Cross-platform office equipment resource dynamic scheduling optimization method and system

Abstract

The invention discloses a cross-platform office equipment resource dynamic scheduling optimization method and system, and relates to the technical field of scheduling optimization. According to the method, a dynamic recoding mechanism based on historical node load information is introduced, so that a more efficient data synchronization effect between cross-platform office equipment is achieved. Through real-time extraction of node over-carrier frequency and dynamic adjustment of a fragmented data path, the method effectively avoids the bottleneck of a hotspot node in a synchronization process, and guarantees the continuity and timeliness of data transmission. Meanwhile, through the comparison feedback of the predicted synchronization delay in advance and the actual synchronization state, the self-adaptive correction of the node performance can be realized, so that the data scheduling strategy between the equipment is gradually optimized, and the overall resource utilization efficiency of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scheduling optimization, and particularly to a dynamic scheduling optimization method and system for cross-platform office equipment resources. Background Art

[0002] With the wide popularization of remote work and multi-device collaboration, enterprises and individual users have put forward higher requirements for the real-time synchronization and dynamic scheduling of office resources in cross-platform and cross-network environments. Traditional office systems often rely on static device configurations and fixed synchronization mechanisms, making it difficult to meet the needs of current complex office scenarios. Especially in the case of frequent interactions between multiple platforms (such as mobile terminals, cloud platforms, intranet servers, etc.) and high-concurrency tasks, there are significant non-equilibrium phenomena in device resource scheduling, resulting in problems such as overload at some nodes and idle resources at some nodes, which seriously restricts the improvement of overall office efficiency. Therefore, how to dynamically optimize the resource scheduling between cross-platform office devices and implement a highly reliable state synchronization mechanism is a technical problem that urgently needs to be solved in the current remote office management system.

[0003] CN119172178B discloses a method for remotely monitoring and managing mobile office devices based on the Internet of Things. By analyzing the operation behavior deviation and identity verification frequency anomaly characteristics of different users, a data prediction model is constructed to optimize the identification and isolation of user behaviors, and improve the security and privacy protection capabilities of the device management system. This solution focuses on the identification and risk control of user behaviors and solves the identity identification problem in a shared device environment. However, its core technical path still remains at the management and differential response on the user side, and does not involve the collaborative scheduling and synchronization mechanism optimization between office equipment resources. In addition, this solution lacks processing strategies for key issues such as uneven resource distribution and synchronization delay control in the multi-node data flow path, and cannot effectively cope with system performance fluctuations caused by resource bottlenecks and synchronization inconsistencies. Therefore, in the face of scenarios of cross-platform high-frequency office state updates and resource synchronization, there are still technical shortcomings such as insufficient processing capacity, inflexible scheduling, and uncontrollable delay.

[0004] CN116668047A discloses a mobile office management method, device, system, electronic device, and storage medium. By connecting the client and the internal server through the cloud server, it realizes the access scheduling and secure communication of office service resources, with the emphasis on ensuring data access security and connection efficiency. However, this method mainly focuses on the access permission management of service resources and the design of connection paths, and does not pay attention to the task status data synchronization and sharding scheduling mechanism issues between office devices. In the actual office scenario, there is a real-time synchronization requirement for multi-source heterogeneous status data between different devices. Without support for data sharding encoding strategies, node overload frequency modeling, and dynamic scheduling mechanisms based on delay prediction, it will be difficult to cope with the load imbalance and synchronization mismatch problems caused by frequent task data interactions. Therefore, although this solution improves the access efficiency, it still does not form the ability of dynamic optimization at the resource scheduling level and cannot achieve real-time intelligent adjustment of status data between nodes. Summary of the Invention

[0005] In view of the problems existing in the above-mentioned background technology, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to fundamentally solve the problems such as low synchronization scheduling efficiency, large state synchronization delay, and unpredictable overloaded nodes between multi-platform office devices.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a dynamic scheduling optimization method for cross-platform office device resources, which includes collecting target status data A to be synchronized between multiple office devices and dividing it into a shard data set F according to a preset sharding rule; extracting the node overload frequency parameter W recorded in the historical status data and dynamically re-encoding the shard data set F to generate a re-encoded shard data set F'; calculating the parameter L of the predicted synchronization delay of the re-encoded shard data set F' on the current data flow path and predicting the synchronization status result B based on the parameter L; synchronizing the re-encoded shard data set F' between cross-platform devices and real-time recording the synchronization status data C; comparing the synchronization status data C with the synchronization status result B, and if the delay deviation exceeds a preset threshold, updating the node overload frequency parameter W and adjusting the subsequent shard re-encoding rule.

[0009] As a preferred solution of the dynamic scheduling optimization method for cross-platform office device resources of the present invention, wherein: the generation of the shard data set F includes: constructing a status record list E = {e1|t1, e2|t2,..., e n |t n}, where, e1~e n are device nodes to be synchronized, and t1~t nis the time stamp of the most recent synchronization record for the corresponding node, where n is the device node; for device nodes e1 to e n In the status buffer, the time stamps from t1 to t n For the newly added target status after that, extract the data set A to be synchronized, and append the original node identifier and acquisition time to each piece of data; perform primary aggregation on the data set A according to the preset content type and node priority dimension, and perform sharding processing according to the source node and data generation time sequence to generate a sharded data set F; append the source node identifier and the time difference Δt from the previous time stamp to each shard in the sharded data set F.

[0010] As a preferred solution of the cross-platform office equipment resource dynamic scheduling optimization method described in the present invention, wherein: the generation of the re-coded sharded data set F' includes: retrieving the data forwarding saturation records of each node recorded in the recent T cycles, extracting the corresponding overload frequency values according to the node numbers to form a node overload frequency parameter set; calculating the forwarding pressure-bearing weight coefficient η of each target node according to the node overload frequency parameter set to form a node weight vector; associating the target synchronization path of each shard in the sharded data set F with the node weight vector, and preferentially allocating high-frequency shards to high-weight nodes to generate the re-coded sharded data set F'.

[0011] As a preferred solution of the cross-platform office equipment resource dynamic scheduling optimization method described in the present invention, wherein: the forwarding pressure-bearing weight coefficient η is inversely proportional to the overload frequency.

[0012] As a preferred solution of the cross-platform office equipment resource dynamic scheduling optimization method described in the present invention, wherein: the association includes: parsing the target synchronization path of each shard in the sharded data set F, and extracting all relay node numbers involved in the path to construct a path node set corresponding to each shard; calculating the path weighted score of each path based on the node weight vector; re-ordering the sharded data set F according to the frequency attribute and path weighted score corresponding to each shard.

[0013] As a preferred solution of the cross-platform office equipment resource dynamic scheduling optimization method of the present invention, wherein: the calculation of the parameter L is as follows: for each shard in the re-encoded shard data set F′, call the relay node set of the synchronization path and the corresponding node pressure-bearing weight vector, and combine the average time-consuming of unit data forwarding of each node in the recent T0 cycle to construct the basic forwarding time-consuming sequence of each path; combine the time difference Δt attached to the shard, and perform weighted accumulation on the basic forwarding time-consuming sequence in the order of nodes to obtain the predicted synchronization delay parameter L of the corresponding shard; the prediction of the synchronization status result B includes: based on the parameter L values corresponding to each shard, combine the current network bandwidth status and queuing buffer parameters of the target device, comprehensively evaluate the predicted duration of shard synchronization completion, and combine the waiting backlog degree characterized by Δt to judge the synchronization status of each shard, and form the synchronization status result B.

[0014] As a preferred solution of the cross-platform office equipment resource dynamic scheduling optimization method of the present invention, wherein: the update of the node overload frequency parameter W and the adjustment of the subsequent shard re-encoding rule include: if the delay deviation exceeds the preset threshold, retrieve the numbers of each relay node in the corresponding path and record them as the set of performance abnormal nodes; count the abnormal occurrence frequencies of each node in the set of performance abnormal nodes, and update the node overload frequency parameter W to complete a historical performance status correction; according to the updated W, recalculate the forwarding pressure-bearing weight coefficient η and correct the node weight vector; based on the corrected node weight vector, adjust the re-encoding rule of the subsequent shard data set to make the high-frequency shards avoid the nodes with unstable performance.

[0015] In a second aspect, the present invention provides a cross-platform office equipment resource dynamic scheduling optimization system, which includes: a status collection module for collecting target status data A to be synchronized among multiple office equipment and dividing it into a shard data set F according to a preset shard rule; a re-encoding module for extracting the node overload frequency parameter W recorded in the historical status data and dynamically re-encoding the shard data set F to generate a re-encoded shard data set F′; a delay prediction module for calculating the parameter L of the predicted synchronization delay of the re-encoded shard data set F′ on the current data flow path and predicting the synchronization status result B based on the parameter L; a synchronization execution module for synchronizing the re-encoded shard data set F′ among cross-platform devices and recording the synchronization status data C in real time; a deviation adjustment module for comparing the synchronization status data C with the synchronization status result B, and if the delay deviation exceeds the preset threshold, updating the node overload frequency parameter W and adjusting the subsequent shard re-encoding rule.

[0016] In a third aspect, the present invention provides a computer device, including a memory and a processor, where: when the computer program instructions are executed by the processor, the steps of the cross-platform office equipment resource dynamic scheduling optimization method as described in the first aspect of the present invention are implemented.

[0017] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program instructions are executed by a processor, the steps of the cross-platform office device resource dynamic scheduling optimization method as described in the first aspect of the present invention are implemented.

[0018] The beneficial effects of the present invention are as follows: by introducing a dynamic re-encoding mechanism based on historical node load information, the present invention achieves a more efficient data synchronization effect among cross-platform office devices. Through the real-time extraction of node overload frequencies and the dynamic adjustment of sharded data paths, the present invention effectively avoids the bottleneck of hot nodes during the synchronization process, ensuring the coherence and timeliness of data transmission. At the same time, through the advance prediction of the expected synchronization delay and the comparison and feedback of the actual synchronization status, the present invention can achieve adaptive correction of node performance, thereby gradually optimizing the data scheduling strategy between devices and improving the overall resource utilization efficiency of the system. In summary, the present invention not only significantly reduces the average delay of data synchronization in cross-platform office scenarios, but also has the capabilities of dynamic load balancing and abnormal node avoidance, ensuring the stability and real-time performance during the multi-device collaboration process, and is applicable to the resource scheduling optimization requirements in complex heterogeneous office network environments. Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a flowchart of the cross-platform office device resource dynamic scheduling optimization method.

[0021] Figure 2 It is a structural diagram of the cross-platform office device resource dynamic scheduling optimization system. Detailed Embodiments

[0022] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0023] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0024] Secondly, the "one embodiment" or "embodiment" mentioned herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that are mutually exclusive of other embodiments.

[0025] As mentioned in the above background art, with the wide promotion of remote work and multi-device collaboration, enterprise and individual users have put forward higher requirements for the real-time synchronization and dynamic scheduling of office resources in cross-platform and cross-network environments. Traditional office systems often rely on static device configurations and fixed synchronization mechanisms, making it difficult to meet the needs of current complex office scenarios. Especially in the case of frequent interactions and high concurrency of tasks among multiple platforms (such as mobile terminals, cloud platforms, intranet servers, etc.), there are significant non-equilibrium phenomena in device resource scheduling, resulting in problems such as overload of some nodes and idle resources of some nodes, which seriously restricts the improvement of overall office efficiency. Therefore, how to dynamically optimize the resource scheduling among cross-platform office devices and implement a highly reliable state synchronization mechanism is a technical problem that urgently needs to be solved in the current remote office management system.

[0026] Figure 1 It is a flowchart of a method for dynamically scheduling and optimizing cross-platform office device resources according to an embodiment of the present invention. As Figure 1 shown, in the method for dynamically scheduling and optimizing cross-platform office device resources, it includes:

[0027] S1: Collect the target state data to be synchronized among multiple office devices and divide it into a shard data set F according to a preset sharding rule.

[0028] Preferably, the generation of the shard data set F includes the following steps:

[0029] (1) Construct a status record list E = {e1|t1, e2|t2,..., e n |t n}, where e1~e n are the device nodes to be synchronized, t1~t n are the time stamps of the most recent synchronization records of the corresponding nodes, and n is the device node.

[0030] It should be noted that this list can be understood as an information table containing the current operating status, unique device numbers, and the time of their last successful synchronization of multiple office devices (such as printers, scanners, shared terminals, etc.). The process of constructing this status record list includes: periodically capturing the operating status information from the status monitoring module of each device (such as system log manager, status probe, etc.); transmitting the captured information to the aggregation server through the intranet control node; performing data fusion at the aggregation server end, establishing an index, and uniformly constructing the list format (such as represented by JSON or a structured database table).

[0031] Meanwhile, for each device node, this step also records the time stamp of its last synchronization record. This time stamp is extracted from the device's historical synchronization log to identify the time point when the device's last data status was successfully synchronized, ensuring that the starting state of the synchronization operation is precisely controllable.

[0032] (2) According to the device nodes e1 to e n The new target status added after the time stamps t1 to t in the status buffer n is used to extract the data set A to be synchronized, and attach the original node identifier and the collection time to each piece of data.

[0033] Exemplarily, based on the last synchronization timestamp of each device recorded in the status list, trace back from the current moment to t n to extract the new target status information added to the status buffer of this device node during this period. The target status information may include new items in the device task queue, configuration file change information, usage log data, and user interaction records, etc. The extraction of this information requires data extraction on each device node, scanning the new data entries in the status buffer according to the time stamp, and each extracted data item needs to be forced to attach two key fields: one is the original node identifier (i.e., the source of this data item), and the other is the current collection time, forming a data entry set with clear traceability and collection timestamp.

[0034] Different from the extensive practices of full synchronization or simple incremental synchronization in traditional methods, based on the time stamp of each node as the boundary, accurately screen the newly generated or changed data within the time period from the node status buffer, so as to achieve fine-grained and time-sequential continuous status collection.

[0035] Specifically, each piece of collected data item is attached with the original node identifier and the collection time, ensuring that the data always retains its source and time series attributes during the subsequent sharding and scheduling processes. Such an operation can effectively improve the data traceability and time sequence, avoid data dislocation or disorder problems during the synchronization process, and significantly optimize the synchronization reliability and accuracy in a multi-device environment.

[0036] (3) Perform primary aggregation on the data set A according to the preset content type and node priority dimension, and perform sharding processing according to the source node and the data generation time sequence to generate a sharded data set F.

[0037] Preferably, in the embodiment of the present invention, primary aggregation refers to performing preliminary merging and organization on the collected target state data set according to the content type (such as printing status, scanning task, fax activity, etc.) and the node priority dimension (such as device model priority, usage frequency priority, management level priority) without changing the original data granularity, so as to facilitate subsequent sharding processing and improvement of synchronization efficiency.

[0038] During the actual operation of primary aggregation, first traverse each record in the data set A and attach a content type label and a node priority score to it. For example, a printing task status data from a certain main printing device, after calculating its node priority score to be 0.85, this record will be marked as printing class - priority 0.85. Subsequently, perform the first round of grouping according to the content type. For example, first group all printing - type data together, and then group scanning - type, exception - type, etc. in turn. This step ensures that the same - type data is aggregated in order, facilitating subsequent sharding to maintain data consistency.

[0039] Next, within each content type group, sort according to the node priority score from high to low. This step is called priority dimension aggregation, and its core lies in improving the timeliness of key device data synchronization.

[0040] Finally, it should be emphasized that the primary aggregation of the present invention also specifically optimizes the time sequence consistency of multi - source device data. During the aggregation process, calculate the acquisition timestamp of each data item and ensure the incrementality in the time sequence within the same - type data group. Even if there is a microsecond - level time difference in the data reported by different devices, it will be aligned through the built - in time correction mechanism to ensure the consistency of the time line in the subsequent synchronization process of the sharded data set F.

[0041] This is different from the problem in the prior art of simply aggregating by device but ignoring the time sequence, significantly improving the logical coherence of multi - device synchronization.

[0042] (4) Attach the source node identifier and the time difference Δt from the previous time mark to each shard in the sharded data set F.

[0043] Among them, the source node identifier is used to quickly locate the original device node of the shard during the synchronization process, facilitating rapid backtracking and accurate error repair in case of an exception.

[0044] S2: Extract the node overload frequency parameter W recorded in the historical state data, and perform dynamic re - encoding on the sharded data set F to generate a re - encoded sharded data set F'.

[0045] S2.1: Retrieve the data forwarding saturation records of each node recorded in the recent T cycles, extract the corresponding overload frequency values according to the node numbers, and form a node overload frequency parameter set.

[0046] Exemplarily, first retrieve the historical status data in the recent T cycles (for example, T can be the most recent 7 days or the most recent 1000 synchronization operations), and filter out the data forwarding saturation records of each node recorded therein. Among them, the data forwarding saturation record refers to the event that a node reaches or exceeds the preset forwarding capacity upper limit within a certain period. For example, if node X triggers the forwarding saturation alarm 5 times within the T cycle, its corresponding overload frequency value is 5. Taking the node number as the index, extract the overload frequency values of all nodes, and finally form a node overload frequency parameter set.

[0047] S2.2: According to the node overload frequency parameter set, calculate the forwarding pressure-bearing weight coefficient η of each target node to form a node weight vector. Among them, the forwarding pressure-bearing weight coefficient η is inversely proportional to the overload frequency and can be expressed as:

[0048]

[0049] where W i is the overload frequency of node i within the T cycle, and λ is the coefficient for adjusting the sensitivity (such as λ

[0050] = 0.1); through this formula, when W i = 0 (that is, the node is not overloaded), η = 1, indicating the full-load pressure-bearing capacity of the node; and when W i increases, η decreases accordingly, dynamically reflecting the decline in the node's bearing capacity. The set of η values of all nodes constitutes the node weight vector in the current cycle.

[0051] S2.3: Associate the target synchronization path of each shard in the shard data set F with the node weight vector, preferentially allocate high-frequency shards to high-weight nodes, and generate a re-encoded shard data set F'.

[0052] Further, the association includes the following steps: parse the target synchronization path of each shard in the shard data set F, extract all relay node numbers involved in the path, and construct a path node set corresponding to each shard; based on the node weight vector, calculate the path weighted score of each path; reorder the shard data set F according to the frequency attribute corresponding to each shard and the path weighted score.

[0053] Exemplarily, first, parse the data structure of each shard in set F, extract its target synchronization path, and identify all relay node numbers involved in the path. For example, a synchronization path of a shard may successively pass through node e, node b, and node c, then the path node set corresponding to this path is {e, b, c}. The system successively performs this parsing operation on all shards in set F to construct a complete shard-path node association table.

[0054] Furthermore, based on the aforementioned node weight vectors, calculate the path weighted score for each synchronization path. Specifically, the path weighted score S k is calculated as follows:

[0055]

[0056] where P k is the path node set of the shard. Considering the dilution effect of the path length (i.e., the node hop count) on the score, in the present invention, based on the preliminary product value, count the node hop count n e in the path, and calculate the dilution correction coefficient according to the node hop count n e to avoid the imbalance phenomenon that the score of a long path is falsely low due to the increase in the hop count and the score of a short path is falsely high due to the small hop count. S k truly reflects the overall carrying capacity of this synchronization path. The higher the value, the smaller the path pressure and the more suitable for allocating frequently synchronized shards; the lower the value, the more vulnerable the path is and the load should be reduced.

[0057] Based on the frequency attribute of each shard (i.e., the expected frequency of synchronization trigger in the service) and the path weighted score S k , perform reordering on the shard data set F to generate a new reordered shard data set F'.

[0058] The specific sorting logic is as follows:

[0059] For high-frequency shards, preferentially allocate them to the path node sets with higher path weighted scores to ensure that high-frequency synchronization tasks obtain sufficient resource support;

[0060] Low-frequency shards are allowed to be allocated to paths with slightly lower weight values, so as to balance the network load.

[0061] According to this rule, perform priority rearrangement on all shards, and finally form the reordered result F'.

[0062] It should be noted that the division of the sharding frequency attribute is completed based on the trigger frequency statistics of the sharding historical synchronization tasks and in combination with a sliding window weighted model over a certain period. Specifically, the actual trigger times of each shard participating in synchronization in the most recent T periods are counted, and the frequency index is determined accordingly. To divide high frequency and low frequency, the present invention adopts the quantile dynamic threshold method to avoid the problem of poor adaptability brought by a fixed threshold. The specific process is as follows: First, the frequency indexes of all shards are sorted in ascending order; the Q-th quantile (such as Q = 0.7, i.e., the 70% quantile, etc., which is not uniquely limited here) is selected as the frequency division threshold; if the frequency index of a certain shard is greater than or equal to the frequency division threshold, it is marked as a high-frequency shard; otherwise, it is marked as a low-frequency shard.

[0063] Through the above re-encoding mechanism, the present invention effectively avoids the unbalanced phenomenon in the prior art where high-load nodes are continuously overloaded and low-load nodes are idle, dynamically improves the resource scheduling efficiency of the synchronization network, and ensures the balanced and stable operation of data synchronization tasks in a large-scale office equipment environment.

[0064] S3: Calculate the parameter L of the estimated synchronization delay of the re-encoded shard data set F' on the current data flow path, and predict the synchronization status result B based on the parameter L.

[0065] S3.1: For each shard in the re-encoded shard data set F', call the relay node set of the synchronization path and the corresponding node bearing weight vector, and combine the average forwarding time per unit data of each node in recent periods to construct the basic forwarding time sequence of each path.

[0066] It should be noted that during the multi-shard data synchronization process, the delay situation directly determines data consistency and the final synchronization timeliness. In the prior art, however, the shard synchronization delay is often only roughly estimated as the sum of the single-node forwarding times, without fully considering the node load differences and the dynamic changes in the shard urgency, resulting in a large error in the prediction result. Therefore, the operation of the present invention first establishes the re-encoded shard data set F' as the basic data object for delay calculation.

[0067] In addition, different from conventional static path time-consuming calculations, such as Hop Count or link bandwidth average method, etc., the present invention adopts the node bearing weight vector as a dynamic adjustment factor. Each node calculates the forwarding bearing capacity index based on the historical overload frequency and the current period load, which can truly reflect the processing capacity fluctuations of the node at different time periods. The weights calculated above are matched one by one with the path node set to establish a path model with dynamic load characteristics for each shard.

[0068] S3.2: Combine the time difference Δt attached to the shard, and perform weighted accumulation on the basic forwarding time sequence in the node order to obtain the estimated synchronization delay parameter L of the corresponding shard.

[0069] In a conventional synchronization system, path delay is mostly measured by link rate or historical average, ignoring the fluctuating time-consuming characteristics of nodes in the recent period. In the present invention, for each relay node, the average forwarding time per unit data packet within the recent T cycles is statistically calculated, and this statistical value is used as the node delay characterization parameter.

[0070] Specifically, for each node in the path, the recent period statistical table is queried to extract the average forwarding time per unit byte or unit shard. Then, the forwarding times of all nodes are concatenated in path order to form a basic forwarding time sequence. This sequence not only reflects the path length and the number of nodes, but also dynamically reflects the current congestion and time-consuming differences of each segment of the path, thus significantly improving the accuracy of delay estimation.

[0071] After the basic time sequence is constructed, an operation of weighted accumulation of the basic forwarding time sequence is further performed in node order by combining the time difference Δt attached to the shards.

[0072] It should be noted that in the data synchronization scenario, some shards may be lagged and backlogged due to network fluctuations, and their synchronization delay tolerance and priority are significantly different from those of real-time shards. Therefore, in the present invention, by normalizing the Δt value and performing weighted adjustment on the time consumption of each node in the basic time sequence, dynamic compensation or improvement of the path delay of the lagged shards is achieved. By summing the weighted sequence in node order, the estimated synchronization delay parameter L of the shard can be obtained, which accurately reflects the estimated complete synchronization time of the shard from the current state to the target device.

[0073] S3.3: The prediction of the synchronization state result B includes the following operation steps:

[0074] (1) Based on the parameter L values corresponding to each shard, combined with the current network bandwidth state and queuing buffer parameters of the target device, comprehensively evaluate the estimated duration for the shard synchronization to be completed.

[0075] It should be noted that since only calculating the delay L cannot directly guide the synchronous scheduling decision, and the synchronous state result B, as the specific judgment result of the synchronous state, can provide a clear classification basis for the system. In the present invention, the prediction operation of the synchronous state result B value is first carried out according to the parameter L values corresponding to each shard, ensuring that the predicted delay quantization data of each shard becomes the decision basis; secondly, it is corrected and evaluated in combination with the current network bandwidth state and queuing buffer parameters of the target device. The final predicted duration is obtained by dividing the parameter L by the average available bandwidth and multiplying by (1 minus the occupancy ratio of the device's current buffer), that is, the parameter L is reduced when the average available bandwidth is large and amplified when the occupancy ratio of the device's current buffer approaches 1, so as to dynamically reflect the bandwidth improvement and congestion risk. A very small positive number (such as 0.01) is added in the calculation to prevent division-by-zero errors.

[0076] (2) Combine the waiting backlog degree characterized by Δt to judge the synchronous state of each shard and form the synchronous state result B.

[0077] Finally, the present invention combines the waiting backlog degree characterized by Δt to judge the synchronous state of each shard and form the synchronous state result B. Perform a comparison operation of the predicted duration and Δt for each shard to judge the success possibility of shard synchronization.

[0078] Specifically, if the predicted duration value is less than the Δt value, that is, the predicted synchronization duration is within the acceptable time limit, it is determined that the shard is in a normal synchronization state, denoted as the normal identifier in the state result B value; if the predicted duration is greater than or close to the Δt value, that is, the predicted synchronization duration has exceeded or approached the synchronization time limit, it is determined that the shard has a risk of synchronization timeout, corresponding to being denoted as the risk or timeout state identifier in the B value.

[0079] After completing the above one-by-one judgment of all shards, summarize the state identification results of each shard to form the overall synchronous state prediction result B.

[0080] S4: Synchronize the re-encoded shard data set F′ among cross-platform devices and record the synchronous state data C in real time.

[0081] First, according to the target synchronization path of each shard, start the cross-platform data transmission task, and schedule each node in the optimized path to forward data step by step in the order of weights. To improve the transmission efficiency, it is preferred to use an intermediate protocol stack based on heterogeneous interface adaptation to be compatible with the data interface limitations of different platform devices.

[0082] During the synchronization process, perform state monitoring on the summary server side, and monitor each synchronization event of the re-encoded shard in real time, recording information including but not limited to: synchronization start time, transmission delay of each relay node in the path, whether a failure retry occurs, actual completion duration, and final confirmation mark, etc.

[0083] Specifically, when recording the synchronization status data C, the present invention particularly incorporates the time difference Δt between shards and the real-time completion duration into the recording system. The purpose of this design is to directly quantify whether a shard has completed synchronization within the time limit during subsequent synchronization status judgment. The above information is uniformly and structurally recorded as the synchronization status data C.

[0084] S5: Compare the synchronization status data C with the synchronization status result B. If the delay deviation exceeds the preset threshold, update the node overload frequency parameter W and adjust the subsequent shard re-encoding rule.

[0085] First of all, the core of the comparison lies in: taking each shard as a unit, comparing the expected synchronization completion duration of each shard with the actual synchronization completion duration, calculating the delay deviation between the two, that is, |actual duration - expected duration|, and comparing it with the preset threshold. If the delay deviation exceeds the preset threshold, it indicates that there is an unexpected synchronization anomaly in this shard, triggering the following adjustment mechanism.

[0086] Preferably, updating the node overload frequency parameter W and adjusting the subsequent shard re-encoding rule includes:

[0087] (1) If the delay deviation exceeds the preset threshold, retrieve the numbers of each relay node in the corresponding path and record them as the set of nodes with performance anomalies; count the occurrence frequencies of anomalies of each node in the set of nodes with performance anomalies, and update the node overload frequency parameter W to complete a correction of the historical performance state.

[0088] (2) According to the updated W, recalculate the forwarding pressure-bearing weight coefficient η and correct the node weight vector.

[0089] (3) Based on the corrected node weight vector, adjust the re-encoding rule of the subsequent shard data set to enable high-frequency shards to avoid nodes with unstable performance.

[0090] It can be seen that the present invention discovers abnormal nodes and performance deviations in network transmission in a timely manner by comparing the predicted synchronization status result B with the actual synchronization status data C, realizes the dynamic correction of the node overload frequency parameter W and the adaptive adjustment of the re-encoding strategy. This mechanism can effectively identify unstable relay nodes, avoid high-frequency data from continuing to pass through performance bottleneck paths, thereby optimizing shard scheduling and network load distribution. By updating the node weight vector and actively avoiding high-risk paths, the stability and success rate of data synchronization are greatly improved, the risk of synchronization failure caused by transmission delay, congestion or node failure is reduced, and the robustness and intelligent scheduling ability of the system are enhanced.

[0091] Furthermore, this embodiment also provides a cross-platform office equipment resource dynamic scheduling optimization system, including,

[0092] A status collection module 100, which is used to collect target status data A that needs to be synchronized among multiple office devices and divide it into a shard data set F according to a preset sharding rule;

[0093] A recoding module 200, which is used to extract a node overload frequency parameter W recorded in historical status data and perform dynamic recoding on the shard data set F to generate a recoded shard data set F';

[0094] A delay prediction module 300, which is used to calculate a parameter L of the predicted synchronization delay of the recoded shard data set F' on the current data flow path and predict a synchronization status result B based on the parameter L;

[0095] A synchronization execution module 400, which is used to synchronize the recoded shard data set F' among cross-platform devices and record synchronization status data C in real time;

[0096] A deviation adjustment module 500, which is used to compare the synchronization status data C with the synchronization status result B. If the delay deviation exceeds a preset threshold, the node overload frequency parameter W is updated and the subsequent shard recoding rule is adjusted.

[0097] This embodiment also provides a computer device, which is applicable to the case of a cross-platform office device resource dynamic scheduling optimization method, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the cross-platform office device resource dynamic scheduling optimization method as proposed in the above embodiment.

[0098] This computer device may be a terminal. This computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device may be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0099] This embodiment also provides a storage medium, on which a computer program is stored, and when this program is executed by a processor, it implements the cross-platform office device resource dynamic scheduling optimization method as proposed in the above embodiment.

[0100] In summary, by introducing a dynamic re-encoding mechanism based on historical node load information, the present invention achieves a more efficient data synchronization effect among cross-platform office devices. Through the real-time extraction of node overload frequencies and the dynamic adjustment of sharded data paths, the present invention effectively avoids the bottleneck of hot nodes during the synchronization process, ensuring the coherence and timeliness of data transmission. At the same time, through the comparison and feedback of the predicted synchronization delay in advance and the actual synchronization status, the present invention can achieve adaptive correction of node performance, thereby gradually optimizing the data scheduling strategy between devices and improving the overall resource utilization efficiency of the system. In summary, the present invention not only significantly reduces the average delay of data synchronization in cross-platform office scenarios, but also has the capabilities of dynamic load balancing and abnormal node avoidance, ensuring the stability and real-time performance during the multi-device collaboration process, and is applicable to the resource scheduling optimization requirements in complex heterogeneous office network environments.

[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. Cross-platform office equipment resource dynamic scheduling optimization method, characterized in that: Including: Collecting target status data to be synchronized among multiple office devices, and dividing it into a shard data set F according to a preset sharding rule; Extracting the node overload frequency parameter W recorded in the historical status data, and dynamically re-encoding the shard data set F to generate a re-encoded shard data set F'; Calculating the parameter L of the predicted synchronization delay of the re-encoded shard data set F' on the current data flow path, and predicting the synchronization status result B based on the parameter L; Synchronizing the re-encoded shard data set F' among cross-platform devices, and recording the synchronization status data C in real time; Comparing the synchronization status data C with the synchronization status result B, if the delay deviation exceeds a preset threshold, updating the node overload frequency parameter W and adjusting the subsequent shard re-encoding rule.

2. The cross-platform office equipment resource dynamic scheduling optimization method according to claim 1, wherein: The generation of the shard data set F includes: Construct a status record list E of all office equipment, E = {e1|t1, e2|t2,..., e n |t n}, where e1 to e n are device nodes to be synchronized, t1 to t n are time stamps of the most recent synchronization records of the corresponding nodes, and n is the device node; According to device nodes e1 to e n In the status buffer, the new target status after time markers t1 to t n After that, extract the data set A that needs to be synchronized, and attach the original node identifier and acquisition time to each item of data; Performing primary aggregation on the data set A according to the preset content type and node priority dimension, and performing sharding processing according to the source node and the data generation time sequence to generate a shard data set F; Attaching the source node identifier and the time difference Δt from the last time stamp to each shard in the shard data set F.

3. The cross-platform office equipment resource dynamic scheduling optimization method according to claim 1, characterized in that: The generation of the re-encoded shard data set F' includes: Retrieving the data forwarding saturation records of each node recorded in the recent T cycles, extracting the corresponding overload frequency values according to the node numbers, and forming a node overload frequency parameter set; Calculating the forwarding pressure-bearing weight coefficient η of each target node according to the node overload frequency parameter set to form a node weight vector; Associating the target synchronization path of each shard in the shard data set F with the node weight vector, preferentially allocating high-frequency shards to high-weight nodes, and generating a re-encoded shard data set F'.

4. The cross-platform office equipment resource dynamic scheduling optimization method according to claim 3, wherein: The forwarding pressure-bearing weight coefficient η is inversely proportional to the overload frequency.

5. The cross-platform office equipment resource dynamic scheduling optimization method according to claim 4, characterized in that: The association includes: Analyzing the target synchronization path of each shard in the shard data set F, and extracting all relay node numbers involved in the path to construct a path node set corresponding to each shard; Calculating the path weighted score of each path based on the node weight vector; Re-ordering the shard data set F according to the frequency attribute and path weighted score corresponding to each shard.

6. The dynamic scheduling optimization method for cross-platform office equipment resources according to claim 1, characterized in that: The calculation of the parameter L is: For each shard in the re-encoded shard data set F', invoking the relay node set of the synchronization path and the corresponding node pressure-bearing weight vector, and combining the average unit data forwarding time of each node in the recent T0 cycles to construct a basic forwarding time sequence for each path; Combining the time difference Δt attached to the shard, and performing weighted accumulation on the basic forwarding time sequence in the node order to obtain the predicted synchronization delay parameter L of the corresponding shard; The prediction of the synchronization status result B includes: Based on the parameter L values corresponding to each shard, combining the current network bandwidth status and queuing buffer parameters of the target device, comprehensively evaluating the predicted duration for the completion of shard synchronization, and combining the waiting backlog degree characterized by Δt, judging the synchronization status of each shard to form the synchronization status result B.

7. The dynamic scheduling optimization method for cross-platform office equipment resources according to claim 1, characterized in that: The updating the node overload frequency parameter W and adjusting the subsequent shard re-encoding rule includes: If the delay deviation exceeds the preset threshold, retrieve the numbers of each relay node in the corresponding path and record them as the set of nodes with abnormal performance; count the abnormal occurrence frequencies of each node in the set of nodes with abnormal performance, and update the node overload frequency parameter W to complete a correction of the historical performance state. According to the updated W, recalculate the forwarding pressure-bearing weight coefficient η and correct the node weight vector. Based on the corrected node weight vector, adjust the recoding rules of the subsequent shard data set to avoid nodes with unstable performance for high-frequency shards.

8. Cross-platform office equipment resource dynamic scheduling optimization system, based on the cross-platform office equipment resource dynamic scheduling optimization method according to any one of claims 1 to 7, characterized in that: It further includes: A status acquisition module, which is used to acquire the target status data A to be synchronized among multiple office devices and divide it into a shard data set F according to the preset shard rules. A recoding module, which is used to extract the node overload frequency parameter W recorded in the historical status data and perform dynamic recoding on the shard data set F to generate a recoded shard data set F'. A delay prediction module, which is used to calculate the parameter L of the predicted synchronization delay of the recoded shard data set F' on the current data flow path and predict the synchronization status result B based on the parameter L. A synchronization execution module, which is used to synchronize the recoded shard data set F' among cross-platform devices and record the synchronization status data C in real time. A deviation adjustment module, which is used to compare the synchronization status data C with the synchronization status result B. If the delay deviation exceeds the preset threshold, update the node overload frequency parameter W and adjust the subsequent shard recoding rules.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the cross-platform office device resource dynamic scheduling optimization method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the cross-platform office device resource dynamic scheduling optimization method according to any one of claims 1 to 7.

Images