, wherein the second nodes in the second node group sequence are lineage nodes having a reverse lineage dependency with the first node, the first node is a lineage node of a node task to be executed, and each second node in the second node group is located on the same lineage path; The execution unit is configured to perform the following processing steps for each second node group in the second node group sequence: determine the second task information corresponding to each second node in the second node group, wherein the second task information represents the real-time task description of the node task corresponding to the second node, and the second task information includes: task description information, task processing status, task start time and task predicted end time; determine the real-time path status information corresponding to the second node group according to the second task information group corresponding to the second node group; a dynamic task decomposition unit, configured to perform dynamic task decomposition on the node task corresponding to the first node according to the obtained real-time path state information set and the first task information corresponding to the first node, so as to generate a third task information set, wherein the third task information corresponds to the sub-node task obtained by decomposing the node task corresponding to the first node, and the third task information in the third task information set corresponds to at least one second sub-node group; a processing unit configured to, for each third task information in the third task information set, process the sub-node task corresponding to the third task information in response to at least one second sub-node group data corresponding to the third task information being ready, so as to generate a sub-node task result; A generating unit is configured to generate a node task result corresponding to the first node according to the obtained sub-node task result set and the third task information set, wherein: The determining the second task information corresponding to each second node in the second node group includes: Determine a task description for the node task corresponding to the second node as task description information included in the second task information corresponding to the second node; Determine an actual task start time of the node task corresponding to the second node as the task start time included in the second task information corresponding to the second node; Determine a current task state of the node task corresponding to the second node as the task processing state included in the second task information corresponding to the second node; Acquire a historical task information set of the second node at a preset time granularity, wherein the historical task information in the historical task information set includes: task start time and task end time; Determine a first distribution graph according to the historical task information set, wherein the first distribution graph represents the duration distribution of task running durations of historical node tasks corresponding to the historical task information; Eliminate outliers from the first distribution graph to obtain a second distribution graph; determining a graph boundary of the second distribution graph to obtain a graph boundary; Extracting boundary features of the boundary of the graph to obtain boundary features; Determining, according to the boundary feature, a distribution type corresponding to the first distribution graph; Perform interval estimation according to the distribution type and the first distribution graph to determine the predicted execution time of the task; The task predicted end time included in the second task information corresponding to the second node is obtained according to the task predicted execution time and the task start time included in the second task information corresponding to the second node.

6. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

7. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images