Object set sealing convergence risk management method and device and storage medium

Abstract

The application discloses a method and device for controlling sealing convergence risk of an object set and a storage medium, relates to the technical field of data storage, and is applied to a control system. The method for controlling sealing convergence risk of an object set comprises the following steps: determining an unsealed object set according to an object set state of each object set; determining a remaining sealing workload corresponding to the unsealed object set based on the object set state; comparing the remaining sealing workload with a safe completion cache capacity of the control system, and determining a convergence risk of the unsealed object set according to a comparison result, wherein the safe completion cache capacity represents a cache capability of the control system for completing object sealing or completing object state conversion; and performing a risk control action corresponding to the convergence risk. Based on the method, fine-grained perception of sealing progress and pre-judgment of convergence risk are realized, and invalid occupation of cache resources by an unsealed object set for a long time is avoided.

Description

Technical Field

[0001] This application relates to the field of data storage technology, and in particular to methods, devices and storage media for risk management of object collection sealing and convergence. Background Technology

[0002] In object storage operations, business workflows typically continuously write objects to a collection of objects and then enter a state-sealing phase or complete a state transition after certain conditions are met.

[0003] In relevant object collection write methods, alerts and cleanups are typically based on cache water level, storage capacity, or single-point object integrity. However, this approach only manages based on static thresholds from a global resource perspective. It is unaware of the business boundaries of the object collection, the sealing trigger rules, and the expected workload. It cannot quantify the remaining write progress and sealing gap of a single object collection, nor can it distinguish the source of cache usage, resulting in the object collection failing to be sealed on schedule.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device and storage medium for risk management of object collection sealing and convergence, which aims to solve the technical problem that object collections are difficult to seal on schedule.

[0006] To achieve the above objectives, this application proposes a method for managing the convergence risk of object set sealing, the method comprising: Determine the set of unsealed objects based on the state of each object set; The remaining sealing workload corresponding to the unsealed object set is determined based on the state of the object set. The remaining sealing workload is compared with the safe completion cache capacity of the control system. Based on the comparison result, the convergence risk of the unsealed object set is determined. The safe completion cache capacity represents the caching capacity of the control system for completing object sealing or completing object state transition. Perform the risk control actions corresponding to the convergence risk.

[0007] In one embodiment, the step of determining the remaining sealing workload corresponding to the unsealed object set based on the state of the object set includes: The unsealed object set is determined based on the lifecycle field in the object set state; Calculate the expected workload and the workload already completed for the set of unsealed objects; Based on the expected workload and the completed workload, determine the individual remaining workload for each set of unsealed objects, and the total remaining workload for all sets of unsealed objects.

[0008] In one embodiment, the step of calculating the expected amount of work completed and the amount of work already completed for the set of unsealed objects includes: Obtain the list of objects corresponding to the set of unsealed objects; When the object list is not empty, the expected workload is determined based on the expected number of objects, the expected total number of bytes, the verification information, and / or the completion flag in the object list. The completed workload is determined based on the number of uploaded objects, the number of bytes written, the verification results, and the persistence confirmation results. When the object list is empty, the expected workload and the completed workload are determined based on the upstream workflow declaration, configuration file, historical statistics, tenant profile, business type profile or conservative statistical quantile of the unsealed object set.

[0009] In one embodiment, after the step of determining the expected workload based on the expected number of objects, the expected total number of bytes, verification information, and / or completion markers in the object list when the object list is not empty, the method for managing the risk of sealing and converging the object set further includes: Based on the object list of the unsealed object set, determine the missing object set between the unsealed object set and the uploaded object set; The workload of missing objects is determined based on the size, quantity, verification information, object version, and / or fragmentation information of the missing objects. Update the expected workload based on the workload of the missing objects.

[0010] In one embodiment, the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system, and determining the convergence risk of the unsealed object set based on the comparison result, includes: If a single remaining sealing workload exceeds the safe completion cache capacity, it is determined that the target unsealed object set corresponding to that single remaining sealing workload is non-convergent; or When the total amount of remaining sealing work exceeds the safe completion cache capacity, it is determined that the sets of unsealed objects are not convergent.

[0011] In one embodiment, before the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: Obtain the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity of the cache pool of the management and control system; The sum of the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity is determined as the safe completion cache capacity; The safe reclaimable cache capacity includes the cache capacity obtained from the object set that has been persistently confirmed by the backend and meets the release conditions.

[0012] In one embodiment, after the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: Based on the object write rate and cache persistence rate of the unsealed object set, the cache release rate of the management system, and the effective free capacity of the cache, the expected cache exhaustion time is determined; The estimated sealing time is determined based on the remaining sealing workload of the unsealed object set and the upload rate of the object set; If the expected cache exhaustion time is less than the expected sealing time, execute the risk control action corresponding to the convergence risk.

[0013] In one embodiment, after the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: When the convergence risk of the unsealed object is convergent, if an object set creation request or a workflow writing request is received, calculate the expected workload of the newly added object set corresponding to the object set creation request or the workflow writing request. Obtain the remaining sealing workload for the entire set of unsealed objects; If the sum of the expected workload to be completed and the remaining workload to be sealed is less than the product of the safe completion cache capacity and the admission safety coefficient, the admission process of the object set creation request or the workflow write request will be executed. Otherwise, data writing for the object collection creation request or the workflow write request will be prohibited.

[0014] Furthermore, to achieve the above objectives, this application also proposes a device for managing the risk of sealing and converging an object set. The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of the method for managing the risk of sealing and converging an object set as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the object set sealing and convergence risk management method as described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: After obtaining the object set status of all object sets and the cache pool status of the management system, the unsealed object set is determined based on the object set status. Then, the remaining sealing workload corresponding to the unsealed object set is determined based on the object set status. The remaining sealing workload is compared with the safe completion cache capacity of the management system to determine the convergence risk of the unsealed object set. Finally, the corresponding risk control actions for convergence risk are executed. This overcomes the limitations of management based solely on global cache static thresholds. It achieves fine-grained perception of sealing progress and advance prediction of convergence risk from the business dimension of object sets, so as to take targeted management measures to resolve sealing blockage caused by cache resource constraints. This ensures that object sets can complete sealing and state transition on schedule, avoids cache resources being occupied ineffectively for a long time by unsealed object sets, ensures stable convergence of business workflows, and improves the utilization efficiency of cache resources in the object storage system and the operational stability of write business. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart illustrating the first embodiment of the risk management method for sealing and convergence of the object set in this application; Figure 2 This diagram illustrates the state of the object set and the cache pool for the risk management method of sealing and convergence of the object set in this application. Figure 3 A simplified flowchart illustrating the risk management method for sealing and converging object sets provided in the first embodiment of this application; Figure 4 A schematic diagram of gap risk based on list determination for the risk management method of sealing convergence for the object set of this application; Figure 5A flowchart comparing cache exhaustion time and sealing time for the sealing convergence risk control method for the object set in this application; Figure 6 A simplified flowchart illustrating the admission control of the new object set provided in the fifth embodiment of this application; Figure 7 This is a schematic diagram of the hardware operating environment involved in the risk management method for sealing and converging object sets in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] In object storage operations, business workflows typically continuously write objects to a collection of objects and then enter a state-sealing phase or complete a state transition after certain conditions are met.

[0023] In relevant object collection write methods, alerts and cleanups are typically based on cache water level, storage capacity, or single-point object integrity. However, this approach only manages based on static thresholds from a global resource perspective. It is unaware of the business boundaries of the object collection, the sealing trigger rules, and the expected workload. It cannot quantify the remaining write progress and sealing gap of a single object collection, nor can it distinguish the source of cache usage, resulting in the object collection failing to be sealed on schedule.

[0024] Based on this, this application provides a solution that, after obtaining the object set status of all object sets and the cache pool status of the management system, determines the unsealed object set based on the object set status, then determines the remaining sealing workload corresponding to the unsealed object set based on the object set status, and compares the remaining sealing workload with the safe completion cache capacity of the management system to determine the convergence risk of the unsealed object set, and finally executes the corresponding risk control action for the convergence risk. This overcomes the limitations of managing based solely on global cache static thresholds, and achieves fine-grained perception of sealing progress and advance prediction of convergence risk from the business dimension of the object set, so as to take targeted management measures to resolve the sealing blockage problem caused by cache resource constraints.

[0025] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, or a risk management device for the sealing and convergence of object sets, etc. The following description uses a risk management device for the sealing and convergence of object sets as an example to illustrate this embodiment and the following embodiments, wherein a management system is installed within the risk management device for the sealing and convergence of object sets.

[0026] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0027] This application provides a method for managing the risk of object collection sealing and convergence, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the risk management method for sealing and convergence of the object set in this application.

[0028] In this embodiment, the method for managing the risk of sealing and convergence of the object set includes steps S10 to S40: Step S10: Determine the unsealed object set based on the object set status of each object set.

[0029] The object collection status describes a structured metadata collection of attributes throughout the entire lifecycle of a single object collection. It covers dimensions such as collection identifier, tenant, associated workflow, expected total write volume, uploaded progress, details of missing objects, lifecycle status, archive trigger rules, and key timestamps. It is used to characterize the write progress, archive conditions, lifecycle status, and remaining archive requirements of an object collection.

[0030] In this embodiment, after obtaining the object collection status, the unsealed object collection can be determined by the lifecycle field in the object collection status. For example, if the lifecycle field indicates "uploading", then the object collection corresponding to the object collection status is considered to belong to the unsealed object collection.

[0031] Furthermore, before step S10, it is necessary to first obtain the object collection status of each object set and the cache pool status of the management system, so as to determine the safe completion cache capacity of the management system based on the cache pool status, and to perform convergence risk comparison based on the safe completion cache capacity. The cache pool status is a structured data set describing the overall resource status of the object storage system's write cache pool, covering dimensions such as total capacity, used capacity, free capacity, safe reclaimable capacity, emergency reserved capacity, tenant / collection-specific reserved capacity, and safe completion capacity, used to characterize the cache pool's free capacity, safe reclaimable capacity, reserved capacity, and reserved capacity.

[0032] Specifically, the state of the object collection and the state of the cache pool can be maintained through an event source state engine, a persistent state table, a centralized database, a distributed consistent state service, a message event stream, an object gateway write-ahead log, or a combination thereof, and determined by handling duplicate events, late events, and event replays through idempotent keys, event sequence numbers, object version numbers, state version numbers, or write-ahead log records.

[0033] As an optional implementation method for obtaining the status of object collections and cache pools, when performing data archiving and management tasks, the status of each currently processed object collection and the cache pool status of the management system can be obtained in real time. Specifically, according to a preset fixed collection cycle, a status query request is initiated to all storage nodes in the cluster, traversing all active object collection metadata entries on each node. Simultaneously, a capacity status query request is initiated to the cache pool management module to obtain real-time resource usage data of the cache pool, and non-recoverable data is pre-filtered according to a preset conservative standard to initially calculate basic capacity parameters. Then, the collected full object collection status and cache pool status are normalized, converted into a standard structure, and written to the cluster-shared status memory table for direct use in subsequent calculation steps.

[0034] For example, an object storage cluster for log archiving scenarios deploys 16 storage nodes, with a metadata collection cycle of 10 seconds. Every 10 seconds, the metadata management node initiates a full object collection status query to all storage nodes, traversing the active collections on each node that are in the writing, pending archiving, or abnormally interrupted states; simultaneously, it pulls real-time capacity data from the cache pool management module, pre-filtering out the space occupied by unconfirmed persistent data and compliant retained data; finally, all returned data is uniformly converted into a standard structure and stored in the cluster's shared memory state library.

[0035] As an alternative implementation for obtaining the state of object collections and cache pools, data collection can be performed using an event-driven state engine. When the state of an object collection changes (e.g., creation, writing, state transition, deletion), the storage nodes proactively report the change event, and the management node only synchronizes the changed data, without needing to fetch the entire collection. Each storage node registers a state change listener for its local object collection. When operations such as creation, write progress updates, or lifecycle state changes occur, a change event is automatically triggered and reported to the metadata management node. Upon receiving the change events reported by each node, the management node deduplicates and verifies the events before updating the state data of the corresponding object collection in the state memory table, without needing to traverse all collections. This approach only processes incremental data, significantly reducing the computational overhead and network transmission pressure on the management node. Simultaneously, the cache pool management module also adopts an event-driven mechanism. When cache usage or reclaimable capacity changes, it proactively reports a capacity change event and subsequently updates the cache pool state data synchronously.

[0036] For example, a high-concurrency object storage cluster manages over 100,000 active object collections. Using a full fetch method, a single data collection takes over 30 seconds. However, by using an event-driven incremental collection method, each storage node automatically reports a progress update event when a write operation occurs on its object collection. After receiving the event, the node only updates fields such as the number of bytes uploaded and the last update time for the corresponding collection, without needing to traverse the entire collection. This allows the cache pool to proactively report its status every time its capacity changes by 1%, ensuring the real-time nature of capacity data.

[0037] Furthermore, the obtained object collection state and cache pool state can be as follows: Figure 2As shown, the ObjectCollectionState can include multiple fields such as identity identifier, expected total quantity, upload progress, archiving and caching, and time control. For example, the identity identifier field can include: collection_id, a unique ID for the object collection corresponding to a complete task; tenant_id, a tenant ID used in multi-tenant scenarios to identify the tenant to which the task belongs, for resource isolation and billing; and workflow_id, a workflow ID used to identify the processing flow to which the task belongs. The expected total quantity field can include expected_object_count (expected total number of objects) and expected_total_bytes (expected total number of bytes), i.e., the original size of the task. The upload progress includes uploaded_object_count (number of shards successfully written to the cache), uploaded_total_bytes (total size of shards written to the cache in bytes), missing_object_set (missing object set, i.e., a list of shards that have not yet been successfully uploaded, used for resuming interrupted uploads and re-uploading), and manifest_gap_workload (workload corresponding to missing entries in the metadata manifest, representing the part that has not been completed at the metadata level). The sealing and caching fields include `remaining_sealing_workload` (remaining sealing workload, representing the remaining processing time for merging cached fragments, writing to disk, and generating the final object after all fragment uploads are complete), `cached_bytes` (the number of cache bytes currently occupied by the object collection in the cache pool), `seal_condition` (the sealing trigger condition, such as all fragment uploads completed, timeout reached, etc.), and `lifecycle_state` (lifecycle state, such as task initialization, uploading, sealing, completed, and failed, etc.). Time control fields include `first_put_time` (the upload time of the first fragment, used for timeout statistics and determination), `last_progress_time` (the time of the last progress update, used to determine if the task is stalled and trigger timeout recycling), and `deadline` (the task deadline; tasks that fail to complete within the timeout period will have their cached resources reclaimed). Therefore, the object collection state includes at least the object collection identifier, workflow identifier, tenant identifier, expected number of objects, expected total number of bytes, number of objects already written, number of bytes already written, remaining sealing workload, current cache usage, sealing conditions, lifecycle state, first write time, most recent write time, and sealing deadline.

[0038] Correspondingly, the cache pool state (CacheCompletionState) includes four fields: total and usage, reclaimable and emergency, resource reservation, and safe capacity. For example, the total and usage field includes `cache_pool_id`, a unique ID identifying an independent cache resource pool; `total_cache_bytes`, the total cache pool capacity in bytes, i.e., the physical / logical upper limit of the cache space; `used_cache_bytes`, the number of used cache bytes, the total amount of cache currently actually occupied by all tasks; and `free_cache_bytes` (the number of free cache bytes, the cache space that is currently completely unoccupied). The reclaimable and emergency fields include `safely_reclaimable_cache` (safely reclaimable cache space, indicating cache space that has been occupied but will not affect the normal completion of tasks in transit after cleanup (such as residual cache from completed tasks)) and `reserved_emergency_cache` (reserved emergency cache space, indicating cache space specifically reserved to cope with sudden traffic surges or anomalies, which should not be occupied by regular tasks). The resource reservation fields include `tenant_reservation_bytes`, the number of bytes reserved for a tenant, used to reserve cache space in advance for a specific tenant to ensure the minimum available resources for that tenant; and `collection_reservation_bytes`, the number of bytes reserved for an object collection, used to reserve cache space for a single upload task to ensure sufficient cache to complete the entire process after the task starts. The safe capacity field is `safe_completion_capacity`, the safe completion capacity, used to determine the available capacity threshold that the cache pool can guarantee for the normal completion of all tasks in transit after deducting emergency reservations, occupied, and mandatory reserved space. Therefore, the cache pool status includes at least the cache pool identifier, total cache capacity, used cache capacity, free cache capacity, cache capacity that can be safely reclaimed, regular reserved capacity, emergency reserved capacity, tenant reserved capacity, object collection reserved capacity, and safe completion cache capacity.

[0039] This embodiment ensures that the system can perform subsequent risk calculations with the goal of completing the sealing of the object set by simultaneously acquiring two types of states.

[0040] Step S20: Determine the remaining sealing workload corresponding to the unsealed object set based on the object set state.

[0041] The remaining sealing workload refers to the amount of data and objects that still need to be written before the unsealed object set can meet the sealing conditions and complete the state transition. It includes both the progress gap from normal writes and the gap from missing objects due to upload failures or omissions. It represents the workload required for writing, verification, persistence, or filling in missing objects before the object set can be sealed. This remaining sealing workload can be the remaining sealing workload for a single unsealed object set, or the total remaining sealing workload for all unsealed object sets.

[0042] In this embodiment, the remaining workload for sealing the unsealed object set can be determined by the expected workload of the unsealed object set and the completed workload. The expected workload can be determined based on the object list, upstream workflow declarations, historical statistics, or conservative quantiles, while the completed workload can be determined based on the number of uploaded objects, the number of bytes written, verification results, and persistence confirmation results. When calculating the remaining workload for sealing, the difference between the expected workload and the completed workload is used to obtain the remaining workload for sealing.

[0043] For example, in a data migration task, the system directly parses the object manifest provided by the upstream workflow, using the total number of files and bytes declared in the manifest as the expected workload. However, for real-time streaming write scenarios without a pre-defined manifest, the system cannot predict the total data volume. In this case, the expected workload can be determined by taking the P95 conservative quantile based on the time and throughput of similar tasks in historical statistics. For each successfully written object, the object gateway returns a persistent acknowledgment (ACK) and records the checksum (MD5). The system accumulates the number of uploaded objects and the number of bytes written in the completed workload. Finally, the difference between the two is used to obtain the remaining archiving workload.

[0044] Step S30: Compare the remaining sealing workload with the safe completion cache capacity of the control system, and determine the convergence risk of the unsealed object set based on the comparison results.

[0045] The "safe completion cache capacity" refers to the caching capacity of the management system used to complete object sealing or object state transitions. This safe completion cache capacity can be determined based on the cache pool's free cache capacity, safe reclaimable cache capacity, regular reserved capacity, emergency reserved capacity, and established cache reservations. It is understood that data in the sealing, recovery protection period, isolation zone, audit retention, or reserved state is not included in the safe completion cache capacity.

[0046] In this embodiment, the remaining workload of the object set dimension and the safety capacity of the cache pool dimension can be compared in multiple dimensions to output a binary classification of convergence risk level, such as convergent and non-convergent. Alternatively, a graded convergence risk level can be output to match control actions of different intensities and achieve refined control.

[0047] As an optional implementation, the safe completion cache capacity can be calculated using the cache pool status field, and this cache capacity can be used as the baseline threshold for risk assessment. Next, two comparative indicators are calculated: the maximum remaining workload of a single unsealed object set and the sum of the remaining workloads of all unsealed object sets. These two indicators are then compared to the safe completion cache capacity. Alternatively, the comparisons can be performed sequentially: first, the maximum remaining workload of a single unsealed object set is compared to the safe completion cache capacity; only when the maximum remaining workload of a single unsealed object set is less than the safe completion cache capacity is the total workload compared. Finally, based on the excess ratio of the comparison results, the corresponding convergence risk level is matched, and the final risk assessment result is output. Specifically, if a single remaining sealing workload exceeds the safe completion cache capacity, the target unsealed object set corresponding to that single remaining sealing workload is determined to be non-convergent; or if the total remaining sealing workload exceeds the safe completion cache capacity, all unsealed object sets are determined to be non-convergent.

[0048] For example, a cache pool has a total capacity of 200GB, and the calculated safe completion cache capacity is 80GB. In the current set of unsealed objects, the maximum remaining workload for a single set is 90GB, and the total remaining workload for all sets is 110GB. Both dimensions exceed the safe capacity, with the total exceeding the safe capacity by 37.5%. Therefore, the system determines that the current cache pool is at a high convergence risk level.

[0049] As an alternative implementation, when determining convergence risk, in addition to comparing workload and cache capacity, dynamic prediction can be made by combining write rate. This involves calculating the estimated time when cache resources will be fully occupied by combining the current remaining available cache with the average write rate of the business. This time-based prediction of convergence risk is a core indicator that reflects future cache occupancy trends. Specifically, the historical average write rate of each unsealed object set can be statistically analyzed. Combined with the remaining sealing workload of the corresponding set, the estimated time required for each set to complete all writes can be calculated. Then, the estimated cache exhaustion time can be calculated by combining the safe completion cache capacity with the current overall write rate. The estimated completion time of the object set is then compared with the cache exhaustion time and the sealing deadline. If the set completion time is later than the cache exhaustion time, convergence risk is identified, and the risk level is determined based on the time difference.

[0050] For example, a cache pool has a safe completion capacity of 80GB, a current total write rate of 10MB / s, and a calculated estimated cache exhaustion time of approximately 2.2 hours. A set of unsealed objects has a remaining workload of 50GB, an average write rate of 5MB / s, and an estimated completion time of approximately 2.8 hours, which is later than the cache exhaustion time. Therefore, the system determines that this set has a medium level of convergence risk.

[0051] This embodiment compares workload metrics with cache security capacity from multiple dimensions, transforming quantitative data into actionable risk decision-making results and improving decision-making accuracy.

[0052] Step S40: Perform risk control actions corresponding to the convergence risk.

[0053] In this embodiment, corresponding risk management strategies can be matched and executed based on the convergence risk level and type. For example, convergence risks can be mitigated from two dimensions: new object collection admission and existing object collection optimization, ensuring that the object set is sealed on schedule. Specifically, admission, rate limiting, routing, priority cache persistence, priority sealing, or alarm control can be executed according to the convergence risk of the object set. That is, the system performs hierarchical control according to the convergence risk level. Low-impact actions can include risk control actions such as alarms, queuing, or suggesting retry; while high-impact actions can include admission rejection, rate limiting, routing to other cache pools, priority cache persistence, or priority sealing, and recording corresponding audit information. It can be understood that admission is the system's process of determining whether the current cache resources can support the sealing of a new object set creation request or a new write request before receiving it, and restricting requests that do not meet security conditions.

[0054] In this embodiment, as an optional implementation, when the convergence risk is non-convergent, a corresponding new admission control strategy can be matched according to the current convergence risk status level. This allows for dynamic adjustment of control intensity based on the risk level, avoiding insufficient control at high risk and excessive control at low risk. In this way, for newly arrived object set creation requests and write requests, corresponding operations are performed according to the admission rules, including direct access, adding to the waiting queue, rate limiting, routing to other cache pools, and direct rejection.

[0055] Optionally, during the risk control process, the system can also monitor the capacity release of the cache pool in real time. When the capacity safely recovers to the corresponding threshold, requests in the waiting queue are released in sequence, and normal access is gradually restored.

[0056] For example, when the system determines a high convergence risk, a level-one control strategy is triggered, which suspends the acceptance of new object collection creation requests and adds new single object write requests to the waiting queue. When the risk is determined to be medium, a level-two control strategy is triggered, which limits the rate of new collection creation requests, allowing only 50% of the requests to enter, and adding the excess to the waiting queue. When an existing object collection is sealed and the corresponding cache space is released, the system detects that the safe completion capacity has rebounded above the threshold, automatically allowing the preceding requests in the waiting queue in order, and gradually restoring normal access.

[0057] As an alternative implementation, after the system locates a missing object entry in the object set, it proactively sends a notification containing details of the missing objects to the upstream business system, reminding the upstream system to re-upload the missing objects and expedite the collection sealing process. Alternatively, it can seal successfully uploaded valid objects into the database first, releasing the occupied cache resources, and handle the remaining missing objects separately later, thereby avoiding a small number of missing objects occupying the cache for a long time and slowing down the overall business convergence. Specifically, when the convergence risk is non-convergence, all unsealed object sets are traversed to identify abnormal sets with missing objects and writing stalls. Then, a detailed list of missing objects is generated for the abnormal sets, and a re-upload notification is sent to the upstream business system. At the same time, for sets with slow writing progress and estimated inability to complete sealing on schedule, their writing progress is accelerated by increasing the writing priority and allocating dedicated cache resources. For sets that still cannot complete the missing objects after exceeding the maximum waiting time, a partial sealing operation is performed, sealing the uploaded valid objects to the backend storage and releasing the corresponding cache space.

[0058] For example, when checking existing unsealed collections, the system finds that a certain collection has been stagnant for 12 hours without any new data being written, and 45 objects are missing after comparison with the list. The system first generates a detailed list of missing objects and sends a resubmission notification to the upstream business system; if the resubmitted data is not received 24 hours after the notification is sent, the system triggers a partial sealing mechanism, sealing the 955 uploaded objects into the database first, releasing the corresponding cache space, and the missing objects will be resubmitted by the upstream and processed separately.

[0059] This embodiment addresses convergence risks rapidly by simultaneously controlling new access controls and optimizing existing systems, achieving rapid closed-loop risk management and ensuring stable system operation. It also adapts to different business priorities and risk scenarios, balancing reliability and user experience.

[0060] For example, to help understand the implementation process of the object set sealing and convergence risk control method obtained by combining this embodiment with the first embodiment described above, please refer to... Figure 3 , Figure 3A simplified flowchart illustrating a method for managing convergence risks in object collection sealing is provided. Specifically: The system first acquires real-time status data of all in-transit object collections to calculate the remaining sealing workload for each object collection. Next, combining the total cache pool capacity, occupied space, various reserved resources, and the amount of cache that can be safely reclaimed, the system calculates the safe completion cache capacity of the current cache pool. Then, the system compares the global remaining sealing workload with the safe completion cache capacity to assess the convergence risk level of the cache resources. Finally, based on the risk result, corresponding control actions are executed, such as new task admission, upload rate limiting, cache reclamation, or task rejection, to ensure the stable completion of all upload sealing tasks within the cache pool.

[0061] This embodiment provides a method for managing the sealing and convergence risks of object collections. By completing the entire chain of full state collection, multi-method remaining workload calculation, multi-dimensional risk judgment, and hierarchical closed-loop handling, it breaks through the limitations of traditional solutions that rely solely on global cache static threshold alarms. It achieves fine-grained perception of sealing progress and advance prediction of convergence risks from the business dimension of object collections. Through targeted control measures, it effectively resolves the sealing blockage problem caused by cache resource constraints, ensures that object collections complete sealing and state transitions on schedule, avoids long-term invalid occupation of cache resources, and maximizes resource utilization efficiency while ensuring cache security. It also improves the operational stability and business adaptability of the object storage system's write business.

[0062] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, step S20 may include steps S21 to S23: Step S21: Determine the set of unsealed objects based on the lifecycle field in the object collection state; In this embodiment, based on the lifecycle status field in the object collection state, all collections in the unsealed state are first filtered out from the full object collection, and invalid collections that have been sealed or deleted are excluded, so as to avoid redundant calculations on invalid collections and improve the overall calculation efficiency.

[0063] Step S22: Calculate the expected amount of work completed and the amount of work already completed for the set of unsealed objects; Step S23: Based on the expected workload and the completed workload, determine the individual remaining workload for each set of unsealed objects, and the total remaining workload for all sets of unsealed objects.

[0064] In this embodiment, each set of unsealed objects contains both expected and completed workloads. The expected workload represents the uncompleted workload. Therefore, after determining the set of unsealed objects, the expected and completed workloads for each set are calculated. Subsequently, the remaining workload for each individual unsealed object set, as well as the total remaining workload for all unsealed object sets, are determined using these figures. The expected and completed workloads can be calculated based on an explicit object list, or, in the absence of an explicit object list, using a conservative estimation method based on a multi-source fusion strategy.

[0065] As an optional implementation for calculating the expected and completed workload of an unsealed object set, an object manifest corresponding to the unsealed object set can be obtained. When the object manifest is not empty, the expected workload is determined based on the expected number of objects, the expected total number of bytes, verification information, and / or completion markers in the object manifest. The completed workload is determined based on the number of uploaded objects, the number of bytes written, verification results, and persistence confirmation results. For example, the explicit object manifest corresponding to the unsealed object set can be extracted, and then the manifest content can be parsed to obtain the total number of expected objects and the expected total number of bytes for that set, which are taken as the expected workload for that set. Simultaneously, the completed workload is determined by statistically analyzing the number of objects successfully uploaded and the total number of bytes, the number of bytes written, the verification results of the written bytes, and the persistence confirmation results for that object set. During the calculation process, the expected workload can be subtracted from the uploaded workload to obtain the final remaining sealing workload for a single object set. In addition to calculating the remaining sealing workload for a single set of objects, it is also necessary to calculate the total remaining sealing workload for all unsealed sets of objects, which means adding up the remaining sealing workload for all unsealed sets of objects to obtain the total remaining sealing workload.

[0066] Furthermore, based on the list of objects in the unsealed object set, the set of missing objects between the unsealed object set and the uploaded object set can be determined. Then, based on the size, quantity, verification information, object version, and / or fragmentation information of the missing objects, the workload of the missing objects is determined. The expected workload is updated based on the workload of the missing objects, such as by comparing and locating the missing object entries one by one, and calculating the workload corresponding to the missing objects based on their size. By separating normal write gaps and abnormal missing gaps in the statistics, this method covers both regular progress gaps and the additional workload of abnormal scenarios such as network fluctuations and upload failures, ensuring the comprehensiveness of the calculation. During the update process, the remaining sealing workload of the object set is obtained by subtracting the uploaded workload from the expected workload and then adding the workload corresponding to the missing objects. This allows for the calculation of the difference to obtain a quantified progress gap, transforming the abstract sealing progress into a standardized indicator that can be directly compared with the cache capacity, providing a unified comparison basis for subsequent risk assessment.

[0067] Specifically, please refer to Figure 4 , Figure 4 This diagram illustrates gap risk identification based on a manifest. Specifically, after obtaining the Manifest object list of the object set or deriving the corresponding list, the successfully uploaded object set data is read. Missing object entries are located by comparing each entry, resulting in the missing object set (missingset). Then, considering the size, quantity, validation information, object version, and fragmentation information of the missing objects, the workload corresponding to each missing object is calculated. This separates normal write gaps from abnormal missing gaps, covering the additional workload from regular progress gaps and abnormal scenarios such as network fluctuations and upload failures. Subsequently, by subtracting the uploaded workload from the expected workload and then adding the missing object workload, the final remaining sealing workload of the object set is updated. This transforms the abstract sealing progress into a standardized indicator directly comparable to cache capacity, ultimately linking it to cache completion capabilities and providing a unified comparative basis for subsequent cache risk assessment.

[0068] For example, when a data backup service creates an object collection, it simultaneously uploads an explicit manifest. The manifest records that the collection should contain 1000 objects, with a total size of 100GB. After the system filters out this unsealed collection, it parses the manifest and finds that the expected workload is 100GB. It then counts that 850 objects have been successfully uploaded, with a total size of 82GB. A step-by-step comparison of the manifest reveals that 30 objects failed to upload due to network fluctuations, corresponding to a total size of 3GB. Finally, the remaining sealing workload for this collection is calculated to be 100GB - 82GB + 3GB = 21GB.

[0069] Alternatively, the object list can be, in addition to being an explicit list, at least one of the following: a marker object, a multipart upload list, a naming rule deduction list, and a workflow output list.

[0070] As an alternative implementation for calculating the expected and completed workload of an unsealed object set, a conservative estimation method based on a multi-source fusion strategy is used. Specifically, when the object list is empty, the expected and completed workloads are determined based on upstream workflow declarations, configuration files, historical statistics, tenant profiles, business type profiles, or conservative statistical quantiles of the unsealed object set. This strategy can utilize information from multiple data sources, such as upstream workflow declarations, system configuration defaults, historical business statistics, tenant profiles, and business type profiles, in descending order of priority, to estimate the expected workload of the object set, thus covering common business scenarios without an object list.

[0071] Specifically, the availability of each data source can be verified sequentially according to preset data source priority rules, and the highest priority available data source can be selected as the basis for estimating the expected workload. Then, based on the selected data source and combined with the historical write characteristics of the corresponding business, a conservative statistical quantile is used to calculate the expected workload for the object set, ensuring that cache resources always have sufficient slack to support the completion of the set's sealing, thus preventing cache deadlock from the calculation perspective. Finally, the currently uploaded workload of the object set is calculated, and combined with the estimated expected workload, an estimated value for the remaining sealing workload is obtained, and this result is labeled as the estimation method.

[0072] For example, a log-writing-in-real-time business object set has no explicit manifest, and the upstream workflow does not declare the total write volume. The system, following priority rules, sequentially verifies and selects the tenant's historical business statistics as the estimation basis; it then retrieves the size data of the tenant's object sets of similar businesses over the past 30 days, selecting the 95th percentile of 120GB as the expected total size. Currently, 70GB of this set has been uploaded, and the final estimated remaining archiving workload is 50GB, the result labeled as an estimate. Similarly, the completed workload is calculated based on historical statistics.

[0073] This embodiment provides a method for risk management of object set sealing and convergence. Based on precise calculation and conservative estimation, it covers both high-precision scenarios with inventory lists and general scenarios without inventory lists. In inventory list scenarios, it ensures calculation accuracy and improves the accuracy of risk assessment. In inventory list-less scenarios, conservative estimation ensures reliability and avoids missed risk assessments. This guarantees adaptability to all business scenarios.

[0074] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Furthermore, before step S30, steps S50-S60 are also included: Step S50: Obtain the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity of the cache pool of the management and control system.

[0075] The recyclable cache capacity includes the cache capacity that has been confirmed for persistence by the backend and meets the release conditions in the object collection. Idle cache capacity refers to the blank cache space in the cache pool that is not currently occupied by any object collection or business data. This space can directly accommodate newly written data and is available immediately without any additional operations; it is a core component of the cache pool's natively available resources. Reserved capacity refers to the dedicated guarantee capacity specifically allocated by the system for the object collection sealing and state transition processes. It is not included in the available capacity statistics of regular business operations and is only used to support the sealing and writing of object collections and state transition operations. It is specifically designed to cope with extreme scenarios such as sudden write peaks and temporary data overflows, preventing regular business operations from crowding out the cache resources required for sealing.

[0076] In this embodiment, the cache pool management module continuously monitors the total physical capacity and allocated / occupied capacity of the cache pool. By subtracting all occupied cache space from the total capacity, the current free cache capacity is calculated and synchronously updated in the cache pool status structure. Next, the cache pool management module conservatively iterates through all occupied cache shards, verifying the persistence confirmation status, compliance retention attributes, and tenant / collection reserved occupancy attributes of the data corresponding to each shard. The capacity of all shards meeting the safe reclamation conditions is summed to obtain the safe reclamation cache capacity. Subsequently, the system calculates the reserved capacity based on preset sealing guarantee configuration rules and the total cache pool capacity. This dedicated reservation provides a safety net for the sealing process, preventing fluctuations in regular business traffic from crowding out the cache resources needed for sealing, and ensuring the basic convergence capability of the sealing process from a resource allocation perspective.

[0077] For example, in an object storage cluster designed for archiving, the total physical capacity of the write cache pool is 200GB. The cache pool management module calculates that 110GB of the current capacity is occupied by various types of data, thus the free cache capacity is calculated to be 200GB - 110GB = 90GB. Then, after iterating through the 110GB of occupied space and verifying each fragment, it is confirmed that 30GB of data has received successful persistence confirmation from the backend, has no compliance retention requirements, and no dedicated reservation constraints; therefore, it is safe to reclaim this capacity, and the safe reclaimable cache capacity is 30GB. Next, the system configures and archives dedicated reserved capacity at a rate of 10% of the total capacity, calculating a reserved capacity of 20GB.

[0078] Step S60: Determine the sum of the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity as the safe cache capacity.

[0079] In this embodiment, the total capacity value is obtained by summing the three parameters: free cache capacity, safe reclaimable cache capacity, and reserved capacity. This aggregates the available capacity under three different attributes and usage conditions to obtain the overall sealing and protection capability of the cache pool, providing a unified quantitative benchmark for subsequent risk comparison. For example, if the system reads 90GB of free cache capacity, 30GB of safe reclaimable cache capacity, and 20GB of reserved capacity from the cache pool status, and performs the summation calculation: 90GB + 30GB + 20GB = 140GB, the current safe cache capacity of the cache pool is determined to be 140GB.

[0080] This embodiment uses three types of capacity summation operations to cover available cache resources with different attributes, improving resource utilization efficiency without increasing hardware investment and achieving a balance between resource efficiency and convergence safety.

[0081] This embodiment provides a method for managing the convergence risk of object collection sealing. By collecting data through classification and aggregating calculations, the safe completion cache capacity is obtained. This method can accurately quantify the total capacity of the cache pool that can support the sealing of object collections. It not only ensures data security and compliance requirements through the recyclable caliber, but also provides a backup resource guarantee for the sealing process through dedicated reserved capacity. The calculated threshold is accurate and reliable, providing a stable and accurate judgment benchmark for subsequent convergence risk assessment. This effectively improves the accuracy of risk identification and avoids misjudgment or omission of risks due to deviations in capacity caliber.

[0082] Based on the first embodiment of this application, in the fourth embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, after step S30, steps S70~S80 are also included: Step S70: Determine the expected cache exhaustion time based on the object write rate and cache persistence rate of the unsealed object set, the system's cache release rate, and the effective free capacity of the cache.

[0083] Object write rate is the total amount of valid data written to the cache pool from all unsealed object sets per unit time. It is the core source of cache usage growth, providing a direct reflection of the speed at which cache resources are consumed. Cache persistence rate is the speed at which data in the cache is flushed to backend permanent storage per unit time. It is one of the core ways to release cache space; once data persistence is confirmed, the corresponding cache space can enter a reclaimable state. Cache release rate is the total speed at which safe reclaimable space in the cache pool is cleaned up and converted into free, usable capacity per unit time. It covers scenarios such as space release after persistence and overall space release after object set sealing. The estimated cache exhaustion time is calculated linearly from the current effective free cache capacity and the net cache growth rate; it is the estimated time until the available cache capacity is completely consumed.

[0084] In this embodiment, the system calculates the average object write rate by statistically analyzing the total amount of data written to all unsealed object sets within a preset collection period. Simultaneously, it calculates the cache persistence rate and cache release rate by statistically analyzing the amount of data persisted in the cache and the total cache release capacity within the same period. Next, the system calculates the net cache growth rate by subtracting the total cache release rate from the total object write rate. If the net growth rate is less than or equal to zero, it indicates that the cache release rate is not lower than the write rate, the available capacity will not continue to decrease, there is no risk of exhaustion, and there is no need to continue calculating the exhaustion time. When the net cache growth rate is greater than zero, the system reads the current effective free cache capacity and divides it by the net cache growth rate to calculate the expected cache exhaustion time.

[0085] For example, the write cache pool of an object storage cluster currently has an effective free capacity of 80GB. The system collects data at a rate of 5 minutes. Within this period, the total write data volume of all unsealed object sets is 3GB, resulting in an average object write rate of 10MB / s. The total capacity of cache persistence and space release within the same period is 1.2GB. Since separate collection is not feasible, this is combined into a cache release rate of 4MB / s. The calculated net cache growth rate is 10MB / s - 4MB / s = 6MB / s, which is greater than zero. Dividing the effective free capacity of 80GB (equivalent to 81920MB) by 6MB / s yields an estimated cache exhaustion time of approximately 13653 seconds, or about 3.79 hours.

[0086] Step S80: Determine the estimated sealing time based on the remaining sealing workload of the unsealed object set and the object set upload rate.

[0087] The upload rate of an object collection is the amount of effective data successfully written to the cache per unit time for a single unsealed object collection. If the upload rate is zero, it means that the collection is in a write standstill state. The estimated sealing time is the estimated time required for a single object collection to complete all the remaining sealing work and meet the sealing conditions at the current upload rate.

[0088] In this embodiment, the system counts the number of successfully uploaded data within the corresponding collection period for each set of unsealed objects and calculates the average upload rate for that set. If the expected cache exhaustion time is less than the expected sealing time, then risk control actions corresponding to the convergence risk are executed.

[0089] Specifically, the flowchart comparing cache exhaustion time and archive time is as follows: Figure 5 As shown, the process includes two aspects. First, it collects the PUT (object write request) rate of data writing and the Flush (cache persistence) / Drain (cache release) rate of cache writes to disk, calculating the net cache growth rate and thus estimating the expected cache exhaustion time. Second, it collects the remaining workload of the object set and the current upload rate to calculate the expected sealing completion time. Then, it compares the two times to determine if the cache exhaustion time is earlier than the sealing time. If so, a high-risk control action is triggered; otherwise, monitoring continues or the status is determined to be low-risk.

[0090] This embodiment provides a method for managing convergence risk in object collections. By calculating the cache exhaustion time and object sealing time in a dual-time dimension, it upgrades static capacity threshold comparison to dynamic trend prediction. This method can more accurately identify the timing and scope of convergence risk, avoid unnecessary alarms and over-control caused by static thresholds, reduce interference with normal business operations, and also detect hidden risks such as slow writes and collection stagnation in advance. It allows sufficient response time for control actions, improves the accuracy and foresight of risk assessment, and better balances cache resource security and business write experience.

[0091] Based on the first embodiment of this application, in the fifth embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, after step S30, steps S90~S110 are also included: Step S90: When the convergence risk of the unsealed object is convergent, if an object set creation request or a workflow writing request is received, calculate the expected workload of the newly added object set corresponding to the object set creation request or workflow writing request.

[0092] When the convergence risk is convergent, it means that the total remaining workload for sealing all currently unsealed object sets does not exceed the safe completion cache capacity, and all existing sets can be sealed under cache resource constraints. The expected workload for a new object set refers to the total amount of cache data that the new object set to be added is expected to occupy from creation to completion of sealing.

[0093] In this embodiment, when the system verifies that the current cache pool is convergent, it can directly trigger new request control without proceeding to subsequent admission calculations. Specifically, when it is confirmed that the cache pool is convergent and a new object set creation request or workflow write request is received, the system uses a multi-source fusion strategy to calculate the expected workload of the new set. Similarly, if there is still no data, an estimate is made based on the tenant's historical data, business type profile, and conservative statistical quantiles.

[0094] Step S100: Obtain the remaining sealing workload for the entire set of unsealed objects.

[0095] After obtaining the expected workload of the newly added object set, the system iterates through all object sets in the cache pool whose lifecycle is in an unsealed state, reads the current remaining sealing workload value of each set one by one, and adds up the remaining sealing workload of all sets to obtain the total remaining sealing workload of the current full set of unsealed objects.

[0096] Step S110: If the sum of the expected completed workload and the remaining sealed workload is less than the product of the safe completion cache capacity and the admission safety coefficient, execute the admission process for the object set creation request or the workflow write request.

[0097] The admission safety factor is a system-preset safety redundancy factor of less than 1. It is used to reserve a certain proportion of buffer space after safely completing the cache capacity, in order to cope with uncertainties such as sudden write traffic, data volume estimation errors, and fluctuations in temporary cache usage, and to ensure overall convergence capability in extreme scenarios. The admission process refers to the complete operation in which the system allocates cache resources, initializes collection metadata, and formally integrates a new object collection creation or write request after it has passed the convergence verification, indicating that the request has been formally accepted by the system.

[0098] In this embodiment, the system reads the current safe completion cache capacity of the cache pool and multiplies it by a preset admission safety coefficient to calculate the admission capacity threshold. Then, it adds the sum of the remaining workload of the existing pool to the expected workload of the new pool to calculate the total sealing workload requirement (existing + incremental). If the total requirement is less than or equal to the threshold, it indicates that after the new collection is added, all collections still have overall convergence capability. The system executes the admission process, allocates cache resources to the new collection, and starts the write process. Similarly, if the total requirement is greater than the threshold, it indicates that the new request will disrupt the convergence capability of the existing collection. The system performs control actions such as rejection, queuing, suggesting retry, rate limiting, or routing to other cache pools based on business priority.

[0099] For example, the current safe completion cache capacity of the cache pool is 100GB, and the system's preset admission safety factor is 0.8. The calculated admission capacity threshold is 100GB × 0.8 = 80GB. The total remaining workload of existing unsealed collections is 50GB, and the expected workload of the new collection is 32GB, for a total workload requirement of 50GB + 32GB = 82GB. Since 82GB is greater than the 80GB admission threshold, it indicates that allowing the new collection would disrupt the overall convergence capability. Therefore, the system does not execute the admission process and writes the request to a waiting queue, allowing it to be released in order once existing collections are sealed and cache space is freed. If the expected workload of the new collection is 25GB, the total requirement is 50GB + 25GB = 75GB, which is less than the 80GB threshold. In this case, the system executes the admission process, creates the object collection, and allows writing.

[0100] This embodiment employs a global inventory plus incremental check approach, combined with an access control mechanism featuring redundancy in security coefficients, to move the control checkpoint forward, shifting from post-event cleanup alerts to pre-event access control. Compared to traditional access control based on the current cache level, this mechanism uses overall convergence as the core criterion, rather than current write space, thus fundamentally avoiding the problem of blocking while entering and ultimately deadlocking. Simultaneously, the security coefficient setting offsets the impact of workload estimation errors and traffic fluctuations, maximizing resource utilization while ensuring cache security.

[0101] For example, to help understand the implementation process of the object set sealing and convergence risk control method obtained by combining this embodiment with the first embodiment described above, please refer to... Figure 6 , Figure 6A simplified flowchart of admission control for a new object collection is provided. Specifically, when a new object collection creation request is received, the system first summarizes the remaining workload of all existing object collections, then estimates the total workload required to complete the entire process of the newly added object collection, and then determines whether the two exceed the safe completion capacity of the cache pool and the preset admission coefficient. If the determination result is safe, the new object collection is allowed to enter the cache pool to start the upload process. If the determination result is unsafe, the creation request is rejected, queued, or migrated to another cache resource pool.

[0102] This embodiment provides a method for managing the risk of cache convergence during object collection sealing. By implementing pre-emptive convergence admission verification, the control point for cache convergence shifts from handling existing risks to preventing new entry points. It uses overall convergence capability as the core judgment criterion, rather than solely relying on real-time cache levels. This allows for control over the total business load of the cache pool from the source, preventing deadlocks caused by new requests crowding out the sealing resources of the existing collection and hindering overall business convergence. Simultaneously, through multi-source workload estimation and redundant safety factor design, it ensures the accuracy of admission judgments while reserving buffer space to cope with business fluctuations. This maximizes the utilization efficiency of cache resources and the business access experience while ensuring cache operation security and timely business sealing.

[0103] This application provides a device for managing the risk of object set sealing and convergence. The device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the object set sealing and convergence risk management method described in the first embodiment above.

[0104] The following is for reference. Figure 7 This document illustrates a structural schematic diagram of a risk management device for sealing and converging object sets suitable for implementing embodiments of this application. The risk management device for sealing and converging object sets in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, personal digital assistants (PDAs), tablet computers (PADs), portable multimedia players (PMPs), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 7 The illustrated object set sealing and convergence risk management device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0105] like Figure 7 As shown, the object set sealing and convergence risk management device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The random access memory 1004 also stores various programs and data required for the operation of the object set sealing and convergence risk management device. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the input / output interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the object set sealing and convergence risk management device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows an object set sealing and convergence risk management device with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented or possessed alternatively.

[0106] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0107] The object set sealing and convergence risk management device provided in this application, employing the object set sealing and convergence risk management method in the above embodiments, can solve the technical problem of difficulty in sealing object sets on schedule. Compared with the prior art, the beneficial effects of the object set sealing and convergence risk management device provided in this application are the same as the beneficial effects of the object set sealing and convergence risk management method provided in the above embodiments, and other technical features in this object set sealing and convergence risk management device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0108] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0109] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0110] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the object set sealing and convergence risk management method in the above embodiments.

[0111] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory (EPROM, or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0112] The aforementioned computer-readable storage medium may be included in the object set sealing and convergence risk management device; or it may exist independently and not be assembled into the object set sealing and convergence risk management device.

[0113] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the object set sealing and convergence risk control device, cause the object set sealing and convergence risk control device to: Determine the set of unsealed objects based on the state of each object set; The remaining sealing workload corresponding to the unsealed object set is determined based on the state of the object set. The remaining sealing workload is compared with the safe completion cache capacity of the control system. Based on the comparison result, the convergence risk of the unsealed object set is determined. The safe completion cache capacity represents the caching capacity of the control system for completing object sealing or completing object state transition. Perform the risk control actions corresponding to the convergence risk.

[0114] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0116] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0117] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for managing the risk of sealing and converging object sets, thereby solving the technical problem of the difficulty in sealing object sets on schedule. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the method for managing the risk of sealing and converging object sets provided in the above embodiments, and will not be repeated here.

[0118] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for managing the convergence risk of object collection sealing, characterized in that, Applied to a management and control system, the method for managing the risk of sealing and convergence of the object set includes: Determine the set of unsealed objects based on the state of each object set; Determining the remaining sealing workload corresponding to the unsealed object set based on the object set state includes: determining the unsealed object set according to the lifecycle field in the object set state; Calculate the expected workload and the workload already completed for the set of unsealed objects; Based on the expected workload and the completed workload, determine the individual remaining workload for each set of unsealed objects, and the total remaining workload for all sets of unsealed objects. The remaining sealing workload is compared with the safe completion cache capacity of the control system. Based on the comparison result, the convergence risk of the unsealed object set is determined. The safe completion cache capacity represents the caching capacity of the control system for completing object sealing or completing object state transition. Perform the risk control actions corresponding to the convergence risk.

2. The method for managing the risk of sealing and convergence of object sets as described in claim 1, characterized in that, The steps for calculating the expected workload and the completed workload of the set of unsealed objects include: Obtain the list of objects corresponding to the set of unsealed objects; When the object list is not empty, the expected workload is determined based on the expected number of objects, the expected total number of bytes, the verification information, and / or the completion flag in the object list. The completed workload is determined based on the number of uploaded objects, the number of bytes written, the verification results, and the persistence confirmation results. When the object list is empty, the expected workload and the completed workload are determined based on the upstream workflow declaration, configuration file, historical statistics, tenant profile, business type profile or conservative statistical quantile of the unsealed object set.

3. The method for managing the risk of sealing and convergence of object sets as described in claim 2, characterized in that, After the step of determining the expected workload based on the expected number of objects, expected total bytes, verification information, and / or completion markers in the object list when the object list is not empty, the method for managing the risk of sealing and convergence of the object set further includes: Based on the object list of the unsealed object set, determine the missing object set between the unsealed object set and the uploaded object set; The workload of missing objects is determined based on the size, quantity, verification information, object version, and / or fragmentation information of the missing objects. Update the expected workload based on the workload of the missing objects.

4. The method for managing the risk of sealing and convergence of object sets as described in claim 1, characterized in that, The step of comparing the remaining sealing workload with the safe completion cache capacity of the control system, and determining the convergence risk of the unsealed object set based on the comparison result, includes: If a single remaining sealing workload exceeds the safe completion cache capacity, it is determined that the target unsealed object set corresponding to that single remaining sealing workload is non-convergent; or When the total amount of remaining sealing work exceeds the safe completion cache capacity, it is determined that the sets of unsealed objects are not convergent.

5. The method for managing the risk of sealing and convergence of object sets as described in claim 1, characterized in that, Before the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system, and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: Obtain the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity of the cache pool of the management and control system; The sum of the free cache capacity, the cache capacity that can be safely reclaimed, and the reserved capacity is determined as the safe completion cache capacity; The safe reclaimable cache capacity includes the cache capacity obtained from the object set that has been persistently confirmed by the backend and meets the release conditions.

6. The method for managing the risk of sealing and convergence of object sets as described in claim 1, characterized in that, After the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system, and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: Based on the object write rate and cache persistence rate of the unsealed object set, the cache release rate of the management system, and the effective free capacity of the cache, the expected cache exhaustion time is determined; The estimated sealing time is determined based on the remaining sealing workload of the unsealed object set and the upload rate of the object set; If the expected cache exhaustion time is less than the expected sealing time, execute the risk control action corresponding to the convergence risk.

7. The method for managing the risk of sealing and convergence of object sets as described in claim 1, characterized in that, After the step of comparing the remaining sealing workload with the safe completion cache capacity of the control system, and determining the convergence risk of the unsealed object set based on the comparison result, the sealing convergence risk control method for the object set further includes: When the convergence risk of the unsealed object is convergent, if an object set creation request or a workflow writing request is received, calculate the expected workload of the newly added object set corresponding to the object set creation request or the workflow writing request. Obtain the remaining sealing workload for the entire set of unsealed objects; If the sum of the expected workload to be completed and the remaining workload to be sealed is less than the product of the safe completion cache capacity and the admission safety coefficient, the admission process of the object set creation request or the workflow write request will be executed. Otherwise, data writing for the object collection creation request or the workflow write request will be prohibited.

8. A risk management device for sealing and converging object sets, characterized in that, The object set sealing and convergence risk management device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the object set sealing and convergence risk management method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the method for managing the risk of sealing and converging the object set as described in any one of claims 1 to 7.

Images