Remote operation and maintenance system and method of entertainment device based on internet of things and SaaS platform

Abstract

The present application relates to the technical field of remote monitoring, in particular to an entertainment device remote operation and maintenance system and method based on the Internet of Things and a SaaS platform, the system comprising: a terminal access verification module, a running load monitoring module, a cache state checking module, a task distribution mapping module and a cross-domain resource allocation module.In the present application, the device number, port and gateway information are associated, collated and deduplicated in the access stage, the consistency and uniqueness of the access data can be controlled at the source, the CPU, memory and bandwidth indicators are continuously collected during operation and compared with the resource standards item by item, the resource load state can be analyzed and trend judged in real time, the task segment identification and data block version are matched before task execution, the version is updated when differences are found, and after the update is completed, the task distribution between cross-domain nodes is kept balanced by combining the real-time resource occupation, the risk of resource idling and conflict backlog is reduced, and the continuity of collaborative processing in a multi-node environment is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote monitoring, in particular to an entertainment device remote operation and maintenance system and method based on an Internet of Things and a SaaS platform. BACKGROUND

[0002] The technical field of remote monitoring belongs to the application range of automation control and information communication fusion, and its core matters include real-time collection, transmission and control instruction issuing of distributed terminal device operation status, operation process and environment information through a communication network, covering communication hardware, embedded control systems, data acquisition devices, cloud management platforms and other technical elements. The overall technical field involves the use of wired or wireless communication methods to realize remote centralized management of field terminals at the center end, and is widely used in industrial equipment monitoring, public service facility management and home and commercial terminal device management scenarios. Among them, the traditional entertainment device remote operation and maintenance system refers to the monitoring, playing control and fault handling of various audio and video playing devices, on-demand terminals and interactive control devices in the home or commercial entertainment scene, relying on a local area network or a special communication network. The operation and maintenance management is usually completed by integrating an embedded processing chip and a storage unit at the device end, and issuing control instructions and operation configurations to the device through local management software or a local area network control end.

[0003] The prior art lacks comprehensive verification of the number and network parameters in the device access process, cannot effectively eliminate repeated or abnormal access terminals, and relies only on local monitoring to find potential bottlenecks in a timely manner. When the cached versions of task data differ between multiple nodes, there is a lack of unified processing mechanism, which can easily lead to data inconsistency and response delay in the task execution process. Under the condition of cross-regional, task allocation is not combined with the real-time resource state of the node to judge, which causes resource idling and uneven task load distribution. SUMMARY

[0004] In order to solve the technical problems existing in the prior art, the embodiments of the present application provide an entertainment device remote operation and maintenance system and method based on an Internet of Things and a SaaS platform. The technical solution is as follows:

[0005] On the one hand, an entertainment device remote operation and maintenance system based on an Internet of Things and a SaaS platform is provided, which includes:

[0006] The terminal access verification module obtains the terminal device number, access port and gateway MAC address, reads the permission number, compares the numbers and analyzes the port and MAC corresponding conditions, removes the numbers, and obtains the access terminal list;

[0007] The running load monitoring module collects CPU utilization, memory occupation and bandwidth occupation based on the accessed terminal list, compares the collected data with resource configuration standards, analyzes resource occupation and bottleneck, and obtains a terminal running state table;

[0008] The cache state checking module matches task fragment identification and data block version based on the terminal running state table, analyzes consistency, marks cache conflict for inconsistent devices, and obtains a cache conflict list;

[0009] The task distribution mapping module filters conflict entries based on the cache conflict list, compares task fragments and data block versions with current versions, updates inconsistent versions to current versions, allocates task fragments to idle nodes after updating, and obtains a task allocation mapping table;

[0010] The cross-domain resource allocation module obtains cross-domain node CPU utilization, memory occupation and bandwidth occupation based on the task allocation mapping table, analyzes resource availability, associates nodes and task fragment identification, and obtains a cross-domain task landing point table.

[0011] As a further scheme of the application, the accessed terminal list includes device number, access port and Internet of Things gateway MAC address, the terminal running state table includes CPU utilization, memory occupation and network bandwidth occupation, the cache conflict list includes device number, task fragment identification and cache conflict state, the task allocation mapping table includes task fragment identification, data block version and processing thread node, and the cross-domain task landing point table includes cross-domain data center node, task fragment identification and node resource availability.

[0012] As a further scheme of the application, the resource configuration standard refers to analyzing target ranges of CPU utilization, memory occupation and network bandwidth occupation in the running load monitoring process, comparing real-time occupation data of terminals and nodes with the target ranges in sequence, and judging whether resource usage is normal.

[0013] The resource occupation and bottleneck refer to comparing each terminal CPU, memory and bandwidth occupation with target ranges of resource configuration standards in sequence, judging resource usage proportion and change trend, and analyzing high occupation links.

[0014] As a further scheme of the application, the conflict entry refers to comparing task fragment identification and corresponding data block version in cache state checking, extracting corresponding device number and task fragment identification when versions are inconsistent, and putting them into subsequent version updating and task allocation processes.

[0015] The cross-domain node CPU utilization refers to that in the cross-domain resource allocation process, the used core number and the total core number of the target cross-domain data center node are acquired first, and the relationship is compared, whether the computing resource of the node meets the allocation condition is analyzed, and the task is arranged to be issued when the resource state is in the executable range.

[0016] As a further scheme of the present application, the terminal access verification module comprises:

[0017] The number acquisition submodule acquires the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, associates the device number with the access port and MAC address corresponding relationship, and obtains the port and gateway corresponding information.

[0018] The information comparison submodule compares the device number with the permitted device number based on the port and gateway correspondence, screens the matching items, and obtains the matching device information according to the SaaS tenant configuration table reading permitted device number entries.

[0019] The access deduplication submodule deduplicates the repeated numbers based on the matching device information, calls the device number of the matching item, and obtains the accessed terminal list.

[0020] As a further scheme of the present application, the running load monitoring module comprises:

[0021] The data acquisition submodule acquires the CPU utilization, memory occupation rate and network bandwidth occupation rate of each terminal based on the accessed terminal list, arranges the terminal occupation rate in sequence according to the device number and access port order corresponding to the same terminal, and appends the current value in the order of acquisition time to obtain the terminal occupation rate sequence.

[0022] The resource comparison submodule compares the corresponding items in the sequence according to the CPU utilization standard value, memory occupation rate standard value and bandwidth occupation rate standard value in the resource configuration standard, calculates the CPU occupation rate ratio, and appends the comparison result according to the terminal number to obtain the resource comparison table based on the terminal occupation rate sequence.

[0023] The occupation analysis submodule analyzes the CPU, memory and bandwidth occupation of the terminal based on the resource comparison table, arranges the comparison ratio and the device number in the state item, and associates the terminal list in the order of access port to obtain the terminal running state table.

[0024] As a further scheme of the present application, the cache state verification module comprises:

[0025] The task extraction submodule extracts the task fragment identifier and data block version of the device based on the terminal running state table, corresponds the task fragment identifier and the corresponding data block version, and obtains the task version corresponding information.

[0026] The version matching submodule analyzes whether the versions are consistent based on the task version correspondence information, version by version task segment identifier and data block, calculates the distribution density of the devices with inconsistent versions, and obtains the version abnormal distribution state.

[0027] The conflict classification submodule filters the device information with inconsistent versions based on the version abnormal distribution state, associates the device number and the task segment identifier, and obtains the cache conflict list.

[0028] As a further scheme of the application, the task distribution mapping module comprises:

[0029] The conflict screening submodule screens the entries with cache conflicts based on the cache conflict list, extracts the task segment identifier and the data block version, associates the data and the device number of the entries, and obtains the conflict task information.

[0030] The version updating submodule obtains the task segment identifier and the data block version based on the conflict task information, compares them with the current version record item by item, selects the current version data block for replacement when the versions are inconsistent, and obtains the updated task information.

[0031] The task distribution submodule screens the nodes in an idle state based on the updated task information and the processing thread occupancy rate of the nodes, distributes the updated task segment to the nodes, and obtains the task distribution mapping table.

[0032] As a further scheme of the application, the cross-domain resource allocation module comprises:

[0033] The resource collection submodule obtains the CPU utilization rate, the memory occupancy rate and the network bandwidth occupancy rate of the target cross-domain data center node based on the task distribution mapping table, associates each data with the corresponding node identifier, and obtains the node resource information.

[0034] The availability analysis submodule analyzes the CPU, memory and bandwidth occupancy of the nodes based on the node resource information, judges the resource availability state, analyzes the corresponding result and the node identifier, and obtains the node availability state.

[0035] The task landing association submodule associates the available nodes with the corresponding task segment identifier based on the node availability state, and obtains the cross-domain task landing table.

[0036] On the other hand, an entertainment device remote operation and maintenance method based on an Internet of Things and a SaaS platform is executed based on the above-mentioned entertainment device remote operation and maintenance system based on an Internet of Things and a SaaS platform, comprising the following steps:

[0037] S1: Obtain the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, read the permitted device number entry according to the SaaS tenant configuration table, compare the device number with the permitted number, analyze the correspondence of the access port and the gateway MAC address, remove the device number, and obtain the accessed terminal list;

[0038] S2: Based on the accessed terminal list, obtain the CPU utilization, memory occupation rate and network bandwidth occupation rate of each terminal, compare the running data with the resource configuration standard, analyze the occupation of the terminal on CPU, memory and bandwidth, and obtain the terminal running state table;

[0039] S3: Based on the terminal running state table, extract the task segment identifier and data block version for matching, analyze whether they are consistent, classify the inconsistent devices as cache conflict items, and obtain the cache conflict list according to the device number and task segment identifier;

[0040] S4: Based on the cache conflict list, filter the conflict entries, obtain the task segment identifier and data block version, compare with the current version record, update to the current version when the versions are inconsistent, assign the task segment to the node with idle processing thread occupation rate, and obtain the task allocation mapping table;

[0041] S5: Based on the task allocation mapping table, obtain the CPU utilization, memory occupation rate and network bandwidth occupation rate of the target cross-domain data center node, analyze the node resource availability, associate the node with the task segment identifier, and obtain the cross-domain task landing point table.

[0042] Compared with the prior art, the advantages and positive effects of the present application are:

[0043] In the present application, the device number, port and gateway information are associated and collated in the access stage, and the consistency and uniqueness of the access data can be controlled at the source. The CPU, memory and bandwidth indicators are continuously collected during operation and compared with the resource standards item by item, the resource load state can be analyzed and trend judged in real time, the task segment identifier and data block version are matched before task execution, the version is updated when differences are found, and the task allocation is completed after the update is completed in combination with the real-time resource occupation, so that the task distribution between cross-domain nodes is kept balanced, the risk of resource idling and conflict backlog is reduced, and the continuity of collaborative processing in a multi-node environment is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0045] Figure 1 System flowchart of the present application;

[0046] Figure 2 System block diagram of the present application;

[0047] Figure 3 Flowchart of the terminal access verification module in the present application;

[0048] Figure 4 Flowchart of the running load monitoring module in the present application;

[0049] Figure 5 Flowchart of the cache state checking module in the present application;

[0050] Figure 6 Flowchart of the task distribution mapping module in the present application;

[0051] Figure 7 Flowchart of the cross-domain resource allocation module in the present application;

[0052] Figure 8 Flowchart of the method steps of the present application. DETAILED DESCRIPTION

[0053] The technical solutions in the present application will be described below with reference to the drawings.

[0054] In the embodiments of the present application, the words such as "example", "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "example" in the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. In fact, the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present application, the meaning expressed by "and / or" can be both, or can be one of the two.

[0055] In the embodiments of the present application, "image" and "picture" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent. "Of", "corresponding" and "corresponding" can be used interchangeably at times. It should be pointed out that when the distinction is not emphasized, the meanings expressed are consistent.

[0056] In the embodiments of the present application, sometimes the subscript such as W1 can be written in the form of non-subscript such as W1. When the distinction is not emphasized, the meanings expressed are consistent.

[0057] In order to make the technical problems, technical solutions and advantages of the present application clearer, the following will be described in detail with reference to the drawings and specific embodiments.

[0058] The embodiment of the present application provides an entertainment device remote operation and maintenance system based on an Internet of Things and a SaaS platform, which comprises Figures 1-2 The schematic diagram of the entertainment device remote operation and maintenance system based on the Internet of Things and the SaaS platform is shown in the figure, and the system comprises:

[0059] The terminal access verification module obtains the device number, the access port and the MAC address of the Internet of Things gateway of the entertainment terminal, reads the permitted device number entry according to the SaaS tenant configuration table, compares the device number with the tenant permitted device number, analyzes whether the access port and the gateway MAC address correspond, removes the matched device entry according to the device number, and obtains the terminal access list;

[0060] The running load monitoring module obtains the CPU utilization, the memory occupation rate and the network bandwidth occupation rate of each terminal based on the terminal access list, compares the data with the resource configuration standard, analyzes the resource occupation condition of each terminal, and obtains the terminal running state table;

[0061] The cache state checking module extracts the task segment identifier and the data block version of the device based on the terminal running state table and performs matching, analyzes the consistency of the task segment identifier and the data block version, marks the device as a cache conflict when the device does not match, classifies the devices according to the device number and the task segment identifier, and obtains the cache conflict list;

[0062] The task distribution mapping module filters the items with the cache conflict based on the cache conflict list, obtains the task segment identifier and the data block version, compares the current version record, selects the current version data block for updating when the versions are inconsistent, and allocates the updated task segment to the node with the idle state of the processing thread occupation rate, and obtains the task distribution mapping table;

[0063] The cross-domain resource configuration module obtains the CPU utilization, the memory occupation rate and the network bandwidth occupation rate of the target cross-domain data center node based on the task distribution mapping table, analyzes the node resource availability, associates the node with the task segment identifier, and obtains the cross-domain task landing point table.

[0064] The terminal access list comprises the device number, the access port and the MAC address of the Internet of Things gateway, the terminal running state table comprises the CPU utilization, the memory occupation rate and the network bandwidth occupation rate, the cache conflict list comprises the device number, the task segment identifier and the cache conflict state, the task distribution mapping table comprises the task segment identifier, the data block version and the processing thread node, and the cross-domain task landing point table comprises the cross-domain data center node, the task segment identifier and the node resource availability.

[0065] Please refer to Figure 2 and Figure 3 , the terminal access verification module comprises:

[0066] The number acquisition submodule acquires the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, associates the device number with the access port and MAC address correspondence, and obtains the port and gateway correspondence information;

[0067] First, the entertainment terminal is connected to the collection device information one by one. The device number can be directly extracted through the unique code reading interface of the device body. The access port needs to be detected through physical connection to determine the corresponding interface number. For example, the occupation of port 1 to port 16 in a certain cabinet is detected, and the occupied port number is corresponded to the unique number of the device. The MAC address of the Internet of Things gateway is obtained through network connection information. The hardware address string of the gateway is read and formatted. Then, in the execution process, the acquired device number, access port number and MAC address are associated two by two. For example, the terminal with number E1001 is corresponded to port 05 and MAC address A1:B2:C3:D4:E5:F6. This step needs to judge whether the port and the MAC address belong to the same physical connection path. The network topology information table can be called for comparison. If it is found that the number, port and MAC information belong to the same link in the topology table, the combination is confirmed as valid matching. All valid matches are confirmed in a one-to-many relationship according to the device number dimension, that is, only the corresponding unique port and unique MAC address combination of a device number is retained. For example, when E1002 detects signals on two different ports, the gateway MAC address connected by the port needs to be compared, and the data consistent with the authorized gateway MAC is filtered out. Finally, all filtered matching relationships are sorted into a set of device number, corresponding port and corresponding MAC address, and the port and gateway correspondence information is obtained.

[0068] The information comparison submodule compares the device number with the permitted device number based on the port and gateway correspondence information according to the SaaS tenant configuration table to read the permitted device number entries, filters the matching entries, and obtains the matching device information.

[0069] First, the port and gateway corresponding information is called piece by piece, the device number in it is extracted, and is compared one by one with the license number list in the SaaS tenant configuration table. Before performing the comparison, the license number list needs to be uniformly processed, such as removing spaces, converting to uppercase or lowercase form, to ensure that the device number format in the port and gateway corresponding information is consistent. In the comparison process, each device number is compared with the license number for string complete matching, and the entries that match successfully are temporarily stored, such as the numbers E2005, E2006, and E2008 in the port and gateway corresponding information, which are compared with E2006, E2007, and E2008 in the license number list. Finally, E2006 and E2008 match successfully. For each matching number, the corresponding port number and gateway MAC address information also need to be associated to ensure that these information can be directly called in the subsequent steps. At the same time, it needs to be judged whether the same device number appears in multiple ports. If there is a repetition, the information consistent with the gateway MAC is retained, and the other inconsistent data is discarded. All the results after comparison and screening are unified into a directly usable data set, which contains device number, corresponding port number, and corresponding gateway MAC address. The matching device information is obtained.

[0070] The access deduplication submodule calls the device number of the matching entry based on the matching device information, deduplicates the repeated numbers, and obtains the list of terminals that have accessed;

[0071] First, call the device number of each record, read these numbers one by one and form a temporary list, and ensure that each number in the list is bound to its corresponding access port and gateway MAC address information during execution, so that the source entry can be accurately located in the deduplication operation, for example, there are E3001, E3002, E3001, E3003 four numbers in the matching device information, among which E3001 appears twice and the port and MAC are consistent, then it should be judged as a duplicate record, in the process of judging the duplicate, the device number can be matched completely by string comparison, and at the same time, the corresponding port number and MAC address field are checked whether they are completely the same, if the three information is completely consistent, it is determined as a duplicate, and only one record is retained in the final set, for the case of different port or MAC but same device number, further judgment is needed whether the device number is allowed multiple access paths in the business rules, for example, if a device is allowed to access through two ports in the entertainment terminal management, the record will not be deleted, if not allowed, the record with port or MAC not meeting the requirements will be deleted, rearrange all the unique device numbers screened out with the corresponding port and MAC information into a standardized record format, for example, finally retain E3001-port 07-MACA1:B2:C3:D4:E5:F6, E3002-port 08-MACA2:B3:C4:D5:E6:F7, E3003-port 10-MACA3:B4:C5:D6:E7:F8, get the list of terminals that have accessed.

[0072] Please refer to Figure 2 and Figure 4 , the running load monitoring module comprises:

[0073] The data acquisition submodule obtains the CPU utilization, memory occupancy and network bandwidth occupancy of each terminal based on the list of terminals that have accessed, and arranges the occupancy of the same terminal in sequence according to the correspondence between the device number and the access port, and appends the current value in sequence according to the collection time, to obtain the terminal occupancy sequence.

[0074] First, the device number and access port information of each record in the list are called in turn, and the current CPU utilization, memory occupancy and network bandwidth occupancy of the corresponding terminal are collected by running the monitoring instruction. The acquisition of each occupancy needs to be associated with the real-time running state of the terminal. For example, on the terminal numbered T5001, port 12, the CPU utilization collection result is 42%, the memory occupancy is 61%, and the network bandwidth occupancy is 38%. These occupancy data need to be synchronized in the same collection process to avoid numerical deviation caused by different time synchronization. Then, for the same terminal, the three occupancy rates collected are one-to-one corresponding according to the device number and access port order, for example, the three occupancy rates of T5001-port 12 are recorded first, and then the three occupancy rates of T5002-port 15 are recorded, forming an ordered data alignment relationship. Then, the latest value is appended to the end of the existing sequence in the order of collection time. If the CPU utilization of T5001 in the last sequence is 40%, the memory occupancy is 60%, and the network bandwidth occupancy is 35%, then 42%, 61% and 38% are directly appended after it. This appending method needs to maintain the consistency of the arrangement of the three occupancy rates in order to facilitate subsequent comparison and analysis. In the execution process, the data length before and after appending also needs to be checked to ensure that the number of occupancy records of each terminal is equal. In this way, in the case of multiple terminals and multiple ports, the sequence can be expanded stably according to the established number and port order, and finally the terminal occupancy sequence is obtained.

[0075] The resource comparison sub-module compares the CPU utilization standard value, the memory occupancy standard value and the bandwidth occupancy standard value in the resource configuration standard with the corresponding items in the sequence based on the terminal occupancy sequence, calculates the CPU occupancy ratio, and appends the comparison result according to the terminal number to obtain the resource comparison table.

[0076] The calculation formula of the CPU occupancy ratio is specifically represented as:

[0077] ;

[0078] Among them, represents the CPU occupancy ratio of the terminal numbered , represents the actual CPU utilization of the terminal , represents the CPU utilization standard value, represents the actual memory occupancy of the terminal , represents the memory occupancy standard value, represents the memory occupancy difference fine tuning item, represents the small constant that avoids the denominator representing zero, represents the actual bandwidth occupancy of the terminal , representative bandwidth occupancy standard value, representative bandwidth ratio adjustment bias, representative logarithmic function base;

[0079] Formula overall operation logic: calculate the absolute difference between the current actual value of CPU and the standard value , and the memory occupancy difference item plus the fine-tuning item multiply, and take the overall root to avoid the explosive interference of excessive resources on the index; then, the total value composed of the CPU standard value and a very small positive number is used as the denominator for normalization processing; multiply the logarithmic function item of bandwidth occupancy rate deviation to map its exponential adjustment ability on the calculation result. The overall multiplication structure maintains three-dimensional coordinated influence, and the absolute value and root number make the abnormal value be softened, and the logarithmic item adjusts the nonlinear expansion trend;

[0080] The following is the monitoring data collection and calculation steps of terminal No. in the sampling period:

[0081] The current resource occupancy value of the terminal is obtained by sampling monitoring reading:

[0082] , which is calculated by reading the CPU user+system value and the ratio of the total number of cores every 10s;

[0083] , which is obtained by the ratio of the used memory value to the total memory value;

[0084] , which is obtained by the ratio of the real-time bandwidth flow to the total limit;

[0085] The reference value of the configuration standard table record is:

[0086] : calculated by the stable period of the previous 30 days;

[0087] : obtained in the same period;

[0088] : average value of data under the normal operation of multiple terminals;

[0089] Coefficient and threshold setting:

[0090] : reference memory peak value fluctuation average in the last 5 minutes;

[0091] : fixed safety constant;

[0092] : bandwidth ratio lower limit protection item;

[0093] : Conventional base 10 is convenient to parse the calculation result range;

[0094] Substitute the parameters in turn to calculate:

[0095] Calculate the difference item:

[0096] ;

[0097] ;

[0098] Product root item:

[0099] ;

[0100] Denominator normalization:

[0101] ;

[0102] Fourth step, logarithmic term:

[0103] ;

[0104] ;

[0105] ;

[0106] Result interpretation and numerical significance: , indicating that the terminal numbered 2 has a very low resource deviation in the current sampling period, and the load deviation is small compared to the configuration standard. The value will be recorded in the resource control table for comparison. The value is in the low segment of the sample threshold 0.00-0.20, indicating that the terminal's current resource usage is reasonable and does not need to be migrated or adjusted. According to the results, the terminal's load balancing action can be skipped.

[0107] Formula innovation point:

[0108] The usefulness of the formula is that by introducing the root processing of the product of CPU and memory difference, and combining the logarithmic function term of bandwidth deviation to participate in the calculation, the coupling and nonlinear adjustment ability between differentiated resource dimensions is achieved. It can identify the offset of multiple resource items and smooth the peak value, thereby enhancing the perception ability of the overall resource scheduling system to the state of heterogeneous resources.

[0109] The occupancy analysis submodule analyzes the CPU, memory and bandwidth occupancy of the terminal based on the resource control table, arranges the comparison ratio and device number as state entries, and associates them to the terminal list in the order of access ports to obtain the terminal running state table.

[0110] Firstly, read the device number and its corresponding CPU, memory and bandwidth occupation ratio in the table one by one, associate the three ratios with the device number, and in the analysis process, the CPU utilization needs to be combined with the pre-set interval standard for judgment, for example, 0% to 40% is judged as low occupation, 41% to 75% is medium occupation, 76% to 100% is high occupation, the memory occupation rate and the bandwidth occupation rate are also classified according to the corresponding interval, for example, the memory occupation rate 0% to 50% is low occupation, 51% to 85% is medium occupation, 86% to 100% is high occupation, the bandwidth occupation rate 0% to 30% is low occupation, 31% to 70% is medium occupation, 71% to 100% is high occupation, these interval values can be obtained by statistics on the resource performance of the entertainment terminal under different loads, for example, the CPU utilization of a device number T8001 is 68%, the memory occupation rate is 77%, and the bandwidth occupation rate is 54%, then the device is judged as medium occupation in CPU item, high occupation in memory item, and medium occupation in bandwidth item, arrange the judgment results of each item according to the device number, form the corresponding state item, and at the same time, the arrangement order of the device number should be consistent with the order of the access port, for example, port 01 corresponds to T8001, port 02 corresponds to T8003, and port 03 corresponds to T8002, then the order of the state item should be T8001, T8003, T8002, ensure that the port order matches the state item order when associated with the accessed terminal list, and finally get the terminal running state table.

[0111] Please refer to Figure 2 and Figure 5 , the cache state checking module comprises:

[0112] The task extraction submodule extracts the task segment identifier and data block version of the device based on the terminal running state table, and obtains the task version corresponding information corresponding to the task segment identifier and the corresponding data block version.

[0113] First, read the unique number of each device in the state table, and retrieve its corresponding task fragment identification field in turn, which is usually a sequence of letters and numbers, used to mark the specific position of the task in the processing flow, and then read the corresponding data block version information of the device, which is a fixed format version number, such as V1.2 or V3.5, used to identify the update batch of the data block, and it is necessary to ensure that the task fragment identification and data block version information come from the same record during extraction to avoid association errors caused by cross-record reading, for example, the device with number D1001, whose task fragment identification is TPX-045 and data block version is V2.1, must be kept in pairs of TPX-045 and V2.1. When performing this process, all device records need to be traversed, and records missing task fragment identification or data block version are skipped or supplemented to ensure data integrity. When there are multiple task fragment identifications for the same device number, the task fragment with the earlier serial number in the access port sequence is prioritized and the other repeated number entries are discarded. This rule can be implemented by maintaining a port priority sequence table, for example, port 03 is prior to port 05, if the same device has records in both ports, the task fragment identification and data block version pair of port 03 is prioritized. After the above matching and screening, each task fragment identification and corresponding data block version are solidified in a one-to-one relationship under the corresponding device number to form an association list indexed by device number, and finally the task version corresponding information is obtained.

[0114] The version matching sub-module analyzes whether the versions are consistent based on the task version corresponding information, version by version task fragment identification and data block, calculates the distribution density of devices with inconsistent versions, and obtains the version abnormal distribution state.

[0115] The calculation formula of the distribution density of devices with inconsistent versions is as follows:

[0116] ;

[0117] Among them, represents the distribution density of devices with inconsistent versions, represents the data block version value corresponding to the i-th task fragment, represents the standard version value in the task version corresponding information, represents the number of devices corresponding to the i-th task fragment, represents the standard deviation of the time series fluctuation of the device resource occupation rate of the i-th task fragment in the version monitoring period, represents the average value of the time series fluctuation standard deviation of all task fragment device resource occupation rates, represents the connectivity value of the device in the i-th task fragment in the network topology, with the dimension of connection number, represents the maximum value of the task fragment device connectivity, with the dimension of connection number, The dimension of the total number of task fragments is the number of fragments, The index of the task fragment;

[0118] Formula operation logic: first, calculate the absolute value of the difference between the actual version value and the standard version value of each task fragment to keep the difference amount non-negative, then multiply the difference value by the number of devices corresponding to the task fragment to reflect the weighted effect of device scale on the difference, then calculate the ratio of the standard deviation of the time series fluctuation of the device resource occupation rate of the task fragment to the average fluctuation standard deviation of all fragments and take the square root as the fluctuation smoothing factor introduced into the aforementioned product, then calculate the ratio of the average device connectivity of the task fragment to the maximum connectivity of all fragments and continue to multiply, so that the network structure characteristics participate in the difference measurement, after the above product calculation, sum the results of all task fragments to get the total global weighted difference, finally, normalize the total amount by dividing it by the total number of task fragments, and thus get the distribution density of version inconsistent devices;

[0119] : The version v1.2.3 corresponds to the value 1.23, and the monitoring obtains the version numbers of the three groups of task fragments as v1.2.3 (1.23), v1.3.0 (1.30), and v1.2.0 (1.20);

[0120] : The current setting is v1.3.0, which corresponds to the value 1.30;

[0121] : The first fragment has 12 devices, the second fragment has 10 devices, and the third fragment has 15 devices;

[0122] : Through 24-hour monitoring, collect CPU, memory, and bandwidth occupation rate every 10 minutes, and calculate the standard deviation;

[0123] The standard deviation of the CPU utilization rate sample sequence obtained by monitoring the first fragment is 4.2%, so ;

[0124] The second fragment is 5.0%, and the third fragment is 3.8%;

[0125] : The average value of the standard deviation of the time series fluctuation of the device resource occupation rate of the task fragment;

[0126] ; : The average connection number of the first fragment is 8, the second fragment is 9, and the third fragment is 7; : The maximum value, i.e. . : The current value is 3.

[0127] Numerical substitution into the formula calculates the absolute version difference of the segment:

[0128] Segment 1: ;

[0129] Segment 2: ;

[0130] Segment 3: ;

[0131] Calculate the square root of the fluctuation ratio of the segment:

[0132] Segment 1: ;

[0133] Segment 2: ;

[0134] Segment 3: ;

[0135] Calculate the connectivity ratio of the segment:

[0136] Segment 1: ;

[0137] Segment 2: ;

[0138] Segment 3: ;

[0139] Calculate the sum part:

[0140] Segment 1 contribution value: ;

[0141] Segment 2 contribution value: ;

[0142] Segment 3 contribution value: ;

[0143] Accumulate: ;

[0144] ;

[0145] Result interpretation and numerical significance: distribution density of version inconsistent devices Units per task segment. If the system benchmark value is 0.50 units per task segment, the result indicates that the distribution density of the current version inconsistent devices is higher than the benchmark, meaning that there is a certain degree of centralized difference distribution in version management, corresponding to the step result "version abnormal distribution state", indicating that part of the task segments need to be adjusted for version consistency;

[0146] Formula innovation point explanation:

[0147] The formula has the advantages that by introducing the square root of the ratio of the standard deviation of the device resource occupancy time series fluctuation and the ratio of the device network topology connectivity, two parameters, the version difference value and the number of devices are jointly involved in the product operation, the quantitative fusion of the device running fluctuation and the network structure influence is referred to in the version difference calculation, so that the influence of the device group running characteristics on the version abnormal set can be reflected in the distribution density evaluation.

[0148] The conflict classification submodule filters the device information with inconsistent versions based on the version abnormal distribution state, obtains the cache conflict list corresponding to the device number and the task segment identifier.

[0149] First, read all the records in the distribution, and retrieve the device numbers marked as version inconsistent in turn, take out these numbers one by one, and cross-query the task segment identifiers corresponding to them in the associated data table, to ensure that the task segment identifier and the device number are the same source record. For example, device number D4502 is marked as version inconsistent in the version abnormal distribution state, it needs to be positioned to its unique task segment identifier TPJ-092 from the task segment record table, and paired with the device number. This step needs to compare whether the combination of device number and task segment identifier exists repeatedly in the execution process. If a repeated combination is found, only the earliest registered record is retained to avoid data redundancy in subsequent processing. When the same device number is associated with multiple task segment identifiers, the priority should be judged according to the device access port order or the storage address order. For example, the task segment of port 02 is prior to the task segment of port 05. The selected combination will directly enter the cache conflict data set. After completing the combination confirmation one by one, all device numbers and task segment identifiers that meet the conditions form a one-to-one correspondence entry, each entry contains device number and task segment identifier two key information, and is arranged in the original input order. Finally, the cache conflict list is obtained.

[0150] Please refer to Figure 2 and Figure 6 The task distribution mapping module includes:

[0151] The conflict screening submodule filters the entries with cache conflicts based on the cache conflict list, extracts the task segment identifier and the data block version, and corresponds the data with the device number of the entry, to obtain the conflict task information.

[0152] Firstly, read each record in turn and judge whether the conflict identification field value is true value, and temporarily store the entries with true conflict identification in the screening list, then extract the task fragment identification and data block version from the screening list one by one, and establish the position correspondence of the two fields with the device number of the original record, and repeat comparison of the task fragment identification during the execution process, if there are multiple same task fragment identifications under the same device number, select the record with the largest data block version difference, and the rest is excluded, to avoid repeated interference, then arrange the device number of the reserved record in sequence, to ensure the stability of the subsequent data correspondence, then merge the task fragment identification and data block version into paired data set in the order of device number, and in the instantiation scenario, reference can be made to the case that device A102, task fragment T56 and data block version V3 have true conflict identification in the conflict list, at this time, they will be bound with device number A102 and stored in the data set, finally complete the processing of all entries meeting the conditions, and get the conflict task information.

[0153] The version updating submodule obtains the task fragment identification and data block version based on the conflict task information, and replaces the current version data block when the versions are inconsistent, to obtain the updated task information;

[0154] Firstly, read the task fragment identification and corresponding data block version of each record in the conflict task data table, use the task fragment identification as the retrieval condition to locate the corresponding current version number in the version record table, call the version field value for accurate matching comparison of characters or values, when the comparison result is the same version, keep the original data block record unchanged, when the comparison result is different versions, perform replacement operation, that is, replace the old version record in the conflict task data table with the version number and corresponding data block content in the current version record table, in this process, it is necessary to ensure that each replacement operation corresponds to a unique task fragment identification and does not affect other non-version related fields under the same identification, for example, when the version of task fragment T015 is recorded as 3.2 in the conflict task information and the current version record table shows the version as 3.5, extract the data block content corresponding to the version 3.5 to directly cover the data block of the original version 3.2, and keep other field information of the task fragment T015 unchanged, if multiple task fragments are involved in one processing, they need to be processed in order according to the task fragment identification number, to ensure that there is no cross reference error in the multiple replacement process, generate an update mark for the replaced record and register the replaced version number in the internal cache, for subsequent checking and preventing repeated replacement, for example, in a batch processing, the versions of task fragments T021, T034 and T045 are replaced from 1.8, 2.0 and 2.4 to 1.9, 2.2 and 2.5 respectively, then the corresponding relationship between the task fragment identification and the new version number is reserved in the update record, after all the processing is completed, the complete record set after replacement is reassembled to obtain the updated task information

[0155] The task distribution sub-module filters the idle nodes based on the updated task information and the processing thread occupancy of the nodes, and distributes the updated task fragments to the nodes to obtain a task distribution mapping table;

[0156] First, the identification, required resource type and resource consumption value of each task fragment in the task information table are read, and these data are compared with the current processing thread occupancy of each node in the node monitoring record. The processing thread occupancy needs to be compared with the set idle judgment reference value in percentage value. The idle judgment reference value can be calculated by referring to 40% of the total number of processing threads, that is, when the current occupancy of a node is lower than 40%, it is marked as idle. For example, node N07 has a current occupancy of 32% and is marked as idle. Then, the nodes are taken out from the idle node list in turn, and the occupancy demand of the task fragments for CPU, memory and bandwidth is matched with the available resources of the nodes one by one. The matching process reserves the binding relationship between the nodes and the task fragments under the condition of meeting all resource requirements. If the node resources are insufficient, the next node is selected for screening. In the matched nodes, the allocation order is arranged according to the priority order of the task fragment identification. The priority can be sorted by task importance level or execution order number. For example, task fragment T305 has a higher priority than T412, and is preferentially allocated to the idle node that meets the condition. Finally, after the matching and order allocation of all task fragments and nodes are completed, the task distribution mapping table is obtained.

[0157] Please refer to Figure 2 and Figure 7 , the cross-domain resource distribution module comprises:

[0158] The resource collection sub-module obtains the CPU utilization, memory occupancy and network bandwidth occupancy of the target cross-domain data center node based on the task distribution mapping table, and obtains the node resource information corresponding to each data and the corresponding node identification.

[0159] First, read the target cross-domain data center node identifier corresponding to each task allocation in the mapping table, take the node identifier as an index one by one, get the CPU utilization, memory occupancy and network bandwidth occupancy of the node from the cross-domain node monitoring record. CPU utilization can be calculated by the number of used cores and the total number of cores, and after the value is obtained, it is expressed in percentage form, for example, node A12 has 3 cores in use among the current 8 cores, then the CPU utilization is 37.5%, the memory occupancy is obtained by comparing and calculating the node real-time memory usage value with the node memory total value, for example, node A12 has a total memory of 64GB, and has used 24GB, then the occupancy is 37.5%, the network bandwidth occupancy needs to call the node real-time bandwidth usage value and the total bandwidth value for comparison and calculation, for example, the total bandwidth of node is 1Gbps, and the current occupancy is 600Mbps, then the bandwidth occupancy is 60%, after obtaining the three data, each data is established one-to-one corresponding relationship with the corresponding node identifier, and is arranged in the unified structure in turn according to the node order in the task allocation mapping table, for the batch data acquisition process of multiple nodes, it is necessary to loop through all the node entries in the task allocation mapping table, to ensure that the CPU utilization, memory occupancy and bandwidth occupancy of all nodes are extracted and associated with their respective node identifiers, for example, the mapping table contains nodes A12, B08 and C15, then the three resource occupancy rates of these three nodes are obtained and associated with the identifiers, and finally the node resource information is obtained.

[0160] The availability analysis submodule analyzes the CPU, memory and bandwidth occupancy of the node based on the node resource information, judges the resource available state, analyzes the corresponding results and node identifiers, and obtains the node available state.

[0161] First, the CPU utilization, memory occupancy and network bandwidth occupancy of each node are read, and the three occupancy data are compared with the preset availability threshold one by one. The availability threshold is set according to the stable operation requirements of the environment where the node is located. For example, the CPU utilization threshold can be set to 80%, the memory occupancy threshold can be set to 75%, and the bandwidth occupancy threshold can be set to 70%. These thresholds are determined by the comprehensive evaluation of the hardware configuration and task load characteristics of the node. The judgment process is to compare whether the CPU utilization is less than 80%, whether the memory occupancy is less than 75%, and whether the bandwidth occupancy is less than 70%. If any of the above exceeds the threshold, it is determined that the resource is unavailable. Otherwise, it is determined that the resource is available. In the case that all resources are available, the node is marked as an available node. If one or more resources are unavailable, the node is marked as an unavailable node. For example, the CPU utilization of node X21 is 72%, the memory occupancy is 68%, and the bandwidth occupancy is 55%. The three values are within the threshold range, so the node is determined to be in an available state. The CPU utilization of node Y09 is 85%, the memory occupancy is 69%, and the bandwidth occupancy is 63%. Since the CPU utilization exceeds the threshold of 80%, the node is determined to be in an unavailable state. After judging the availability of all nodes, the availability judgment result of each node is output corresponding to its node identifier to form a paired data set of node identifier and availability state, and finally the node availability state is obtained.

[0162] The task landing point association submodule associates the available nodes with the corresponding task segment identifiers based on the node availability state to obtain a cross-domain task landing point table.

[0163] First, read the node identification set determined to be in a usable state, and in the data table, call the task fragment identification information corresponding to each available node in turn, locate the task fragment data corresponding to the node with the node identification as the index parameter, if a node has multiple task fragment identifications, arrange them in order according to the numbering of the task fragment identification, and form a one-to-one or one-to-many pairing relationship with the node identification, in this process, cross-check all node task fragment identifications to ensure that there are no duplicate pairing records, for example, node A13 corresponds to task fragments T001, T003, node B07 corresponds to task fragment T002, and node C19 corresponds to task fragments T004, T005, then in the pairing record, node A13 forms an association record with T001, and another association record with T003, and so on, when the number of nodes is large, the nodes can be sorted first according to the alphabetical sequence of the node identification, and then sorted according to the size of the task fragment identification number within each node, so as to maintain the order consistency of the pairing record, after the pairing is completed, all the pairing records are combined into a cross-domain task landing point record set, each record contains the direct correspondence between the node identification and the corresponding task fragment identification, for example, forming A13-T001, A13-T003, B07-T002, C19-T004, C19-T005 such group data, and finally obtaining the cross-domain task landing table.

[0164] Referring to Figure 8 , the entertainment device remote operation and maintenance method based on the Internet of Things and the SaaS platform is executed based on the above-mentioned entertainment device remote operation and maintenance system based on the Internet of Things and the SaaS platform, comprising the following steps:

[0165] S1: Obtain the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, read the permitted device number entries according to the SaaS tenant configuration table, compare the device number with the permitted number, analyze the correspondence between the access port and the gateway MAC address, and remove duplicates according to the device number to obtain a list of terminals that have been accessed;

[0166] S2: Based on the list of terminals that have been accessed, obtain the CPU utilization, memory occupancy and network bandwidth occupancy of each terminal, compare the running data with the resource configuration standard, analyze the occupancy of the terminal in CPU, memory and bandwidth, and obtain a terminal running state table;

[0167] S3: Based on the terminal running state table, extract the task fragment identification and data block version for matching, analyze whether they are consistent, and classify the devices that are not consistent as cache conflict items according to the device number and task fragment identification, and obtain a cache conflict list;

[0168] S4: Based on the cache conflict list, the conflict entries are screened, the task fragment identifier and the data block version are obtained, and the current version record is compared. When the version is inconsistent, the current version is updated, the task fragment is assigned to the node with idle processing thread occupancy, and the task allocation mapping table is obtained;

[0169] S5: Based on the task allocation mapping table, the CPU utilization, memory occupancy and network bandwidth occupancy of the target cross-domain data center node are obtained, the node resource availability is analyzed, the node is associated with the task fragment identifier, and the cross-domain task landing table is obtained.

[0170] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A remote operation and maintenance system for entertainment equipment based on an Internet of Things and a SaaS platform, characterized in that, The system comprises: The terminal access verification module obtains terminal device number, access port and gateway MAC address, reads permission number, compares numbers and analyzes port and MAC corresponding conditions, removes duplicates according to numbers, and obtains a terminal access list; The running load monitoring module collects CPU utilization, memory occupancy and bandwidth occupancy based on the terminal access list, compares the collected data with resource configuration standards, analyzes resource occupancy and bottlenecks, and obtains a terminal running state table; The cache state verification module matches task fragment identifiers and data block versions based on the terminal running state table, analyzes consistency, marks cache conflicts for inconsistent devices, and obtains a cache conflict list; The task distribution mapping module filters conflict entries based on the cache conflict list, compares task fragments and data block versions with the current version, updates inconsistent versions to the current version, and updates the task fragments to idle nodes after the update, thereby obtaining a task distribution mapping table; The cross-domain resource allocation module obtains cross-domain node CPU utilization, memory occupancy and bandwidth occupancy based on the task distribution mapping table, analyzes resource availability, associates nodes and task fragment identifiers, and obtains a cross-domain task landing point table; The cache state verification module comprises: The task extraction submodule extracts device task fragment identifiers and data block versions based on the terminal running state table, corresponding task fragment identifiers and corresponding data block versions, and obtains task version corresponding information; The version matching submodule analyzes whether the versions are consistent based on the task version corresponding information, calculates the distribution density of devices with inconsistent versions, and obtains a version abnormal distribution state; The conflict classification submodule filters device information with inconsistent versions based on the version abnormal distribution state, associates device numbers and task fragment identifiers, and obtains a cache conflict list. 2.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform according to claim 1, characterized in that: The terminal access list includes device number, access port and Internet of Things gateway MAC address, the terminal running state table includes CPU utilization, memory occupancy and network bandwidth occupancy, the cache conflict list includes device number, task fragment identifier and cache conflict state, the task distribution mapping table includes task fragment identifier, data block version and processing thread node, and the cross-domain task landing point table includes cross-domain data center node, task fragment identifier and node resource availability. 3.The IoT and SaaS platform-based remote operation and maintenance system for entertainment equipment according to claim 1, characterized in that, The resource configuration standard refers to analyzing the target range of CPU utilization, memory occupancy and network bandwidth occupancy during the running load monitoring process, comparing the real-time occupancy data of terminals and nodes with the target range in sequence, and determining whether the resource usage is normal; The resource occupancy and bottleneck refer to comparing the CPU, memory and bandwidth occupancy rates of each terminal with the target range of the resource configuration standard, determining the usage proportion and change trend of the resource, and analyzing the high occupancy link. 4.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform of claim 1, wherein, The conflict entry refers to comparing task fragment identifiers and corresponding data block versions in the cache state verification, extracting corresponding device numbers and task fragment identifiers when the versions are inconsistent, and putting them into the subsequent version update and task distribution process. The cross-domain node CPU utilization refers to that in the cross-domain resource allocation process, the used core number and the total core number of the target cross-domain data center node are acquired first, and the relationship is compared, whether the computing resource of the node meets the allocation condition is analyzed, and the task is arranged for dispatch when the resource state is in the executable range. 5.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform according to claim 1, characterized in that, The terminal access checking module comprises: The number acquisition submodule acquires the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, associates the device number with the access port and MAC address corresponding relationship, and obtains the port and gateway corresponding information; The information comparison submodule compares the device number with the permitted device number based on the port and gateway corresponding information, screens the matching items, and obtains the matching device information according to the SaaS tenant configuration table reading permitted device number entries; The access deduplication submodule deduplicates the repeated numbers based on the matching device information, calls the device number of the matching item, and obtains the accessed terminal list. 6.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform of claim 1, wherein, The running load monitoring module comprises: The data acquisition submodule acquires the CPU utilization, memory occupation rate and network bandwidth occupation rate of each terminal based on the accessed terminal list, arranges the terminal occupation rate in sequence according to the device number and access port corresponding order for the same terminal occupation rate, appends the current value in the order of acquisition time, and obtains the terminal occupation rate sequence; The resource comparison submodule compares the corresponding items in the sequence according to the CPU utilization standard value, memory occupation rate standard value and bandwidth occupation rate standard value in the resource configuration standard, calculates the CPU occupation rate ratio, and obtains the resource comparison table by appending the comparison result according to the terminal number; The occupation analysis submodule analyzes the CPU, memory and bandwidth occupation of the terminal based on the resource comparison table, arranges the comparison ratio and device number in corresponding state items, and associates the terminal list in the order of access port to obtain the terminal running state table. 7.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform according to claim 1, characterized in that, The task distribution mapping module comprises: The conflict screening submodule screens the items with cache conflicts based on the cache conflict list, extracts the task fragment identifier and data block version, corresponds the data and the device number of the item, and obtains the conflict task information; The version updating submodule obtains the task fragment identifier and data block version based on the conflict task information, compares the items according to the current version record, selects the current version data block for replacement when the versions are inconsistent, and obtains the updated task information; The task distribution submodule selects the nodes in the idle state based on the updated task information and the processing thread occupation rate of the nodes, distributes the updated task fragments to the nodes, and obtains the task distribution mapping table. 8.The entertainment device remote operation and maintenance system based on the Internet of Things and SaaS platform of claim 1, wherein, The cross-domain resource allocation module comprises: The resource acquisition submodule acquires the CPU utilization, memory occupation rate and network bandwidth occupation rate of the target cross-domain data center node based on the task distribution mapping table, corresponds each data to the corresponding node identifier, and obtains the node resource information; The availability analysis submodule analyzes the CPU, memory and bandwidth occupation of the node based on the node resource information, judges the resource available state, analyzes the corresponding result and node identifier, and obtains the node available state; The task landing point association submodule associates the available nodes with the corresponding task segment identifiers based on the node available state, to obtain a cross-domain task landing point table.

9. A remote operation and maintenance method for entertainment equipment based on an Internet of Things and a SaaS platform, characterized in that, The entertainment device remote operation and maintenance system based on the Internet of Things and the SaaS platform according to any one of claims 1-8, comprising the following steps: S1: Obtain the device number, access port and Internet of Things gateway MAC address of the entertainment terminal, read the permitted device number entries according to the SaaS tenant configuration table, compare the device number with the permitted number, analyze the correspondence between the access port and the gateway MAC address, remove duplicates according to the device number, and obtain a list of terminals that have been accessed; S2: Based on the list of terminals that have been accessed, obtain the CPU utilization, memory occupancy and network bandwidth occupancy of each terminal, compare the running data with the resource configuration standard, analyze the terminal occupancy in CPU, memory and bandwidth, and obtain a terminal running state table; S3: Based on the terminal running state table, extract the task segment identifier and data block version for matching, analyze whether they are consistent, classify the devices that are not consistent as cache conflict items, classify according to the device number and task segment identifier, and obtain a cache conflict list; S4: Based on the cache conflict list, filter the conflict entries, obtain the task segment identifier and data block version, compare with the current version record, update to the current version when the versions are inconsistent, assign the task segment to the node with idle processing thread occupancy, and obtain a task allocation mapping table; S5: Based on the task allocation mapping table, obtain the CPU utilization, memory occupancy and network bandwidth occupancy of the target cross-domain data center node, analyze the node resource availability, associate the node with the task segment identifier, and obtain a cross-domain task landing point table.

Images