Parallel world simulation resource dynamic scheduling optimization method and system

Abstract

The application provides a parallel world simulation resource dynamic scheduling optimization method and system. In the method, the dependence relationship between tasks is detected in the parallel world simulation resource to obtain topology data, and the resource competition intensity and the sequence constraint are quantified to form an association feature matrix; a cross-dimension optical signal transmission path is constructed by analyzing the dynamic change of the matrix through a programmable photon array, and a low-conflict resource transmission channel atlas is formed; a global task execution sequence is generated by synchronously processing the matrix and the channel atlas; a resource occupation relationship atlas is generated by performing a multi-world resource occupation operation; the cross-dimension resource conflict probability in the atlas is calculated, and the task execution sequence is dynamically adjusted to avoid the simulation deadlock risk. The application realizes the nanosecond-level dynamic resolution of the cross-dimension resource conflict, the simulation deadlock rate is zeroed, and the task throughput is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dynamic scheduling optimization, in particular to a parallel world simulation resource dynamic scheduling optimization method and system. BACKGROUND

[0002] Parallel world simulation needs to synchronize and coordinate massive heterogeneous computing tasks in multidimensional dynamic environment, facing three core challenges: resource heterogeneity and dynamics (real-time fluctuations of CPU / GPU / FPGA computing units and storage, network link states, need to dynamically match task demand and resource availability); task dependency and timeliness constraints (cross-world interaction events trigger real-time changes in task dependency topology, delays in critical path tasks will cascade to global simulation progress, and critical resources need to be supplied in real time); multi-objective optimization conflicts.

[0003] The current targeted solution is an adaptive scheduling framework based on predicted execution time. This scheme collects dynamic state information of simulation computing nodes in real time, builds a prediction model combining task characteristics, estimates the time consumption of each node executing tasks; based on the prediction results, an adaptive allocation strategy is used to dynamically allocate tasks to the optimal node, and redundant resources are reserved to respond to sudden dependency changes, in order to optimize resource utilization and task completion time. However, the core deficiency is that the prediction model is highly dependent on the quality of historical task data, and is not adaptable to new types of tasks or sudden cross-world interaction events, and real-time state collection in super-large node clusters can easily cause communication storms, leading to increased scheduling delays. SUMMARY

[0004] The present application provides a parallel world simulation resource dynamic scheduling optimization method and system to solve the problems of resource conflict detection delay, cross-dimension deadlock out of control and insufficient task scheduling throughput in the prior art.

[0005] In a first aspect, the present application provides a parallel world simulation resource dynamic scheduling optimization method, comprising:

[0006] Detecting the dependency relationship between tasks in the parallel world simulation resource to obtain task dependency topology data, and quantifying the correlation feature matrix formed between tasks due to resource competition intensity and sequence constraints based on the task dependency topology data;

[0007] Analyzing the dynamic changes of the correlation feature matrix through a programmable photon array to construct a transmission path of cross-dimension optical signals, and forming a low-conflict resource transmission channel map matching the resource competition intensity according to the transmission path;

[0008] Synchronously processing the correlation feature matrix and the low-conflict resource transmission channel map to generate a global task execution sequence;

[0009] based on the global task execution sequence, performing multi-world resource occupation operations and recording resource occupancy states, and generating a multi-world resource occupation relationship graph according to the resource occupancy states;

[0010] Calculate the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship graph, and dynamically adjust the parallel world simulation resource task execution order according to the occurrence probability of cross-dimension resource conflicts to avoid parallel world simulation deadlock risk.

[0011] Optionally, the inter-task dependency relationship in the parallel world simulation resource is detected to obtain task dependency topology data, and an associated feature matrix formed by resource competition intensity and sequence constraints between tasks is quantified based on the task dependency topology data, including:

[0012] Scan the port connection state of each task node in the parallel world simulation resource, and record the physical link of the data transmitted between the task nodes to form task dependency topology data describing the inter-task dependency relationship;

[0013] According to the task dependency topology data, the number of task nodes sharing the same resource is counted, and the number of task nodes sharing the same resource is taken as the quantification value of resource competition intensity;

[0014] Detect the predecessor task and the successor task of the task dependency topology data transmission, and measure the time window length from the completion of the predecessor task to the start of the successor task as the duration value of the sequence constraint;

[0015] Combine the quantification value of the resource competition intensity and the duration value of the sequence constraint into a two-dimensional vector, simultaneously obtain the node arrangement order in the task dependency topology data, and construct all two-dimensional vectors into an associated feature matrix according to the node arrangement order.

[0016] Optionally, the dynamic changes of the associated feature matrix are analyzed by a programmable photon array to construct a transmission path of cross-dimension optical signals, and a low-conflict resource transmission channel graph matching the resource competition intensity is formed according to the transmission path, including:

[0017] The programmable photon array receives the real-time data stream of the associated feature matrix, and maps the row and column coordinates of the associated feature matrix as element values to the spatial light modulation unit of the programmable photon array;

[0018] The spatial light modulation unit dynamically adjusts the offset amount of the optical signal phase by the element value, and maintains a positive proportional relationship between the offset amount and the element value to form a phase distribution map;

[0019] Identify the area where the optical signal phase continuously changes in the phase distribution map, and connect the area to generate a transmission path of cross-dimension optical signals;

[0020] Obtaining the resource contention intensity value of the transmission path coverage area, and converting the resource contention intensity value into an optical signal intensity adjustment coefficient, and loading the optical signal intensity adjustment coefficient to the transmission path;

[0021] Combining the transmission path loaded with the optical signal intensity adjustment coefficient to construct a low-conflict resource transmission channel map.

[0022] Optionally, synchronously processing the correlation feature matrix and the low-conflict resource transmission channel map to generate a global task execution sequence, comprising:

[0023] Combining the resource contention intensity value in the correlation feature matrix and the sequence-constrained duration value into a task node feature vector, and synchronously extracting the optical intensity attribute value of the transmission path of the corresponding task node in the low-conflict resource transmission channel map as a channel feature vector;

[0024] Splicing the task node feature vector and the channel feature vector into a feature combination of the task node, and calculating the feature combination difference of the task node and the adjacent task node;

[0025] Calculating the position score of the task node in the feature combination according to the feature combination difference and the optical intensity attribute value, and generating a global task execution sequence according to the numerical order of the position score.

[0026] Optionally, based on the global task execution sequence, a multi-world resource occupation operation is performed and a resource occupation state is recorded, and a multi-world resource occupation relationship map is generated according to the resource occupation state, comprising:

[0027] According to the order of the global task execution sequence, each task is driven to perform an occupation operation in the multi-world resource;

[0028] Recording the hardware resource identifier and the occupation time window in the execution process of each task to form a resource occupation state data unit containing the hardware resource identifier and the occupation time window;

[0029] Integrating the resource occupation state data unit of each world to form an occupation state set, and detecting the overlap of the occupation time window of the same hardware resource identifier in the occupation state sets of different worlds, and marking as a resource occupation conflict event when there is an overlap of the occupation time window of the same resource in different worlds;

[0030] Taking the world as a node and the resource occupation conflict event as a connecting edge, integrating all nodes and connecting edges to construct a multi-world resource occupation relationship map.

[0031] Optionally, overlap of occupation time windows of the same hardware resource identifier in the set of world occupation states is detected, and when there is overlap of occupation time windows of the same resource in different worlds, a resource occupation conflict event is marked, including:

[0032] Occupation records of the same hardware resource identifier in different world occupation states are scanned, and start and end time stamps of occupation time windows of the resource in different worlds are extracted;

[0033] The difference between the start time stamps of the same hardware resource identifier in different worlds is calculated, and the difference between the end time stamps is calculated, and the difference between the start time stamps is superimposed on the difference between the end time stamps as a time window overlap determination value;

[0034] When the time window overlap determination value is less than a preset threshold, it is determined that there is overlap of occupation time windows of the resource in different worlds;

[0035] For the occupation record of the resource with overlapping occupation time windows, the occupation record of the resource is taken as a resource occupation conflict event.

[0036] Optionally, the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship graph is calculated, and the occurrence probability of the cross-dimension resource conflicts is used to dynamically adjust the parallel world simulation resource task execution order to avoid parallel world simulation deadlock risk, including:

[0037] The connection edges in the multi-world resource occupation relationship graph are scanned, and the number of occurrence times of resource occupation conflict events represented by the connection edges is counted as a conflict event count;

[0038] The ratio of the conflict event count to the total number of connection edges in the graph is calculated, and the ratio is taken as the occurrence probability of cross-dimension resource conflicts;

[0039] When the occurrence probability of the resource exceeds a preset warning threshold, the position of the task node corresponding to the resource in the global task execution sequence is located;

[0040] The task order of the global task execution sequence is adjusted according to the position, so that the occurrence probability is lower than a risk tolerance threshold, to avoid parallel world simulation deadlock risk.

[0041] In a second aspect, the application provides a parallel world simulation resource dynamic scheduling optimization system, including:

[0042] A quantization module is configured to detect task interdependence in parallel world simulation resources to obtain task dependence topology data, and to quantize an associated feature matrix formed by resource competition intensity and order constraints between tasks based on the task dependence topology data;

[0043] A construction module is configured to analyze dynamic changes of the correlation feature matrix by a programmable photon array to construct a transmission path of a cross-dimension optical signal, and form a low-collision resource transmission channel atlas matching the resource competition intensity according to the transmission path;

[0044] A processing module is configured to synchronize processing of the correlation feature matrix and the low-collision resource transmission channel atlas to generate a global task execution sequence;

[0045] A generation module is configured to perform a multi-world resource occupation operation and record a resource occupation state based on the global task execution sequence, and generate a multi-world resource occupation relationship atlas according to the resource occupation state;

[0046] An adjustment module is configured to calculate an occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship atlas, and dynamically adjust a parallel world simulation resource task execution order according to the occurrence probability of the cross-dimension resource conflicts to avoid a parallel world simulation deadlock risk.

[0047] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the parallel world simulation resource dynamic scheduling optimization method in the first aspect.

[0048] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program, and the computer program is executed by a computer to implement the parallel world simulation resource dynamic scheduling optimization method in the first aspect.

[0049] In the embodiment of the present application, intelligent scheduling of parallel world simulation resources is realized through cooperative analysis of task dependency topology and photon transmission path. In the embodiment, dynamic analysis of the correlation feature matrix accurately quantifies resource competition relationships among tasks; the low-collision transmission channel constructed by the programmable photon array significantly improves resource allocation efficiency; and the conflict probability calculation of the multi-world resource occupation atlas effectively prevents the simulation deadlock risk. The method breaks through the limitations of traditional static scheduling, realizes real-time prediction and dynamic avoidance of cross-dimension resource conflicts, ensures efficient and stable operation of the parallel world simulation system, and provides an intelligent solution for resource optimization of complex computing tasks.

[0050] Further, the parallel world simulation resource is accurately modeled by task node topology analysis and resource competition quantification. The high-fidelity task dependency relationship network is constructed by physical link scanning based on port connection state; the two-dimensional vectorization of resource competition intensity and sequential constraint duration significantly improves the expression ability of task association features; the matrix construction method of node sequential arrangement provides a structured input for subsequent resource scheduling. This method breaks through the limitations of traditional dependency analysis, realizes the digital representation of complex relationships between simulation tasks, lays a data foundation for parallel world resource optimization allocation, and effectively improves the running efficiency and stability of large-scale simulation systems.

[0051] Further, dynamic optical signal mapping of simulation resource scheduling is realized by programmable photon array. The optical representation of the association feature matrix is accurately constructed based on the phase shift control of the spatial light modulation unit; intelligent identification of the phase continuous region significantly improves the generation efficiency of cross-dimension transmission paths; the dynamic matching mechanism of optical signal intensity and resource competition intensity effectively reduces the channel conflict probability. This method breaks through the bandwidth limitation of traditional electronic scheduling, realizes the optical processing of high-concurrency resource requests, provides a low-latency and high-throughput resource transmission solution for parallel world simulation, and significantly improves the execution efficiency of complex computing tasks.

[0052] These aspects or other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the 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 some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0054] Figure 1 A flowchart of a parallel world simulation resource dynamic scheduling optimization method provided by the present application is shown;

[0055] Figure 2 A structural schematic diagram of a parallel world simulation resource dynamic scheduling optimization system provided by the present application is shown;

[0056] Figure 3 A structural schematic diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0057] In order to make those skilled in the art better understand the technical solutions of the present application, the technical solutions in the present application will be clearly and completely described below in combination with the drawings in the present application.

[0058] In some of the flowcharts described in the specification and claims of the present application and in the above description of the drawings, a plurality of operations are included in the order in which they occur, but it should be clearly understood that these operations can be executed or performed in parallel or in the order in which they occur in the text, and the serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these flowcharts can include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this text are used to distinguish different messages, devices, modules, etc., and do not represent the order of precedence, nor do "first" and "second" represent different types.

[0059] The technical solutions in the present application will be described clearly and completely in the present application in combination with the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0060] Figure 1 A flowchart of a parallel world simulation resource dynamic scheduling optimization method is provided for the embodiments of the present application, as shown in Figure 1 The method comprises:

[0061] Step 101, detecting the inter-task dependency relationship in the parallel world simulation resource to obtain task dependency topology data, and quantifying the correlation feature matrix formed by the resource competition intensity and the sequence constraint between tasks based on the task dependency topology data.

[0062] In this step, the inter-task dependency relationship refers to the mutual restraint relationship caused by the sharing of resources or the execution sequence limitation when the task is executed. The task dependency topology data is a data structure formed by analyzing the inter-task dependency relationship to describe the connection mode between tasks.

[0063] In the present embodiment, first, the directed graph traversal algorithm is used to detect the inter-task dependency relationship in the parallel world simulation resource, the system scans the input / output resource requirements of each task node, and identifies the execution constraints between tasks. Second, the detection results are converted into task dependency topology data, the adjacency list is used to store the dependency links between task nodes, and the predecessor and successor relationship is clarified. Finally, the correlation feature matrix formed by the resource competition intensity and the sequence constraint between tasks is quantified, and the weighted calculation is used to realize: for each task pair in the dependency topology, the competition intensity of the shared resource type is analyzed, and the rigidity of the sequence constraint is superimposed, to generate an N×N numerical matrix reflecting the comprehensive correlation degree between tasks.

[0064] For example, in a digital twin assembly workshop of a certain automobile manufacturer, the system collects simulation task data of four parallel virtual production lines in real time. Through embedded probe detection, it is found that the body welding task depends on the supply state of the stamping parts, and the painting task needs to wait for the welding quality detection results, thereby constructing the dependency topology data. Based on the topology, the competition intensity of the painting robot cluster for the paint pump resource is quantified, and the sequence constraints of the quality inspection task and the assembly task are identified. These dynamic relationships are encoded as a correlation feature matrix, in which the intersection value of the painting task row and the quality inspection task column is marked as a high-intensity sequence constraint, and the paint pump resource competition intensity is visualized in a gradual color scale.

[0065] Step 102, analyze the dynamic changes of the correlation feature matrix through a programmable photon array to construct a transmission path of cross-dimension light signals, and form a low-conflict resource transmission channel atlas matching the resource competition intensity according to the transmission path.

[0066] In this step, the programmable photon array is a programmable hardware device that processes data through optical signals. Cross-dimension light signals refer to optical signals transmitted in different dimensions. The transmission path is the propagation route of the light signal in the programmable photon array.

[0067] In this embodiment, first, the programmable photon array analyzes the dynamic changes of the correlation feature matrix, uses the optical interference characteristics of the photon chip to map the numerical values in the matrix to the phase modulation parameters of the optical waveguide, and converts the matrix updates to the phase difference of the optical signal in real time. Second, construct a transmission path of cross-dimension light signals, start a dynamic optical switch array based on the phase difference distribution, establish multiple wavelength multiplexing optical paths in the photon chip, and each optical path corresponds to the feature correlation intensity of different task groups. Finally, form a low-conflict resource transmission channel atlas matching the resource competition intensity, select the transmission path with the smallest interference through optical signal signal-to-noise ratio analysis, output each channel in the atlas labeled with the task group it serves and its anti-conflict level, and ensure that high-competition-intensity task groups exclusively occupy independent channels.

[0068] Taking the digital twin assembly workshop of the above-mentioned automobile manufacturer as an example, the programmable photon array receives real-time update data of the correlation feature matrix. When it detects a sudden increase in the resource competition intensity of the painting task, the photon chip dynamically reconstructs the light signal transmission path: allocates a low-delay straight-through optical path for high-priority quality inspection tasks, and establishes a bypass buffer channel for the painting task group. Through the interferometer array modulation of the light phase, a low-conflict resource transmission channel atlas matching the current competition intensity is generated - the bandwidth of the optical path of the painting task group is compressed to relieve the pressure on the air pump, ensuring that the timing constraints are not violated.

[0069] Step 103, synchronize the correlation feature matrix and the low-conflict resource transmission channel atlas to generate a global task execution sequence.

[0070] In this step, the global task execution sequence is generated by comprehensively analyzing the correlation feature matrix and the low-conflict resource transmission channel atlas to generate the task scheduling order.

[0071] In this embodiment, first, the correlation feature matrix and the low-conflict resource transmission channel atlas are synchronized, and the task correlation degree data in the matrix and the channel allocation scheme in the atlas are input into the dynamic scheduler. Second, the global task execution sequence is generated, and the scheduler performs double-rule sorting: for the task group marked as "high conflict risk" in the channel atlas, the order constraint strength is arranged in descending order according to the correlation feature matrix; for the tasks sharing the same low-conflict channel, the resource competition strength is arranged in ascending order. Finally, the global sequence is generated, so that high dependency and high competition tasks preferentially occupy the channel, and the time position of the conflict source task is dispersed.

[0072] Continuing the case of the digital twin assembly workshop of the automobile OEM, the task scheduling engine synchronously processes the process constraints in the correlation feature matrix and the physical limitations of the low-conflict resource transmission channel atlas. When the photon atlas shows that the paint pump channel is overloaded, the system offsets the start time of the three painting robots and inserts the assembly task as a buffer. In the finally generated global task execution sequence, the high-conflict painting task is dispersed to three non-consecutive time periods, and the quality inspection task is strictly arranged to start after the baking process is completed, realizing the double balance of resource competition and sequence constraints.

[0073] In step 104, based on the global task execution sequence, a multi-world resource occupation operation is performed and a resource occupation state is recorded, and a multi-world resource occupation relationship atlas is generated according to the resource occupation state.

[0074] In this step, the multi-world resource occupation operation refers to the allocation and use operation of resources in multiple parallel world simulation environments. The resource occupation state is used to describe whether the resource is occupied and its occupation duration, etc. The multi-world resource occupation relationship atlas is generated by recording the resource occupation state to describe the resource allocation situation.

[0075] In this embodiment, first, based on the global task execution sequence, a multi-world resource occupation operation is performed, and the simulation is advanced according to the sequence time step. When each task starts, it applies for the required resource type (such as memory block, processor core). Second, the resource occupation state is recorded, and the resource monitor captures the occupation / release time stamp and the task ID of each physical resource in real time. Finally, the multi-world resource occupation relationship atlas is generated, and all resource state records are aggregated. The heat atlas is drawn with time axis as vertical coordinate and resource type as horizontal coordinate. The color block density in the graph reflects the contention frequency of multiple tasks for the same resource at the same space-time point.

[0076] For example, when executing the global sequence, the system performs multi-world resource occupation operations on the parallel virtual production line: during the locking of the air pump resource by the painting robot of production line No. 1, the corresponding equipment of production line No. 2 is marked as an occupied state at the same time. The resource occupation state is displayed in real time on the three-dimensional topology interface - the painting air pump node extends four occupation links of different colors, respectively representing the resource application queues of the four production lines. In the multi-world resource occupation relationship graph generated based on this, the painting task link of production lines No. 1 and No. 3 crosses at the air pump node, indicating a potential conflict risk.

[0077] In step 105, the occurrence probability of cross-dimension resource conflict in the multi-world resource occupation relationship graph is calculated, and the parallel world simulation resource task execution order is dynamically adjusted according to the occurrence probability of the cross-dimension resource conflict to avoid the risk of parallel world simulation deadlock.

[0078] In this step, the cross-dimension resource conflict refers to the conflict caused by tasks competing for resources in different dimensions. The risk of parallel world simulation deadlock refers to the risk that multiple parallel world simulation tasks cannot continue to execute due to improper resource allocation.

[0079] In this embodiment, first, the occurrence probability of cross-dimension resource conflict in the multi-world resource occupation relationship graph is calculated, the space-time coordinates corresponding to the high-density color block area in the heat map are counted, and a probability model is trained in combination with historical deadlock data to predict the resource occupation conflict rate. Second, the parallel world simulation resource task execution order is dynamically adjusted according to the occurrence probability of the cross-dimension resource conflict: when the conflict probability exceeds the threshold value, the scheduler rearranges the global sequence of step 103, inserts the task with high conflict probability into the low duty cycle period or migrates it to an alternative channel. Finally, the risk of parallel world simulation deadlock is avoided through an iterative feedback loop, ensuring that there is no long-term locking state in the resource occupation relationship graph.

[0080] For example, in the case of the digital twin assembly workshop of the automobile OEM mentioned above, it is found through analysis of the resource occupation graph that the painting task of production line No. 1 holds the air pump waiting for the quality inspection result, while the quality inspection task of production line No. 3 is blocked due to waiting for the air pump to be released, forming a cross-line deadlock loop. The system calculates that the conflict occurrence probability reaches the threshold value, and immediately dynamically adjusts the task order: interrupts the painting task of production line No. 1 and inserts a buffer process, forcing the release of the air pump resource for use by the quality inspection task of production line No. 3. After adjustment, the deadlock probability decreases, and the two production lines resume parallel operation. Continuous monitoring shows that dynamic scheduling improves the overall efficiency of the production line and reduces resource conflict events.

[0081] In order to analyze the dependency relationship between parallel world simulation tasks, construct task dependency topology data and quantify resource competition intensity and sequential constraint characteristics, form an association feature matrix to reveal the interaction rules between tasks, provide data support for subsequent resource scheduling, realize accurate prediction and conflict prediction of task execution order, and thus improve the resource utilization efficiency and task collaboration ability of the simulation system. According to step 101, the dependency relationship between tasks in the parallel world simulation resource is detected to obtain task dependency topology data, and an association feature matrix formed by resource competition intensity and sequential constraint between tasks is quantified based on the task dependency topology data, including:

[0082] Step 201, scan the port connection state of each task node in the parallel world simulation resource, and record the physical link for data transmission between the task nodes to form task dependency topology data describing the dependency relationship between tasks.

[0083] In this step, the port connection state of the task node refers to the interface state of the task node in the simulation environment for communication with other nodes. The physical link is the actual connection path for data transmission between task nodes. The task dependency topology data is a data structure describing the dependency relationship between tasks formed by analyzing the port connection state and the physical link.

[0084] In this embodiment, first, the port connection state of each task node in the parallel world simulation resource is scanned by a distributed probe tool: the system sends a handshake signal to the communication port of each task node, and judges the node activity according to the response state. Secondly, the physical link for data transmission between task nodes is recorded by using network flow mirroring technology: a packet capture agent is deployed at the data exchange node to capture TCP / IP data packets between task nodes, extract the source address, target address and transmission protocol, and generate a physical link connection log. Finally, the task dependency topology data describing the dependency relationship between tasks needs to parse the physical link log: according to the data flow direction, the predecessor and successor relationship of the task node is identified, the directed dependency relationship is converted into a graph structure stored in an adjacency list, and the node represents the task and the edge represents the dependency direction.

[0085] Step 202, according to the task dependency topology data, the number of task nodes sharing the same resource is counted, and the number of task nodes sharing the same resource is taken as the quantitative value of the resource competition intensity.

[0086] In this step, the number of task nodes sharing the same resource refers to the total number of task nodes that need to use the same resource at the same time in the simulation environment. The quantitative value of the resource competition intensity is a competition degree index obtained by counting the number of task nodes sharing the same resource.

[0087] In the embodiment, first, the number of task nodes sharing the same resource is counted according to the task dependency topology data: the adjacency list of the task dependency topology data is traversed to identify all task edges pointing to the same resource node, and the total number of tasks converging to the resource is counted. Second, the number of task nodes sharing the same resource is taken as the quantitative value of the resource contention intensity. The larger the value, the more intense the resource contention.

[0088] In step 203, the predecessor task and the successor task of the task dependency topology data transmission are detected, and the time window length from the completion of the predecessor task to the start of the successor task is measured as the duration value of the sequential constraint.

[0089] In this step, the predecessor task is a task that must be completed before the current task is executed. The successor task is a task that must be started after the current task is completed. The time window length is the time interval from the completion time of the predecessor task to the start time of the successor task. The duration value of the sequential constraint is the interval time between the execution of the predecessor task and the successor task obtained by measuring the time window length.

[0090] In the embodiment, first, the predecessor task and the successor task of the task dependency topology data transmission are detected: the edge set of the task dependency topology data is traversed to identify all task pairs with direct dependency relationship. Second, the time window length from the completion of the predecessor task to the start of the successor task is measured: the time difference between the state of the predecessor task changing to "completed" and the state of the successor task changing to "started" is calculated through the log timestamp of the task scheduler. Finally, the time window length is taken as the duration value of the sequential constraint, reflecting the forced waiting interval of the task chain.

[0091] In step 204, the quantitative value of the resource contention intensity and the duration value of the sequential constraint are combined into a two-dimensional vector, the node arrangement order in the task dependency topology data is obtained, and all two-dimensional vectors are constructed into an associated feature matrix according to the node arrangement order.

[0092] In this step, the node arrangement order is the arrangement order of the task nodes in the task dependency topology data. The associated feature matrix is a task relationship matrix formed by combining all two-dimensional vectors according to the node arrangement order.

[0093] In the embodiment, first, the quantitative value of the resource contention intensity and the duration value of the sequential constraint are combined into a two-dimensional vector: for each task node, the resource contention intensity value and the duration value of the sequential constraint with the direct successor task are extracted to form a vector. Second, the node arrangement order in the task dependency topology data is obtained: the task dependency topology data is processed by the topological sorting algorithm to generate a linear sequence of task nodes without circular dependency. Finally, all two-dimensional vectors are constructed into an associated feature matrix according to the node arrangement order.

[0094] In order to utilize the dynamic modulation characteristics of the programmable photon array, the correlation feature matrix is mapped to the phase distribution of the optical signal, the transmission path of the cross-dimension optical signal is generated, and the low-conflict resource transmission channel atlas is constructed. According to step 102, the dynamic changes of the correlation feature matrix are analyzed by the programmable photon array to construct the transmission path of the cross-dimension optical signal, and the low-conflict resource transmission channel atlas matching the resource competition intensity is formed according to the transmission path, including:

[0095] Step 301, using the programmable photon array to receive the real-time data stream of the correlation feature matrix, and mapping the row and column coordinates of the correlation feature matrix as element values to the spatial light modulation unit of the programmable photon array.

[0096] In this step, the real-time data stream refers to the dynamic input data of the correlation feature matrix in the programmable photon array. The spatial light modulation unit is an optical element in the programmable photon array for controlling the phase and intensity of the optical signal. The element value refers to the numerical value corresponding to each row and column coordinate in the correlation feature matrix.

[0097] In this embodiment, first, the programmable photon array is used to receive the real-time data stream of the correlation feature matrix, and the electronic signal of the matrix is converted to optical signal input through a high-speed photoelectric conversion interface. Second, the row and column coordinates of the correlation feature matrix are mapped as element values to the spatial light modulation unit of the programmable photon array: each row of the matrix corresponds to a group of modulation units in the photon array, and the row and column coordinate values are converted to the physical position coordinates of the modulation units, and the element values at the coordinates are written into the control registers of the modulation units.

[0098] Step 302, the spatial light modulation unit dynamically adjusts the offset of the phase of the optical signal according to the element value, and maintains a positive proportional relationship between the offset and the element value to form a phase distribution map.

[0099] In this step, the offset of the phase of the optical signal refers to the adjustment amount of the phase of the optical signal during propagation. The phase distribution map is a phase change atlas formed after the spatial light modulation unit adjusts the phase of the optical signal.

[0100] In this embodiment, first, the spatial light modulation unit dynamically adjusts the offset of the light signal phase by the element value: each modulation unit generates a corresponding driving voltage according to the element value in the register, which changes the arrangement angle of the liquid crystal molecules, thereby controlling the light signal phase through the unit. Second, keep the offset proportional to the element value: the larger the element value, the higher the driving voltage, the larger the deflection angle of the liquid crystal molecules, resulting in a proportional increase in the light signal phase offset. Finally, form a phase distribution map: after all the modulation units work synchronously, a light field phase distribution map is presented at the output end of the photon array, and the phase offset in different regions of the map intuitively reflects the value distribution of the original matrix.

[0101] Step 303, identify the region where the light signal phase changes continuously in the phase distribution map, and connect the region to generate the transmission path of the cross-dimension light signal.

[0102] In this step, the region where the light signal phase changes continuously refers to the optical region where the phase value in the phase distribution map changes smoothly. The transmission path of the cross-dimension light signal is the light signal propagation route formed by connecting the phase continuously changing region.

[0103] In this embodiment, first, identify the region where the light signal phase changes continuously in the phase distribution map: use the optical gradient detection algorithm to scan the phase distribution map and mark the region where the phase value changes smoothly. Second, connect the region to generate the transmission path of the cross-dimension light signal: physically connect the marked continuous region through a dynamic light switch array, for example, connect the strip region where the phase gradually changes from 0° to 90° as a curved light path, forming a signal transmission channel across different dimensions of the photon chip.

[0104] Step 304, obtain the resource competition intensity value of the transmission path coverage area, and convert the resource competition intensity value into a light signal intensity adjustment coefficient, and load the light signal intensity adjustment coefficient to the transmission path.

[0105] In this step, the resource competition intensity value refers to the competition degree value quantified by the number of task nodes sharing the same resource. The light signal intensity adjustment coefficient is a light signal power adjustment parameter calculated according to the resource competition intensity value.

[0106] In this embodiment, first, obtain the resource competition intensity value of the transmission path coverage area: backtrack the element value mapped to the modulation unit in step 301, and extract the original data of the resource competition intensity corresponding to the modulation unit through which the transmission path passes. Second, convert the resource competition intensity value into a light signal intensity adjustment coefficient: generate a coefficient according to the preset conversion rule, which determines the output power of the laser source. Finally, load the light signal intensity adjustment coefficient to the transmission path: adjust the light signal intensity injected into the path through the laser controller, for example, load a strong light signal to a high competition intensity path and a weak light signal to a low competition path.

[0107] Step 305, combine the transmission path loaded with the light signal intensity adjustment coefficient to construct a low conflict resource transmission channel map.

[0108] In this step, the low conflict resource transmission channel map is an optimized optical signal transmission network formed by integrating the transmission path loaded with the light signal intensity adjustment coefficient.

[0109] In this embodiment, first, combine the transmission path loaded with the light signal intensity adjustment coefficient: integrate all modulated transmission paths, and allocate optical signals of different wavelengths to independent paths through wavelength division multiplexing technology. Second, construct a low conflict resource transmission channel map: generate a map based on the relationship between optical signal intensity and path topology, and label each channel in the map with its service task group, optical signal intensity level, and path physical coordinates, to finally form an optical transmission network blueprint that avoids high-competition task sharing channels.

[0110] To fuse the dual features of the correlation feature matrix and the low conflict resource transmission channel map, a global task execution sequence is generated by calculating the task node feature combination difference and position score, which realizes the collaborative optimization of task priority and resource channel characteristics, ensures that critical tasks obtain the optimal resource path, and improves the efficiency and reliability of multi-task parallel execution. According to step 103, the correlation feature matrix and the low conflict resource transmission channel map are processed synchronously to generate a global task execution sequence, including:

[0111] Step 401, combine the resource competition intensity value in the correlation feature matrix with the sequence constraint duration value into a task node feature vector, and synchronously extract the optical intensity attribute value of the transmission path of the corresponding task node in the low conflict resource transmission channel map as a channel feature vector.

[0112] In this step, the task node feature vector is a multi-dimensional numerical representation composed of resource competition intensity value and sequence constraint duration value. The optical intensity attribute value refers to the optical signal intensity parameter of the transmission path in the low conflict resource transmission channel map. The channel feature vector is a numerical representation of the transmission characteristics of the task node with the optical intensity attribute value as the core.

[0113] In this embodiment, first, the resource competition intensity value in the association feature matrix is combined with the sequence constraint duration value to form a task node feature vector: traverse each row of the association feature matrix, extract the resource competition intensity value and the sequence constraint duration value corresponding to the task node, and directly splice them into a binary vector. Second, the light intensity attribute value of the transmission path corresponding to the task node in the low-conflict resource transmission channel graph is extracted synchronously: according to the task node identifier marked in the graph, the light signal intensity level of the transmission path to which it belongs is located, and the level is mapped to a numerical light intensity attribute value. Finally, the light intensity attribute value is taken as the channel feature vector: the channel feature vector of each task node is a single-dimensional numerical vector, which is stored together with the task node feature vector.

[0114] Step 402, splice the task node feature vector and the channel feature vector into a feature combination of the task node, and calculate the feature combination difference degree of the task node and the adjacent task node.

[0115] In this step, the feature combination is a comprehensive feature representation formed by splicing the task node feature vector and the channel feature vector. The feature combination difference degree is used to quantify the numerical difference degree between adjacent task nodes.

[0116] In this embodiment, first, the task node feature vector and the channel feature vector are spliced into a feature combination of the task node: for each task node, its task node feature vector (binary) and channel feature vector (unary) are connected head-to-tail to form a three-dimensional vector. Second, the feature combination difference degree of the task node and the adjacent task node is calculated: according to the task dependency topology data, the directly adjacent nodes are identified, the three-dimensional feature combination vectors of the two task nodes are compared through the cosine similarity function, and the similarity result is taken to obtain the difference value. This difference degree reflects the difference between resource demand and channel allocation between nodes.

[0117] Step 403, calculate the position score of the task node in the feature combination according to the feature combination difference degree and the light intensity attribute value, and generate a global task execution sequence according to the numerical order of the position score.

[0118] In this step, the position score is a task node priority evaluation value calculated according to the feature combination difference degree and the light intensity attribute value. The global task execution sequence is a task scheduling order list generated by sorting the position score.

[0119] In this embodiment, first, the position score of the task node is calculated according to the feature combination difference degree and the light intensity attribute value: a score function is designed to fuse the two inputs: the feature combination difference degree is converted into a conflict risk weight, and the light intensity attribute value is superimposed to generate a comprehensive score. Second, the global task execution sequence is generated according to the numerical order of the position score: all task nodes are arranged in descending order of position score, and the same score nodes are arranged in ascending order of sequence constraint duration, and finally the global task execution sequence is generated, which ensures that high-score tasks are executed first and low-score tasks are scheduled later.

[0120] In order to record the multi-world resource occupation state based on the global task execution sequence, the resource occupation relationship graph is constructed by detecting the overlap of the cross-world resource occupation time window, and the multi-world resource conflict event is visualized, which provides a structured data basis for resource conflict analysis and deadlock prevention, and ensures the stable operation of the parallel world simulation. According to step 104, based on the global task execution sequence, the multi-world resource occupation operation is performed and the resource occupation state is recorded, and the multi-world resource occupation relationship graph is generated based on the resource occupation state, including:

[0121] Step 501, according to the order of the global task execution sequence, drive each task to execute the occupation operation in the multi-world resource.

[0122] In this step, the multi-world resource refers to the set of computing resources that can be called by tasks in the parallel world simulation environment. The occupation operation refers to the process of applying and using hardware resources by tasks.

[0123] In this embodiment, first, according to the order of the global task execution sequence, drive each task to execute the occupation operation in the multi-world resource: the task scheduler activates the task node in sequence time steps, and sends a request for exclusive use of hardware resources to the multi-world resource pool through the resource application interface. Second, the resource allocation module audits the request based on the preset priority strategy, and sends a locking instruction to the corresponding hardware resource after the audit is passed, so that the task obtains the exclusive use of the resource. Finally, the task execution engine starts the task operation process under the resource locking state, ensuring the exclusive use of the hardware resource by the task during the occupation period.

[0124] Step 502, record the hardware resource identifier and the occupation time window during the execution of each task to form a resource occupation state data unit containing the hardware resource identifier and the occupation time window.

[0125] In this step, the hardware resource identifier is a code that uniquely identifies a specific hardware resource. The occupation time window is the start and end time range of the task occupying the hardware resource. The resource occupation state data unit is a data structure that records the resource occupation of a single task.

[0126] In the embodiment, first, record the hardware resource identifier and the occupied time window in each task execution process: the resource monitoring agent captures the unique identifier of the hardware resource at the moment when the task occupies the resource, and starts a timer to record the starting time stamp of the occupation. Second, capture the ending time stamp again when the task releases the resource, and form a structured record with <hardware resource identifier, starting time, ending time> as the field. Finally, form a resource occupation state data unit containing the hardware resource identifier and the occupation time window: encapsulate each record as an independent data unit, and store it in the time series database according to the task execution order, where the time window is accurate to the millisecond level and strictly bound with the hardware resource identifier.

[0127] Step 503, integrate the resource occupation state data unit of each world to form an occupation state set, and detect the overlap of the occupation time window of the same hardware resource identifier in the occupation state set of different worlds. When there is an overlap in the occupation time window of the same resource in different worlds, mark it as a resource occupation conflict event.

[0128] In this step, the occupation state set is a resource usage data set formed by integrating multiple resource occupation state data units. The resource occupation conflict event refers to the case where tasks in different worlds compete for the same hardware resource at the same time.

[0129] In the embodiment, first, integrate the resource occupation state data unit of each world to form an occupation state set: the cross-world data aggregator classifies and combines all resource occupation state data units according to world number, generating a resource occupation set with world as the dimension. Second, detect the overlap of the occupation time window of the same hardware resource identifier in the occupation state set of different worlds: the conflict detection engine traverses all the sets of different worlds, compares the occupation time window of the same hardware resource identifier in different worlds, and if there is an intersection between any two time windows, a conflict is triggered. Finally, when there is an overlap in the occupation time window of the same resource in different worlds, mark it as a resource occupation conflict event: generate a mark record containing the conflict resource identifier, the conflict world pair, and the overlapping time range for each conflict event.

[0130] Step 504, integrate all nodes and connection edges to construct a multi-world resource occupation relationship graph with world as node and the resource occupation conflict event as connection edge.

[0131] In this step, the multi-world resource occupation relationship graph is a resource competition relationship network graph constructed with world as vertex and resource conflict as edge.

[0132] In this embodiment, first, taking the world as a node and the resource occupation conflict event as a connecting edge: the graph constructor abstracts each world as a graph node and converts the conflict events marked in step 503 into connecting edges. Second, integrating all nodes and connecting edges: traversing all world nodes and conflict event edges, merging multiple conflict events between nodes into a single edge (avoiding duplicate edges), and attaching conflict details to each edge. Finally, constructing a multi-world resource occupation relationship graph: storing node and edge data based on a graph database, generating a visual relationship network, and reflecting conflict frequency through the thickness of edges in the graph and the total number of conflict events in the world through the size of the node.

[0133] In order to accurately identify cross-world resource occupation conflicts, the resource competition intensity is quantified by calculating the time window overlap determination value, a conflict detection mechanism based on timestamp difference is established, and the automatic marking and classification of resource conflict events are realized. According to step 503, the overlap of the occupation time windows of the same hardware resource identifier in the occupation state set of different worlds is detected, and when the occupation time windows of the same resource in different worlds overlap, it is marked as a resource occupation conflict event, including:

[0134] Step 601, scanning the resource occupation records with the same hardware resource identifier in the occupation state set of different worlds, and extracting the start timestamp and end timestamp of the occupation time window of the resource in different worlds.

[0135] In this step, the resource occupation record is a data entry that describes the time range of the task occupying the hardware resource. The start timestamp is the exact time point at which the task starts to occupy the hardware resource. The end timestamp is the exact time point at which the task ends occupying the hardware resource.

[0136] In this embodiment, first, scanning the resource occupation records with the same hardware resource identifier in the occupation state set of different worlds: the conflict detection engine traverses the resource occupation state set of all worlds, quickly locates the records of the same hardware resource identifier in different worlds through hash indexing. Second, extract the start timestamp and end timestamp of the occupation time window of the resource in different worlds: for each matching record, parse its time window field, and store the start timestamp and end timestamp into a temporary buffer queue respectively, forming a time sequence data pool grouped by resources.

[0137] Step 602, calculate the difference value of the start timestamp of the same hardware resource identifier in different worlds, and calculate the difference value of the end timestamp, and superimpose the difference value of the start timestamp and the difference value of the end timestamp as the time window overlap determination value.

[0138] In this step, the time window overlap determination value is a numerical index obtained by superimposing the start time difference and the end time difference of the same resource in different worlds.

[0139] In this embodiment, first, the difference of the start time stamp of the same hardware resource identifier in different worlds is calculated: from the time series data pool generated in step 601, take the start time stamp of any two worlds, calculate the absolute time difference. Second, calculate the difference of the end time stamp: synchronize the absolute time difference calculation of the end time stamp of the same world. Finally, superimpose the difference of the start time stamp and the difference of the end time stamp as the time window overlap judgment value: directly add the two differences to generate a scalar overlap judgment value, the larger the value, the higher the time window misalignment degree.

[0140] Step 603, when the time window overlap judgment value is less than the preset threshold, it is determined that the occupation time window of the resource by different worlds exists overlap.

[0141] In this step, the preset threshold is a critical value for judging whether the time window overlaps or not, which is set in advance.

[0142] In this embodiment, first, the preset threshold is defined by the resource conflict tolerance strategy. Second, when the time window overlap judgment value is less than the preset threshold: compare the judgment value output by step 602 with the threshold in real time, if the judgment value is less than the threshold, trigger the overlap flag. Finally, it is determined that the occupation time window of the resource by different worlds exists overlap: output a Boolean result (True / False) to mark the conflict risk state of the resource in the world pair.

[0143] Step 604, for the resource occupation records with overlapping occupation time windows, the resource occupation records are taken as resource occupation conflict events.

[0144] In this step, the resource occupation conflict event is the record of the case that different world tasks compete for the same hardware resource in the same time period.

[0145] In this embodiment, first, for the resource occupation records with overlapping occupation time windows: select all resource records marked True in step 603. Second, take the resource occupation records as resource occupation conflict events: generate a structured event object for each record, including the conflict resource identifier, the conflict world pair, and the overlapping time range, and finally aggregate them into a resource occupation conflict event set, which is used to construct the conflict edge of the multi-world resource occupation relationship graph.

[0146] To quantify the probability of multi-world resource conflicts, establish a dynamic adjustment mechanism based on risk thresholds, and optimize the global task execution sequence in real time to avoid deadlock risks, thereby achieving adaptive resource scheduling in the parallel world simulation system and ensuring stable system operation and efficient resource utilization under complex task environments, as described in step 105, the probability of cross-dimensional resource conflicts in the multi-world resource occupancy relationship graph is calculated, and the execution order of parallel world simulation resource tasks is dynamically adjusted based on the probability of cross-dimensional resource conflicts to avoid parallel world simulation deadlock risks, including:

[0147] Step 701: Scan the connecting edges in the multi-world resource occupation relationship graph and count the number of resource occupation conflict events represented by the connecting edges as the conflict event count.

[0148] In this step, connecting edges are graphical elements representing resource conflict events in a multi-world resource occupancy graph. The conflict event count is the total number of resource conflict events obtained through statistics.

[0149] In this embodiment, firstly, the connecting edges in the multi-world resource occupation graph are scanned: all edges connecting different world nodes are traversed, with each edge representing a marked resource occupation conflict event. Secondly, the number of resource occupation conflict events represented by the connecting edges is counted: a counter is maintained for each edge, and the count is incremented whenever a resource preemption occurs between the world pairs corresponding to that edge. Finally, the accumulated value is output as the conflict event count, which reflects the total conflict frequency of a specific resource across multiple worlds.

[0150] Step 702: Calculate the ratio of the conflict event count to the total number of connected edges in the graph, and use the ratio as the probability of cross-dimensional resource conflict.

[0151] In this step, the probability of cross-dimensional resource conflicts is calculated as a conflict likelihood index by the ratio of the number of conflict events to the total number of connected edges.

[0152] In this embodiment, firstly, the ratio of the conflict event count to the total number of connected edges in the graph is calculated: sum the conflict event counts of all edges (e.g., the total number of conflicts can be 50), and divide by the total number of edges in the graph (e.g., 100 edges). Secondly, the ratio is used as the probability of cross-dimensional resource conflict (e.g., 50 / 100 = 0.5), and this probability value quantifies the overall risk level of resource contention between multiple worlds.

[0153] Step 703: When a resource whose occurrence probability exceeds a preset warning threshold is detected, locate the position of the task node corresponding to the resource in the global task execution sequence.

[0154] In this step, the preset alert threshold is a preset resource conflict probability alert line. Positioning refers to determining the specific position of the task node corresponding to the conflict resource in the global task execution sequence.

[0155] In this embodiment, first, the preset alert threshold is set. Second, resources with a probability of occurrence exceeding the alert threshold are detected: all resource types are traversed to filter high-risk resources with a probability of occurrence greater than the threshold. Finally, the position of the task node corresponding to the high-risk resource in the global task execution sequence is located: the resource occupancy state data unit is traced back to determine the task node identifier currently occupying the resource, and the position is indexed according to the node order of the global task execution sequence.

[0156] Step 704, adjusting the task order of the global task execution sequence according to the position, so that the probability of occurrence is lower than the risk tolerance threshold, to avoid parallel world simulation deadlock risk.

[0157] In this step, the risk tolerance threshold is the maximum upper limit of the resource conflict probability allowed by the system. Parallel world simulation deadlock risk refers to the possibility that multiple parallel world tasks cannot continue to execute due to resource conflicts.

[0158] In this embodiment, first, the task order of the global task execution sequence is adjusted according to the position: for the task node associated with the high-risk resource, it is moved backward in the sequence through a rescheduling algorithm to make it execute in a low-conflict period. Second, the probability of occurrence is made lower than the risk tolerance threshold: the conflict probability of the adjusted sequence is calculated again by executing steps 701-703 until the probability value is reduced to within the tolerance threshold. Finally, the parallel world simulation deadlock risk is avoided: through task order optimization, concurrent preemption of multiple worlds to the same resource is eliminated, avoiding system-level deadlock caused by circular waiting.

[0159] Figure 2 A structure diagram of a parallel world simulation resource dynamic scheduling optimization system is provided for the present application, as shown in Figure 2 The system comprises:

[0160] The quantization module 21 is configured to detect the dependency relationship between tasks in the parallel world simulation resource to obtain task dependency topology data, and quantize the correlation feature matrix formed by the resource competition intensity and sequence constraints between tasks based on the task dependency topology data.

[0161] The construction module 22 is configured to analyze the dynamic changes of the correlation feature matrix through a programmable photon array to construct a transmission path of cross-dimension optical signals, and form a low-conflict resource transmission channel atlas matched with the resource competition intensity according to the transmission path.

[0162] The processing module 23 is configured to synchronously process the correlation feature matrix and the low-conflict resource transmission channel atlas to generate a global task execution sequence.

[0163] The generating module 24 is configured to perform multi-world resource occupation operations and record resource occupancy states based on the global task execution sequence, and generate a multi-world resource occupation relationship graph according to the resource occupancy states;

[0164] The adjusting module 25 is configured to calculate the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship graph, and dynamically adjust the parallel world simulation resource task execution order according to the occurrence probability of the cross-dimension resource conflicts to avoid the risk of parallel world simulation deadlock.

[0165] Figure 2 The parallel world simulation resource dynamic scheduling optimization system can perform Figure 1 The parallel world simulation resource dynamic scheduling optimization method of the embodiments has been described above. The specific operation of each module and unit of the parallel world simulation resource dynamic scheduling optimization system in the embodiments has been described in detail in the embodiments of the method, and will not be described in detail here.

[0166] In one possible design, Figure 2 The parallel world simulation resource dynamic scheduling optimization system of the embodiments can be implemented as a computing device, such as a server. Figure 3 As shown in the figure, the computing device can include a storage component 31 and a processing component 32.

[0167] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0168] The processing component 32 is configured to perform the above Figure 1 The parallel world simulation resource dynamic scheduling optimization method of the embodiments.

[0169] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A parallel world simulation resource dynamic scheduling optimization method, characterized in that, The method comprises the following steps: detecting the inter-task dependency in the parallel world simulation resource to obtain task dependency topology data, and quantifying the inter-task correlation feature matrix formed by resource competition intensity and sequence constraints based on the task dependency topology data; analyzing the dynamic changes of the correlation feature matrix by a programmable photon array to construct a transmission path of cross-dimension optical signals, and forming a low-conflict resource transmission channel atlas matching the resource competition intensity according to the transmission path; synchronously processing the correlation feature matrix and the low-conflict resource transmission channel atlas to generate a global task execution sequence; based on the global task execution sequence, performing multi-world resource occupation operation and recording resource occupation state, and generating a multi-world resource occupation relationship atlas according to the resource occupation state; calculating the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship atlas, and dynamically adjusting the parallel world simulation resource task execution sequence to avoid parallel world simulation deadlock risk according to the occurrence probability of cross-dimension resource conflicts.

2. The method of claim 1, wherein, The method comprises the following steps: scanning the port connection state of each task node in the parallel world simulation resource, and recording the physical link of data transmission between the task nodes to form task dependency topology data describing the inter-task dependency; based on the task dependency topology data, counting the number of task nodes sharing the same resource, and taking the number of task nodes sharing the same resource as the quantization value of resource competition intensity; detecting the predecessor task and the successor task of the task dependency topology data transmission, and measuring the time window length from the completion of the predecessor task to the start of the successor task as the duration value of sequence constraint; combining the quantization value of resource competition intensity and the duration value of sequence constraint into a two-dimensional vector, and obtaining the node arrangement order in the task dependency topology data, and constructing all two-dimensional vectors into a correlation feature matrix according to the node arrangement order.

3. The method of claim 1, wherein, The method comprises the following steps: receiving the real-time data stream of the correlation feature matrix by a programmable photon array, and mapping the row and column coordinates of the correlation feature matrix to the spatial light modulation unit of the programmable photon array as element values; the spatial light modulation unit dynamically adjusts the offset amount of the optical signal phase by the element value, and maintains a proportional relationship between the offset amount and the element value to form a phase distribution map; identifying the area of continuous change of the optical signal phase in the phase distribution map, and connecting the area to generate a transmission path of cross-dimension optical signals; obtaining the resource competition intensity value of the transmission path coverage area, and converting the resource competition intensity value into an optical signal intensity adjustment coefficient, and loading the optical signal intensity adjustment coefficient to the transmission path; Combine the transmission path loaded with the light signal intensity adjustment coefficient to construct a low-conflict resource transmission channel atlas.

4. The method of claim 1, wherein, Synchronously process the correlation feature matrix and the low-conflict resource transmission channel atlas to generate a global task execution sequence, including: Combine the resource competition intensity value in the correlation feature matrix and the sequence-constrained duration value into a task node feature vector, and synchronously extract the light intensity attribute value of the transmission path of the corresponding task node in the low-conflict resource transmission channel atlas as a channel feature vector; Splice the task node feature vector and the channel feature vector into a feature combination of the task node, and calculate the feature combination difference between the task node and adjacent task nodes; According to the feature combination difference and the light intensity attribute value, calculate the position score of the task node in the feature combination, and generate a global task execution sequence according to the numerical value of the position score.

5. The method as claimed in claim 1, wherein, Based on the global task execution sequence, perform multi-world resource occupation operations and record resource occupancy state, and generate a multi-world resource occupation relationship atlas according to the resource occupancy state, including: According to the order of the global task execution sequence, drive each task to perform occupation operations in the multi-world resource; Record the hardware resource identifier and the occupation time window in each task execution process to form a resource occupancy state data unit containing the hardware resource identifier and the occupation time window; Integrate the resource occupancy state data unit of each world to form an occupancy state set, and detect the overlap of the occupation time window of the same hardware resource identifier in the occupancy state set of different worlds. When the occupation time windows of the same resource in different worlds overlap, mark it as a resource occupation conflict event; Integrate all nodes and connection edges to construct a multi-world resource occupation relationship atlas with the world as the node and the resource occupation conflict event as the connection edge.

6. The method of claim 5, wherein, Detect the overlap of the occupation time window of the same hardware resource identifier in the occupancy state set of different worlds. When the occupation time windows of the same resource in different worlds overlap, mark it as a resource occupation conflict event, including: Scan the resource occupation records with the same hardware resource identifier in the occupancy state set of different worlds, and extract the start timestamp and end timestamp of the occupation time window of the resource in different worlds; Calculate the difference value of the start timestamp of the same hardware resource identifier in different worlds, and calculate the difference value of the end timestamp at the same time. Add the difference value of the start timestamp and the difference value of the end timestamp as the time window overlap judgment value; When the time window overlap judgment value is less than a preset threshold, it is determined that the occupation time windows of the same resource in different worlds overlap; For the resource occupation records with overlapping occupation time windows, the resource occupation records are taken as resource occupation conflict events.

7. The method as claimed in claim 1, wherein, Calculate the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship atlas, and dynamically adjust the parallel world simulation resource task execution order according to the occurrence probability of cross-dimension resource conflicts to avoid parallel world simulation deadlock risk, including: Scan the connection edges in the multi-world resource occupation relationship atlas, and count the number of resource occupation conflict events represented by the connection edges as conflict event counts; Calculate the occurrence probability of cross-dimension resource conflicts in the multi-world resource occupation relationship atlas, and dynamically adjust the parallel world simulation resource task execution order according to the occurrence probability of cross-dimension resource conflicts to avoid parallel world simulation deadlock risk, including: Scan the connection edges in the multi-world resource occupation relationship atlas, and count the number of resource occupation conflict events represented by the connection edges as conflict event counts; calculating a ratio of the conflict event count to a total number of connection edges in the graph, and taking the ratio as an occurrence probability of cross-dimension resource conflict; when detecting that the resource whose occurrence probability exceeds a preset alert threshold, locating a position of a task node corresponding to the resource in the global task execution sequence; adjusting a task order of the global task execution sequence according to the position, so that the occurrence probability is lower than a risk tolerance threshold, to avoid a parallel world simulation deadlock risk.

8. A parallel world simulation resource dynamic scheduling optimization system, characterized in that, comprising: a quantification module configured to detect inter-task dependency in parallel world simulation resources to obtain task dependency topology data, and to quantize an associated feature matrix formed by resource competition intensity and sequence constraints between tasks based on the task dependency topology data; a construction module configured to analyze dynamic changes of the associated feature matrix by a programmable photon array, to construct a transmission path of a cross-dimension optical signal, and to form a low-conflict resource transmission channel graph matching the resource competition intensity according to the transmission path; a processing module configured to synchronously process the associated feature matrix and the low-conflict resource transmission channel graph to generate a global task execution sequence; a generation module configured to execute multi-world resource occupation operations and record resource occupation states based on the global task execution sequence, and to generate a multi-world resource occupation relationship graph according to the resource occupation states; an adjustment module configured to calculate an occurrence probability of cross-dimension resource conflict in the multi-world resource occupation relationship graph, and to dynamically adjust a parallel world simulation resource task execution order according to the occurrence probability of the cross-dimension resource conflict to avoid a parallel world simulation deadlock risk.

9. A computing device, comprising: comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a parallel world simulation resource dynamic scheduling optimization method according to any one of claims 1-7.

10. A computer storage medium, characterized in that, a computer program is stored, and when the computer program is executed by a computer, a parallel world simulation resource dynamic scheduling optimization method according to any one of claims 1-7 is implemented.

Images