Storage area network oriented nvme device dynamic mapping system

Abstract

The application discloses a storage area network-oriented NVMe device dynamic mapping system and relates to the technical field of storage area network and high-performance storage management, and comprises a dynamic mapping construction module, access continuity indexes, release consistency indexes, bearing stability indexes and switching disturbance indexes are extracted from access relationship state data, release relationship state data and network bearing state data, a corresponding cost matrix is constructed through CRITIC objective weighting, Jonker-Volgenant cross-domain correspondence, access object association and current mapping are executed based on the matrix, and dynamic mapping baseline data are obtained. In the application, a unified dynamic mapping mechanism for host access relationship, release relationship and network bearing relationship is constructed, so that the dynamic mapping no longer depends on single-dimensional judgment, but can form a more coordinated mapping result under the joint action of multiple types of running states, thereby improving the integrity, accuracy and stability of the mapping establishment process and reducing mapping inaccuracy and corresponding deviation caused by relationship fragmentation.

Description

Technical Field

[0001] This invention relates to the fields of storage area networks and high-performance storage management technology, and in particular to a dynamic mapping system for NVMe devices for storage area networks. Background Technology

[0002] As the application of storage area networks and NVMe protocols deepens in high-performance storage scenarios, NVMe access mapping technology for storage area networks is gradually being used to support host access to namespace resources. Existing technologies typically perform mapping processing around host access sessions, subsystem publishing relationships, controller and target port visibility, and network path carrying status, and adjust the mapping relationship according to access requests and status changes to maintain the availability and continuity of the access process.

[0003] Existing NVMe access mapping technologies for storage area networks (SNRs) often handle host access, publishing relationships, and network bearers separately, lacking a continuous mapping mechanism around host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier. This makes it difficult for various relationships to move in stably. When the attach state, visibility state, publishing state, and link state change, it is difficult to form a dynamic mapping result that takes into account access continuity, publishing consistency, bearer stability, and handover disturbances, which can easily lead to inaccurate mapping, cross-domain mapping instability, and handover interruption. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a dynamic mapping system for NVMe devices in storage area networks.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a dynamic mapping system for NVMe devices in storage area networks, comprising: The access session aggregation module performs access object merging, session correspondence organization, and access object binding on the host NQN, initiating port identifier, connection session identifier, discovery connection return information, and host access request information to obtain access relationship status data. The publishing relationship awareness module performs publishing relationship association, attachment location organization, and visibility range aggregation on subsystem identifiers, namespace identifiers, controller identifiers, namespace attachment status, access visibility status, and target publishing status to obtain publishing relationship status data. The network bearer awareness module performs path correspondence organization, bearer continuity relationship expansion, and availability merging on target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status to obtain network bearer status data. The dynamic mapping construction module extracts access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators from access relationship status data, publication relationship status data, and network bearer status data. After objective weighting by CRITIC, it constructs the corresponding cost matrix and performs Jonker-Volgenant cross-domain mapping, access object association, and current mapping adjustment based on the matrix to obtain dynamic mapping baseline data. The candidate mapping generation module performs reachability filtering, conflict resolution, publication continuity verification, and candidate relationship sorting on the dynamic mapping baseline data to obtain candidate mapping result data. The phased switching control module performs mapping version generation, shadow mapping loading, phase confirmation switching, and old mapping release on dynamic mapping baseline data and candidate mapping result data to obtain mapping switching execution data. Access the continuity verification module to perform access continuity verification, switch result confirmation, and failure rollback triggering on the mapping switch execution data, connection session identifier, and host access request information, and obtain the mapping execution result data.

[0006] As a further description of the above technical solution: The access session aggregation module receives the host NQN, initiating port identifier, connection session identifier, discovery connection return information and host access request information. It performs corresponding reading of the host NQN, initiating port identifier, connection session identifier, discovery connection return information and host access request information according to the host NQN, so that the initiating port identifier, connection session identifier, discovery connection return information and host access request information corresponding to the same host NQN enter the same host access location. The access session aggregation module performs item-by-item correspondence processing on each connection session identifier under the same host NQN according to the initiating port identifier, so that each connection session identifier and its corresponding initiating port identifier are in a one-to-one correspondence, and different connection session identifiers are arranged consecutively in the access order under the same host NQN. The access session aggregation module performs synchronous correspondence and organization of the discovery connection return information and host access request information according to the connection session identifier, so that the access entry content in the discovery connection return information and the access target content in the host access request information maintain a corresponding relationship under the same connection session identifier, and the discovery connection return information and host access request information under the same connection session identifier maintain a continuous binding relationship. The access session aggregation module performs access object merging, session correspondence organization, and access object binding on the host NQN, initiating port identifier, connection session identifier, connection discovery return information, and host access request information that have been read and organized, generating access relationship status data.

[0007] As a further description of the above technical solution: The publishing relationship awareness module receives the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status. It performs corresponding readings on the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status according to the subsystem identifier, so that the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status under the same subsystem identifier are placed in the same publishing location. The release relationship awareness module performs item-by-item correspondence processing on the namespace attachment status and controller identifier according to the namespace identifier, so that each namespace identifier maintains a corresponding relationship with the corresponding namespace attachment status and controller identifier, and different namespace identifiers are arranged consecutively according to the attachment position under the same subsystem identifier. The publishing relationship awareness module performs synchronous correspondence and organization on the target publishing status according to the access visibility status, so that the access visibility status and the target publishing status maintain a corresponding relationship under the same namespace identifier and the same controller identifier, and the access visibility status and the namespace attachment status form a continuous connection relationship under the same publishing ownership location. The publishing relationship awareness module performs publishing relationship association processing, attachment location organization processing, and visibility range aggregation processing on the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status that have been read and organized, generating publishing relationship status data.

[0008] As a further description of the above technical solution: The network bearer awareness module receives the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. It performs corresponding readings on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status according to the target port identifier, so that the path identifier, link delay, transmission retry status, port congestion status, and ANA status under the same target port identifier enter the same bearer location. The network bearer perception module performs item-by-item correspondence processing on link delay, transmission retry status and port congestion status according to path identifier, so that each path identifier is associated with the corresponding link delay, transmission retry status and port congestion status, and different path identifiers are arranged continuously according to path position under the same target port identifier. The network bearer awareness module performs synchronous correspondence processing on the path identifier and the target port identifier according to the ANA status, so that the ANA status is consistent with the corresponding path identifier and the target port identifier, and the ANA status forms a continuous bearer relationship with link delay, transmission retry status and port congestion status under the same path identifier. The network bearer awareness module performs path correspondence processing, bearer continuity relationship expansion processing, and availability merging processing on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status that have been read and processed, generating network bearer status data.

[0009] As a further description of the above technical solution: The dynamic mapping construction module receives access relationship status data, publication relationship status data, and network bearer status data. It performs corresponding readings on the access relationship status data, publication relationship status data, and network bearer status data according to the host NQN and connection session identifier, so that the subsystem identifier, namespace identifier, controller identifier, target port identifier, and path identifier that are consecutively corresponding to the same host NQN and the same connection session identifier enter the same corresponding position. Based on host access request information, namespace attachment status, access visibility status, target publication status, link latency, transmission retry status, port congestion status, and ANA status, extract access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators from access relationship status data, publication relationship status data, and network bearer status data, and ensure that access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators maintain a continuous correspondence in the same corresponding position; Objective weighting based on the CRITIC algorithm is performed according to access continuity index, publication consistency index, bearer stability index and handover disturbance index to construct the corresponding cost matrix, and to ensure that each corresponding cost value in the corresponding cost matrix has a continuous correspondence with the host NQN, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding location. According to the corresponding cost matrix, the Jonker-Volgenant algorithm is used to perform cross-domain correspondence processing on the access relationship status data, the publication relationship status data and the network bearer status data, so that the host NQN, namespace identifier, controller identifier, target port identifier and path identifier form a continuous correspondence. Based on the host access request information, the content that has completed the cross-domain correspondence processing of the Jonker-Volgenant algorithm is processed to perform access object association processing and current mapping reorganization processing, so that the namespace identifier, controller identifier, target port identifier and path identifier that are continuously corresponding under different host NQN are arranged continuously in the corresponding order of the host access request information, and dynamic mapping baseline data is generated.

[0010] As a further description of the above technical solution: The objective weighting process based on the CRITIC algorithm is as follows: Synchronous reading of access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators is performed according to the same corresponding location, so that access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators maintain a continuous correspondence under the same host NQN, the same connection session identifier, the same namespace identifier, the same controller identifier, the same target port identifier, and the same path identifier; Perform consistency processing according to the value direction of access continuity indicators, publication consistency indicators, bearer stability indicators and handover disturbance indicators, so that access continuity indicators, publication consistency indicators, bearer stability indicators and handover disturbance indicators maintain a corresponding relationship in the same direction before entering the subsequent weighting process; Discreteness calculations are performed on the access continuity index, publication consistency index, bearer stability index, and handover disturbance index under the same corresponding location to maintain a continuous distinguishing relationship between the value differences of the access continuity index, publication consistency index, bearer stability index, and handover disturbance index. Perform correlation-based processing on access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators under the same corresponding location, so that access continuity indicators maintain a correlation with publication consistency indicators, bearer stability indicators, and handover disturbance indicators; publication consistency indicators maintain a correlation with bearer stability indicators and handover disturbance indicators; and bearer stability indicators maintain a correlation with handover disturbance indicators. Based on the results of the discreteness calculation and the results of the correlation, objective weighting is applied to the access continuity index, the publication consistency index, the bearer stability index, and the handover disturbance index, so that the access continuity index, the publication consistency index, the bearer stability index, and the handover disturbance index form a weight relationship that is continuously corresponding to the same corresponding position. According to the weight relationship, the access continuity index, publication consistency index, bearer stability index and handover disturbance index are combined and mapped so that the access continuity index, publication consistency index, bearer stability index and handover disturbance index under the same corresponding position jointly form the corresponding cost value in the corresponding cost matrix, and the corresponding cost value maintains a continuous correspondence with the host NQN, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding position.

[0011] As a further description of the above technical solution: The steps for handling cross-domain mappings in the Jonker-Volgenant algorithm are as follows: The corresponding cost values ​​in the corresponding cost matrix are read sequentially according to the same corresponding position, and each corresponding cost value is synchronized with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding position, so that each corresponding cost value has a complete correspondence when entering the cross-domain correspondence processing of the Jonker-Volgenant algorithm. According to the host NQN and connection session identifier, the namespace identifier, controller identifier, target port identifier and path identifier in each corresponding position are initially mapped and sorted, so that each host NQN and each connection session identifier is first mapped into a current corresponding position, and the remaining namespace identifier, controller identifier, target port identifier and path identifier are mapped into the corresponding position to be adjusted in the order of their corresponding values; Based on the corresponding value in the current corresponding position, the namespace identifier, controller identifier, target port identifier and path identifier under the same host NQN and the same connection session identifier are compared item by item. The content with the smaller corresponding value is retained in the current corresponding position, and the content with the larger corresponding value is removed from the current corresponding position and moved to the corresponding position to be adjusted. Based on the namespace identifier, controller identifier, target port identifier, and path identifier of the exiting current corresponding position, perform a continuous search on the remaining corresponding positions to be adjusted, so that the content of the exiting current corresponding position is compared with the corresponding value in different corresponding positions to be adjusted in turn, and a new continuous correspondence adjustment direction is formed in the comparison process; According to the adjustment direction formed by continuous search, the namespace identifier, controller identifier, target port identifier and path identifier in the current corresponding position and the corresponding position to be adjusted are replaced. The content that exits the current corresponding position and has a small corresponding value enters the new current corresponding position, and the replaced content continues to enter the remaining corresponding positions to be adjusted for subsequent comparison. Based on the corresponding generation value after position replacement, the namespace identifier, controller identifier, target port identifier, and path identifier under different host NQNs and different connection session identifiers are repeatedly adjusted so that each host NQN and each connection session identifier gradually forms a namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the smaller corresponding generation value. Perform conflict elimination and sorting according to the current corresponding positions after repeated adjustments, so that the same namespace identifier, the same controller identifier, the same target port identifier, or the same path identifier no longer occupies the same corresponding position between different host NQNs and different connection session identifiers, and return the conflicting content to the corresponding position to be adjusted to continue to perform continuous search and position replacement. The results of the current corresponding positions are merged according to the results of the conflict resolution and sorting, so that the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier and path identifier form a stable and continuous correspondence.

[0012] As a further description of the above technical solution: The candidate mapping generation module receives dynamic mapping baseline data and performs corresponding reading on the dynamic mapping baseline data according to the host NQN and host access request information, so that the namespace identifier, controller identifier, target port identifier and path identifier that are consecutively corresponding to the same host NQN and the same host access request information enter the same candidate corresponding position. Based on the access visibility status, target publication status, and namespace attachment status, reachability filtering is performed on the namespace identifiers and controller identifiers in the same candidate corresponding location to ensure that the access visibility status, target publication status, and namespace attachment status maintain a continuous correspondence with the namespace identifiers and controller identifiers in the same candidate corresponding location, and namespace identifiers and controller identifiers that satisfy the continuous correspondence are retained. Based on link delay, transmission retry status, port congestion status, and ANA status, bearer conflict elimination processing is performed on the target port identifier and path identifier in the same candidate corresponding location to ensure that the link delay, transmission retry status, port congestion status, and ANA status maintain a continuous bearer relationship with the target port identifier and path identifier in the same candidate corresponding location, and target port identifiers and path identifiers that are inconsistent with the continuous bearer relationship are excluded. Based on the host access request information, access visibility status, target publication status, and namespace attachment status, the retained namespace identifier and controller identifier are subjected to publication continuity verification processing to ensure that the namespace identifiers and controller identifiers corresponding to consecutively under the same host NQN maintain a continuous connection relationship at the corresponding position in the same host access request information. After completing reachability filtering, bearer conflict resolution, and publication continuity verification, the namespace identifier, controller identifier, target port identifier, and path identifier are sorted according to their corresponding order in the host NQN and host access request information, thus generating candidate mapping result data.

[0013] As a further description of the above technical solution: The phased switching control module receives dynamic mapping baseline data and candidate mapping result data, and performs corresponding reading on the dynamic mapping baseline data and candidate mapping result data according to the host NQN, connection session identifier and host access request information, so that the namespace identifier, controller identifier, target port identifier and path identifier corresponding to the same host NQN and the same connection session identifier respectively enter the same switching corresponding position. Mapping version generation processing is performed on the dynamic mapping baseline data and candidate mapping result data according to the namespace identifier, controller identifier, target port identifier and path identifier in the same switching location, so that the namespace identifier, controller identifier, target port identifier and path identifier in the dynamic mapping baseline data and the namespace identifier, controller identifier, target port identifier and path identifier in the candidate mapping result data form a continuous corresponding mapping version relationship under the same host NQN and the same connection session identifier; According to the mapping version relationship, the namespace identifier, controller identifier, target port identifier and path identifier in the same handover corresponding position are subjected to shadow mapping loading process, so that the namespace identifier, controller identifier, target port identifier and path identifier in the candidate mapping result data enter the shadow mapping corresponding position without deviating from the continuous correspondence of the dynamic mapping baseline data, and the shadow mapping corresponding position maintains a continuous connection relationship with the current handover corresponding position. According to the host access request information and connection session identifier, the namespace identifier, controller identifier, target port identifier and path identifier entering the corresponding position of the shadow mapping are processed by phase confirmation and switching, so that the namespace identifier, controller identifier, target port identifier and path identifier in the corresponding position of the shadow mapping replace the corresponding content in the dynamic mapping baseline data in sequence, and the replaced namespace identifier, controller identifier, target port identifier and path identifier maintain a continuous correspondence with the host NQN and connection session identifier. Based on the namespace identifier, controller identifier, target port identifier, and path identifier after the completion phase confirmation switchover process, the old mapping release process is performed, so that the namespace identifier, controller identifier, target port identifier, and path identifier that were replaced in the dynamic mapping baseline data are removed from the current switchover corresponding position, and the namespace identifier, controller identifier, target port identifier, and path identifier after the replacement are arranged continuously in the corresponding order of host NQN, connection session identifier, and host access request information, generating mapping switchover execution data.

[0014] As a further description of the above technical solution: The access continuous verification module receives the mapping switch execution data, connection session identifier, and host access request information. It performs corresponding reading on the mapping switch execution data according to the connection session identifier and host access request information, so that the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the same connection session identifier and the same host access request information enter the same verification corresponding position. The access continuity verification process is performed on the mapping switch execution data according to the host NQN, namespace identifier, controller identifier, target port identifier and path identifier in the same verification corresponding position, so that the namespace identifier, controller identifier, target port identifier and path identifier corresponding to the same connection session identifier and the same host access request information maintain a continuous correspondence before and after the switch, and the content that maintains the continuous correspondence is retained in the current verification corresponding position; Based on the host access request information and connection session identifier, perform a handover result confirmation process on the namespace identifier, controller identifier, target port identifier, and path identifier stored in the current verification corresponding position, so that the namespace identifier, controller identifier, target port identifier, and path identifier continuously corresponding to the same host NQN form a continuous confirmation relationship in the corresponding position of the current host access request information; Based on the access continuity verification processing result and the switching result confirmation processing result, the namespace identifier, controller identifier, target port identifier and path identifier that have not formed a continuous confirmation relationship are subjected to failure rollback trigger processing, so that the content that has not formed a continuous confirmation relationship is removed from the current verification corresponding position, and the namespace identifier, controller identifier, target port identifier and path identifier that are continuously corresponding to the same connection session identifier and the same host access request information are re-entered into the current verification corresponding position. The execution results of the host NQN, connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier after the completion of the handover result confirmation process or the failure rollback trigger process are merged. This process arranges the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different connection session identifiers in a continuous order according to the corresponding order of the host NQN and host access request information, generating mapping execution result data.

[0015] The present invention has the following beneficial effects: 1. In this invention, a unified dynamic mapping mechanism for host access relationships, publishing relationships, and network bearer relationships is first constructed. This improves the problem of scattered processing of various states and difficulty in stable linkage in the prior art. The dynamic mapping no longer relies on a single dimension judgment, but can form a more coordinated mapping result under the joint action of multiple operating states. This improves the integrity, accuracy and stability of the mapping establishment process and reduces the mapping inaccuracy and correspondence deviation caused by the fragmentation of relationships.

[0016] 2. In this invention, by introducing candidate screening, phased switching, access continuity verification and failure rollback control, the problems of abrupt switching, process instability and access interruption that are easy to occur in the mapping adjustment process of the prior art are improved. The dynamic mapping adjustment process is made smoother and more controllable, and can maintain good access continuity and switching reliability under state change conditions, thereby enhancing the system's adaptability and operational stability in complex operating environments. Attached Figure Description

[0017] Figure 1 This is a system architecture diagram of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Reference Figure 1 One embodiment of the present invention provides a dynamic mapping system for NVMe devices in a storage area network, comprising: The access session aggregation module performs access object merging, session correspondence organization, and access object binding on the host NQN, initiating port identifier, connection session identifier, discovery connection return information, and host access request information to obtain access relationship status data. The publishing relationship awareness module performs publishing relationship association, attachment location organization, and visibility range aggregation on subsystem identifiers, namespace identifiers, controller identifiers, namespace attachment status, access visibility status, and target publishing status to obtain publishing relationship status data. The network bearer awareness module performs path correspondence organization, bearer continuity relationship expansion, and availability merging on target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status to obtain network bearer status data. The dynamic mapping construction module extracts access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators from access relationship status data, publication relationship status data, and network bearer status data. After objective weighting by CRITIC, it constructs the corresponding cost matrix and performs Jonker-Volgenant cross-domain mapping, access object association, and current mapping adjustment based on the matrix to obtain dynamic mapping baseline data. The candidate mapping generation module performs reachability filtering, conflict resolution, publication continuity verification, and candidate relationship sorting on the dynamic mapping baseline data to obtain candidate mapping result data. The phased switching control module performs mapping version generation, shadow mapping loading, phase confirmation switching, and old mapping release on dynamic mapping baseline data and candidate mapping result data to obtain mapping switching execution data. Access the continuity verification module to perform access continuity verification, switch result confirmation, and failure rollback triggering on the mapping switch execution data, connection session identifier, and host access request information, and obtain the mapping execution result data.

[0020] In this embodiment, the access session aggregation module receives the host NQN, initiating port identifier, connection session identifier, connection discovery return information, and host access request information. During the host's connection initiation and access processes, it acquires the host NQN, initiating port identifier, connection session identifier, connection discovery return information, and host access request information. When a host initiates a connection, it acquires the host NQN, and after acquiring the host NQN, it synchronously acquires the initiating port identifier and connection session identifier continuously corresponding to that host NQN, ensuring a continuous correspondence between the host NQN, initiating port identifier, and connection session identifier before proceeding to subsequent processing. When the host performs connection discovery, it acquires the connection discovery return information and maintains a continuous correspondence between the connection discovery return information and the corresponding connection session identifier. When a host initiates access, it acquires the host access request information and maintains a continuous correspondence between the host access request information and the corresponding connection session identifier and connection discovery return information. It first performs a step-by-step reading of the host NQN, and synchronously reads the current host NQN. Retrieve the initiating port identifier, connection session identifier, discovery connection return information, and host access request information corresponding to the current host NQN; after reading the previous host NQN, continue reading the next host NQN, ensuring that the initiating port identifier, connection session identifier, discovery connection return information, and host access request information corresponding to the previous host NQN are sequentially expanded in the order of reading; during the reading of the same host NQN, ensure that the initiating port identifier, connection session identifier, discovery connection return information, and host access request information corresponding to that host NQN always enter the same host access location, and keep the content of different host NQNs separately arranged in their respective host access locations; after completing the corresponding reading of all host NQNs, ensure that the initiating port identifier, connection session identifier, discovery connection return information, and host access request information corresponding to the same host NQN form a continuous correspondence within the same host access location.

[0021] After completing the corresponding read, each connection session identifier under the same host NQN is processed item by item. During processing, the first connection session identifier under the host NQN is read first, and then the initiating port identifier corresponding to the first connection session identifier is read, so that the first connection session identifier and the initiating port identifier are kept in correspondence. Then, the next connection session identifier under the host NQN is read, and the initiating port identifier corresponding to the next connection session identifier is read simultaneously, so that the next connection session identifier and the corresponding initiating port identifier are kept in correspondence. When the previous connection session identifier and the next connection session identifier enter the processing process one after another, the previous connection session identifier and the next connection session identifier are arranged consecutively under the same host NQN in the order of access, and each connection session identifier is kept in a one-to-one correspondence with its corresponding initiating port identifier. After all connection session identifiers under the same host NQN are processed, the arrangement order between the connection session identifiers is consistent with the access order, and each connection session identifier maintains an independent and definite initiating port correspondence.

[0022] After establishing a one-to-one correspondence between each connection session identifier and its corresponding initiating port identifier, the discovery connection return information and host access request information are synchronously organized according to the connection session identifier. During the organization, the current connection session identifier is read first, followed by the discovery connection return information corresponding to the current connection session identifier, and the access entry content is read from the discovery connection return information. After reading the access entry content, the host access request information corresponding to the current connection session identifier is read, and the access target content is read from the host access request information, so that the access entry content and the access target content form a correspondence under the same connection session identifier. When reading the next connection session identifier, the discovery connection return information and host access request information corresponding to the next connection session identifier are read synchronously, so that the access entry content and access target content under the next connection session identifier enter the corresponding position of the next connection session. During the continuous organization of connection session identifiers, the discovery connection return information and host access request information under the same connection session identifier are kept continuously bound, and the discovery connection return information and host access request information under different connection session identifiers are continuously expanded according to the order of the connection session identifiers.

[0023] After completing the corresponding reading and organization of host NQN, initiating port identifier, connection session identifier, discovery connection return information, and host access request information, access object merging processing is performed on all organized content. During merging, initiating port identifiers, connection session identifiers, discovery connection return information, and host access request information that have entered the same host access location under the same host NQN are centrally merged according to the host access location, so that the access content corresponding to the same host NQN maintains a continuous aggregation relationship within the same merging location. After the content corresponding to one host NQN is merged, the same merging is performed on the content corresponding to the next host NQN, so that the content corresponding to different host NQNs forms separate merging results.

[0024] After completing the access object merging process, the merging results are processed for session correspondence organization. During organization, according to the access order of the connection session identifiers under the same host NQN, the order of each connection session identifier and its corresponding initiating port identifier, discovery connection return information, and host access request information are re-corrected and maintained, so that the same connection session identifier remains unchanged in the merging results with its original corresponding initiating port identifier, discovery connection return information, and host access request information. In the state of consecutively arranged connection session identifiers, each connection session identifier forms a continuous session correspondence within the same host NQN range, and this continuous session correspondence is kept consistent with the host access location.

[0025] After completing the session mapping and processing, the merged results are processed for access object binding. During processing, the access entry content and access target content under the same connection session identifier are bound together, and this binding relationship is fixed at the corresponding position of the connection session identifier. Then, the binding relationships formed under the previous connection session identifier and the binding relationships formed under the next connection session identifier are arranged continuously in the access order, so that the access entry content and access target content corresponding to each connection session identifier under the same host NQN form a continuous binding structure. After all connection session identifiers have completed the access object binding processing, a stable and continuous correspondence is formed between the host NQN, the initiating port identifier, the connection session identifier, the discovery connection return information, and the host access request information, generating access relationship status data.

[0026] After generating access relationship status data, the access relationship status data maintains the continuous merging relationship of the initiating port identifier, connection session identifier, discovery connection return information, and host access request information corresponding to the same host NQN. It also maintains the one-to-one correspondence between each connection session identifier and its respective initiating port identifier under the same host NQN, the continuous binding relationship between the access entry content and the access target content under the same connection session identifier, and the continuous session correspondence relationship of different connection session identifiers arranged in the access order. This allows the access relationship status data to directly carry out the subsequent processing flow in the claims.

[0027] In this embodiment, the publishing relationship awareness module receives subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status. During the target publishing and attachment processes, it acquires the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status. When a subsystem enters the publishing process, it acquires the subsystem identifier, and after acquiring the subsystem identifier, it simultaneously acquires the namespace identifier and controller identifier that continuously correspond to that subsystem identifier, ensuring a continuous correspondence between the subsystem identifier, namespace identifier, and controller identifier before subsequent processing. When a namespace enters the attachment process, it acquires the namespace attachment status and maintains a continuous correspondence between the namespace attachment status and the corresponding namespace identifier and controller identifier. During the target's external publishing process, it acquires the access visibility status and target publishing status, and maintains a continuous correspondence between the access visibility status, target publishing status, and the corresponding namespace identifier and controller identifier. The device identifiers maintain a continuous correspondence. First, all input content is read according to the subsystem identifier. During reading, the current subsystem identifier is read first, followed by the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publication status corresponding to the current subsystem identifier. This ensures that the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publication status corresponding to the current subsystem identifier are placed in the current publication location. After completing the reading of the content corresponding to the current subsystem identifier, the system continues to read the next subsystem identifier and its corresponding namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publication status. This ensures that the content corresponding to the next subsystem identifier is placed in the next publication location, and that the content corresponding to different subsystem identifiers is kept separate in its respective publication location. At the same time, the content under the same subsystem identifier forms a continuous correspondence within the same publication location.

[0028] After completing the corresponding reading according to the subsystem identifier, the namespace identifier, namespace attachment status, and controller identifier under the same subsystem identifier are processed item by item for correspondence. During processing, the first namespace identifier under the same subsystem identifier is read first, and then the namespace attachment status and controller identifier corresponding to the first namespace identifier are read to maintain the correspondence between the first namespace identifier and the corresponding namespace attachment status and controller identifier. Then, the next namespace identifier under the same subsystem identifier is read, and the namespace attachment status and controller identifier corresponding to the next namespace identifier are read simultaneously to maintain the correspondence between the next namespace identifier and the corresponding namespace attachment status and controller identifier. When the namespace identifiers of the preceding and following names enter the processing process one after another, the different namespace identifiers are arranged continuously according to the attachment position under the same subsystem identifier, and each namespace identifier maintains a one-to-one correspondence with the corresponding namespace attachment status and the corresponding controller identifier.

[0029] After completing the item-by-item mapping of namespace identifier, namespace attachment state, and controller identifier, the target publication state is synchronously mapped according to the access visibility state. During mapping, the current access visibility state is read first, and then the target publication state corresponding to the current access visibility state is read, so that the current access visibility state and the current target publication state maintain a mapping relationship under the current namespace identifier and the current controller identifier. After completing the mapping between the current access visibility state and the current target publication state, the next access visibility state and the next target publication state corresponding to the next access visibility state are read, so that the next access visibility state and the next target publication state maintain a mapping relationship under the next namespace identifier and the next controller identifier. During the continuous mapping of access visibility states, the access visibility state and the target publication state maintain a continuous mapping relationship under the same namespace identifier and the same controller identifier.

[0030] After the access visibility state and the target publication state are synchronized and aligned, the access visibility state and the namespace attachment state are made to form a continuous connection under the same publication location. During connection, the access visibility state in the current publication location and the namespace attachment state in the same publication location are kept in the same position, so that the access visibility state in the current publication location always corresponds to the namespace attachment state in the current publication location when entering subsequent processing. When continuing to process the access visibility state and namespace attachment state in the next publication location, the access visibility state in the next publication location and the namespace attachment state in the next publication location are kept in the same continuous connection, so that the access visibility state and namespace attachment state in different publication locations can be continuously expanded along the arrangement order of the publication locations.

[0031] After completing the corresponding reading, item-by-item correspondence organization, and synchronous correspondence organization, the publishing relationship association processing is performed on the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status. During the processing, the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status in the same publishing location are continuously associated according to the existing correspondence, so that the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status under the same subsystem identifier form a stable publishing correspondence relationship. After the association is completed in one publishing location, the same association is performed on the corresponding content in the next publishing location, so that the content corresponding to different subsystem identifiers forms separate publishing relationship association results.

[0032] After completing the publication relationship association processing, the namespace identifiers, namespace attachment states, and controller identifiers that have formed association results are processed to organize their attachment positions. During the processing, according to the order of attachment positions of each namespace identifier under the same subsystem identifier, the order of each namespace identifier and its corresponding namespace attachment state and controller identifier are maintained and the position is corrected, so that the attachment position order between the previous namespace identifier and the next namespace identifier remains continuous, and each namespace identifier always corresponds to its original namespace attachment state and controller identifier. After all namespace identifiers have completed the attachment position organization processing, the different namespace identifiers under the same subsystem identifier are arranged in a continuous relationship according to their attachment positions.

[0033] After completing the attachment location sorting process, the visibility range aggregation process is performed on the access visibility status and target publication status. During the process, the access visibility status and target publication status corresponding to the same publication location, namespace identifier, and controller identifier are aggregated according to the existing correspondence, so that the access visibility status and target publication status form a continuous aggregation relationship within the same publication location. After completing the aggregation of the access visibility status and target publication status in the current publication location, the same aggregation is performed on the access visibility status and target publication status in the next publication location, so that the access visibility status and target publication status in different publication locations form continuous and independent visibility range aggregation results. After completing the publication relationship association processing, attachment location sorting processing, and visibility range aggregation processing, publication relationship status data is generated.

[0034] After generating the publication relationship status data, the publication relationship status data maintains the correspondence between the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publication status under the same subsystem identifier, ensuring they all enter the same publication location. It also maintains a one-to-one correspondence between each namespace identifier and its corresponding namespace attachment status and controller identifier, the correspondence between the access visibility status and the target publication status under the same namespace identifier and controller identifier, and the continuous connection between the access visibility status and the namespace attachment status under the same publication location. This allows the publication relationship status data to directly enter the subsequent processing flow.

[0035] In this embodiment, the network bearer awareness module receives target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. During path bearer monitoring, it acquires the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. When the target port participates in bearer operations, it acquires the target port identifier, and after acquiring the target port identifier, it simultaneously acquires the path identifier continuously corresponding to that target port identifier, ensuring a continuous correspondence between the target port identifier and the path identifier before subsequent processing. During path bearer operations, it acquires link delay, transmission retry status, and port congestion status, maintaining a continuous correspondence between these statuses and their corresponding path identifiers. During changes in path availability, it acquires the ANA status, maintaining a continuous correspondence between the ANA status and its corresponding path identifier and target port identifier. First, it processes the target port identifier, path identifier, and link identifier according to the target port identifier. The system reads the corresponding information for latency, transmission retry status, port congestion status, and ANA status. During reading, it first reads the current target port identifier, then simultaneously reads the path identifier, link latency, transmission retry status, port congestion status, and ANA status corresponding to the current target port identifier, ensuring that these information are entered into the current bearer location. After completing the reading of the information corresponding to the current target port identifier, it continues to read the next target port identifier and its corresponding path identifier, link latency, transmission retry status, port congestion status, and ANA status, ensuring that the information corresponding to the next target port identifier is entered into the next bearer location. This process maintains that the information corresponding to different target port identifiers is arranged separately within their respective bearer locations, while ensuring that the path identifier, link latency, transmission retry status, port congestion status, and ANA status under the same target port identifier form a continuous correspondence within the same bearer location.

[0036] After completing the corresponding reading according to the target port identifier, the link delay, transmission retry status, and port congestion status are systematically organized according to the path identifier. During the organization, the first path identifier under the same target port identifier is read first, followed by the link delay, transmission retry status, and port congestion status corresponding to the first path identifier, ensuring that the first path identifier corresponds to the corresponding link delay, transmission retry status, and port congestion status. Subsequently, the next path identifier under the same target port identifier is read, and the link delay, transmission retry status, and port congestion status corresponding to the next path identifier are read simultaneously, ensuring that the next path identifier corresponds to the corresponding link delay, transmission retry status, and port congestion status. When the previous path identifier and the next path identifier are continuously entered into the organization process, the different path identifiers are arranged consecutively according to their path positions under the same target port identifier, and each path identifier maintains a one-to-one correspondence with the corresponding link delay, corresponding transmission retry status, and corresponding port congestion status.

[0037] After completing the item-by-item mapping of path identifiers, link delays, transmission retry statuses, and port congestion statuses, a synchronous mapping is performed on the path identifiers and target port identifiers according to the ANA status. During the mapping, the current ANA status is read first, followed by the current path identifier and current target port identifier corresponding to the current ANA status, ensuring that the current ANA status maintains a consistent relationship with its corresponding current path identifier and current target port identifier. After completing the mapping of the current ANA status, the next ANA status and its corresponding next path identifier and next target port identifier are read, ensuring that the next ANA status maintains a consistent relationship with its corresponding next path identifier and next target port identifier. During the continuous mapping of ANA statuses, different ANA statuses maintain a stable mapping with their corresponding path identifiers and target port identifiers, and the ANA statuses never deviate from their original mapping positions during the mapping process.

[0038] After ensuring the ANA state maintains a consistent relationship with its corresponding path identifier and target port identifier, the ANA state, link delay, transmission retry status, and port congestion status are made to form a continuous bearing relationship under the same path identifier. During this formation, the ANA state under the current path identifier is kept in the same position as the corresponding link delay, transmission retry status, and port congestion status under that current path identifier. This ensures that the ANA state under the current path identifier always corresponds to the link delay, transmission retry status, and port congestion status under the current path identifier during subsequent processing. When processing the content corresponding to the next path identifier, the ANA state under the next path identifier maintains the same continuous bearing relationship as the corresponding link delay, transmission retry status, and port congestion status under that next path identifier. This allows the ANA state, link delay, transmission retry status, and port congestion status under different path identifiers to be continuously unfolded along the path position arrangement order.

[0039] After completing the corresponding reading, item-by-item corresponding organization, and synchronous corresponding organization, path corresponding organization is performed on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. During the processing, each path identifier under the same target port identifier, arranged continuously according to its path position, and the corresponding link delay, transmission retry status, port congestion status, and ANA status are continuously organized according to the existing correspondence, so that each path identifier under the same target port identifier forms a clear path correspondence, and each path identifier always corresponds to the original link delay, transmission retry status, port congestion status, and ANA status. After the path corresponding organization is completed for the content corresponding to one target port identifier, the same processing is performed on the content corresponding to the next target port identifier, so that the content corresponding to different target port identifiers forms separate path corresponding organization results.

[0040] After completing the path mapping and organization process, the path identifier, link delay, transmission retry status, port congestion status, and ANA status are subjected to bearer continuity expansion processing. During the processing, the link delay, transmission retry status, port congestion status, and ANA status corresponding to the same path identifier are continuously expanded along the arrangement position of the path identifier, so that the link delay, transmission retry status, port congestion status, and ANA status under the current path identifier form a continuous bearer relationship. After the current path identifier is expanded, the same expansion is performed on the link delay, transmission retry status, port congestion status, and ANA status under the next path identifier, so that the bearer relationship under different path identifiers is expanded sequentially according to the path position, and the previous path identifier and the next path identifier maintain a continuous connection relationship.

[0041] After completing the bearer continuity expansion process, availability merging is performed on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. During this process, the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status in the same bearer location are centrally merged according to the existing correspondence, so that each path identifier under the same target port identifier and the corresponding link delay, transmission retry status, port congestion status, and ANA status form a continuous merging relationship. After completing the merging in the current bearer location, the same merging is performed on the corresponding content in the next bearer location, so that the content in different bearer locations forms continuous and independent availability merging results. After completing the path correspondence sorting process, bearer continuity expansion process, and availability merging process, network bearer status data is generated.

[0042] After generating network bearer status data, the network bearer status data maintains the correspondence between path identifiers, link delays, transmission retry statuses, port congestion statuses, and ANA statuses under the same target port identifier, ensuring they enter the same bearer location. It also maintains a one-to-one correspondence between each path identifier and its corresponding link delay, transmission retry status, and port congestion status, and maintains the consistency between the ANA status and its corresponding path identifier and target port identifier. Furthermore, it maintains the continuous bearer relationship between the ANA status and the link delay, transmission retry status, and port congestion status under the same path identifier, thereby enabling the network bearer status data to directly enter the subsequent processing flow.

[0043] In this embodiment, after receiving access relationship status data, publication relationship status data, and network bearer status data, the dynamic mapping construction module performs corresponding readings on the access relationship status data, publication relationship status data, and network bearer status data according to the host NQN and connection session identifier. This ensures that the subsystem identifier, namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the same host NQN and the same connection session identifier maintain the corresponding relationship, and that the subsystem identifier, namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to different host NQNs and different connection session identifiers are arranged continuously in the reading order.

[0044] After completing the corresponding read, based on the host access request information, namespace attach status, access visibility status, target publication status, link latency, transmission retry status, port congestion status, and ANA status, access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators are extracted from the access relationship status data, publication relationship status data, and network bearer status data. This ensures that the host access request information maintains a continuous correspondence with the namespace attach status, access visibility status, target publication status, link latency, transmission retry status, port congestion status, and ANA status, and that the access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators maintain a continuous correspondence with the corresponding host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier.

[0045] After extracting access continuity, publication consistency, bearer stability, and handover disturbance metrics, objective weighting based on the CRITIC algorithm is performed on these metrics. During this objective weighting process, the access continuity, publication consistency, bearer stability, and handover disturbance metrics are synchronously read according to the same correspondence, ensuring a continuous correspondence between these metrics and the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier. Synchronous reading: , : No. Synchronous read results under each correspondence : Correspondence sequence number. : No. The host NQN under the corresponding relationship, : No. Connection session identifiers under each mapping relationship : No. Namespace identifiers under each correspondence. : No. The controller identifier under each correspondence. : No. The target port identifier under each mapping relationship. : No. Path identifiers under each correspondence. : No. Access continuity indicators under a correspondence relationship : No. Release consistency metrics under each correspondence relationship : No. The load-bearing stability index under the corresponding relationship : No. The switching disturbance index under the correspondence.

[0046] After synchronous reading is complete, a consistency adjustment is performed according to the value direction of the access continuity index, publication consistency index, bearer stability index, and handover disturbance index. This ensures that the access continuity index, publication consistency index, bearer stability index, and handover disturbance index maintain a unidirectional correspondence across different host NQNs and different connection session identifiers, and that the access continuity index, publication consistency index, bearer stability index, and handover disturbance index are arranged consecutively in their original corresponding order. When the index is a positive index: When the indicator is a contrarian indicator: ; : No. Under the corresponding relationship, the first Consistency results of the indicators : No. Under the corresponding relationship, the first The original values ​​of the indicators, : Correspondence sequence number. : Indicator serial number, where Indicates access continuity metrics, This indicates the release of consistency metrics. Indicates the load-bearing stability index. This indicates a switch to a different disturbance indicator. Total number of corresponding relationships : No. The minimum value of the item indicator in all correspondences. : No. The maximum value of the item indicator in all corresponding relationships.

[0047] After standardization is completed, dispersion calculation is performed on the access continuity index, publication consistency index, bearer stability index, and handover disturbance index to ensure that the differences in the values ​​of each index remain continuously distinguishable, and that the differences in values ​​under different host NQNs and different connection session identifiers are arranged continuously in the order corresponding to the indices. Dispersion calculation processing: ; ; : No. The mean of the item indicator in all corresponding relationships. : No. The results of the dispersion calculation of the indicators. : No. Under the corresponding relationship, the first Consistency results of the indicators Total number of corresponding relationships : Correspondence sequence number. Indicator number.

[0048] After the discreteness calculation is completed, correlation mapping is performed according to the access continuity index, publication consistency index, bearer stability index, and handover disturbance index. This ensures that the access continuity index maintains a correlation with the publication consistency index, bearer stability index, and handover disturbance index; the publication consistency index maintains a correlation with the bearer stability index and handover disturbance index; and the bearer stability index maintains a correlation with the handover disturbance index. Furthermore, the correlation relationships formed under different host NQNs and different connection session identifiers are arranged sequentially according to their original order. Correlation mapping processing: ; : No. Item and the first The correlation between the indicators corresponds to the results. : No. Under the corresponding relationship, the first Consistency results of the indicators : No. Under the corresponding relationship, the first Consistency results of the indicators : No. The mean of the item indicator in all corresponding relationships. : No. The mean of the item indicator in all corresponding relationships. : Correspondence sequence number. The first indicator number. The second indicator number, : Total number of corresponding relationships.

[0049] After the results of the discreteness calculation and correlation correspondence processing are formed, objective weighting processing is performed on the access continuity index, publication consistency index, bearer stability index, and handover disturbance index according to these results. This ensures that the access continuity index, publication consistency index, bearer stability index, and handover disturbance index form a continuous weight relationship with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier, respectively, and that the weight relationships formed under different correspondences are arranged continuously in their original corresponding order. Objective weighting processing: ; ; : No. The amount of information in each indicator : No. The weighting relationship of the indicators : No. The results of the dispersion calculation of the indicators. : No. Item and the first The correlation between the indicators corresponds to the results. Indicator number : with the The index number used for relevant calculations.

[0050] After the weighting relationships are established, a combined mapping process is performed on the access continuity index, publication consistency index, bearer stability index, and handover disturbance index according to the weighting relationships. This ensures that the access continuity index, publication consistency index, bearer stability index, and handover disturbance index together form the corresponding cost values ​​in the corresponding cost matrix. The corresponding cost values ​​maintain a continuous correspondence with the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier. Simultaneously, the corresponding cost values ​​formed under different correspondence relationships are arranged continuously in the corresponding cost matrix according to their original correspondence order. Combined mapping process: ; ; : No. The corresponding value formed under a certain correspondence relationship : No. The weighting relationship of the indicators : No. Under the corresponding relationship, the first Consistency results of the indicators : Corresponding cost matrix, : No. line, number The corresponding value at the column position, : Correspondence sequence number. : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier. Total number of rows : Total number of columns.

[0051] After the corresponding cost matrix is ​​formed, the Jonker-Volgenant algorithm is used to perform cross-domain correspondence processing on the access relationship status data, publication relationship status data, and network bearer status data according to the corresponding cost matrix. When entering the Jonker-Volgenant algorithm cross-domain correspondence processing, the corresponding generation values ​​in the corresponding cost matrix are read sequentially according to the same correspondence relationship. Each corresponding generation value is synchronized with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier, ensuring that each corresponding generation value carries a complete correspondence relationship when entering the Jonker-Volgenant algorithm cross-domain correspondence processing. Sequential reading: ; ; : No. The sequential reading sequence corresponding to the cost values ​​of each row. : No. line, number The corresponding value at the column position, : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier. : List the total number, : An operation that sorts the data in ascending order and returns the original position sequence.

[0052] After sequential reading is completed, the namespace identifier, controller identifier, target port identifier, and path identifier are initially mapped according to the host NQN and connection session identifier, so that the host NQN and connection session identifier maintain the initial mapping relationship with the namespace identifier, controller identifier, target port identifier, and path identifier, and the namespace identifier, controller identifier, target port identifier, and path identifier are arranged consecutively in the order of their corresponding values.

[0053] After the initial mapping is completed, the namespace identifier, controller identifier, target port identifier, and path identifier under the same host NQN and the same connection session identifier are compared item by item according to their corresponding value. Items with lower corresponding values ​​are retained in the original mapping, while items with higher corresponding values ​​are removed from the original mapping and continue to participate in subsequent comparisons. Simultaneously, the comparison results under different host NQNs and different connection session identifiers are arranged consecutively according to the original mapping order. Initial mapping: ; ; : No. The row corresponds to the column number selected during the initial sorting phase. : No. Read the first column index of the sequence in row-by-row order. : The initial assignment variable corresponds to the rearranged assignment variable. : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of current namespace identifier, controller identifier, target port identifier, and path identifier. Item-by-item comparison: ; ; : No. During the item-by-item comparison, the first... The column index corresponding to the smaller value retained in the row. : No. During the item-by-item comparison, the first... The set of columns that a row can be compared with. : No. line, number The corresponding value at the column position, : No. During the item-by-item comparison, the first... Line 1 The difference between the relatively smaller corresponding cost values ​​in the column. : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier. : Compare round numbers.

[0054] After each comparison is completed, a sequential lookup is performed based on the namespace identifier, controller identifier, target port identifier, and path identifier that exited the original mapping. This ensures that the content exiting the original mapping is compared sequentially with the corresponding value in different mappings, establishing a direction for adjusting the mapping during the comparison process. Simultaneously, the adjustment directions formed under different host NQNs and different connection session identifiers are sequentially linked according to the original mapping order. Continuous Lookup: ; ; : No. The reduced cost value during consecutive searches. : No. line, number The corresponding value at the column position, : No. During the first consecutive search The momentum value of the line. : No. During the first consecutive search The column power value of the column. : No. During the first consecutive search The minimum relaxation of the column, : No. The set of rows that have already entered the search process during each consecutive search. : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier.

[0055] After a continuous search establishes an adjustment direction, position replacements are performed on the namespace identifier, controller identifier, target port identifier, and path identifier according to this direction. This allows content with lower cost to enter the new correspondence, while the replaced content continues to participate in subsequent comparisons. Furthermore, the position replacement results in different correspondences are arranged sequentially according to their original order. Position Replacement: ; : No. The assigned variable before the next position replacement. : No. The assignment variable after the second position replacement. : No. The adjustment path is formed by a series of consecutive searches. : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier.

[0056] After the location replacement is completed, the namespace identifier, controller identifier, target port identifier, and path identifier under different host NQNs and different connection session identifiers are repeatedly adjusted according to the corresponding value after the location replacement. This ensures that different host NQNs and different connection session identifiers gradually form namespace identifiers, controller identifiers, target port identifiers, and path identifiers that continuously correspond to the smaller corresponding value, and that the repeatedly adjusted content under different correspondences is continuously connected in the original corresponding order. Repeated adjustment: ; ; : No. Total agency value after repeated adjustments : No. line, number The corresponding value at the column position, : No. The assigned variable after repeated adjustments. Total number of rows : List the total number, Repeatedly adjust the round sequence number.

[0057] After the repetitive adjustments are completed, conflict resolution and sorting are performed according to the adjusted correspondences. This ensures that identical namespace identifiers, controller identifiers, target port identifiers, or path identifiers no longer occupy the same correspondence between different host NQNs and different connection session identifiers. Conflicting content is then reinstated for continuous lookup and position replacement, and the conflict-resolved content under different correspondences is arranged sequentially according to its original order. Conflict Resolution and Sorting: ; ; : Assignment variables under the current correspondence : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier. Total number of rows : Total number of columns.

[0058] After conflict resolution and reorganization, the results are merged according to the correspondence after conflict resolution and reorganization, so that host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier form a continuous correspondence, and the continuous correspondences formed under different host NQNs and different connection session identifiers are arranged in the original correspondence order. Result merging: ; ; The final set of correspondences after conflict resolution and organization. : The assignment variable under the final correspondence The final set of continuous correspondence results after merging. : No. The corresponding host NQN for the row, : No. The connection session identifier corresponding to the row. : No. The namespace identifier corresponding to the column, : No. The corresponding controller identifier for the column, : No. The target port identifier corresponding to the column. : No. The path identifier corresponding to the column, : The row number corresponding to the current host NQN and the current connection session identifier. : The column number corresponding to the combination of the current namespace identifier, controller identifier, target port identifier, and path identifier.

[0059] After completing the cross-domain mapping processing of the Jonker-Volgenant algorithm, access object association processing is performed on the content that has completed the cross-domain mapping processing of the Jonker-Volgenant algorithm according to the host access request information. When performing access object association processing, the host access request information is synchronously read with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier to maintain a continuous correspondence between the host access request information and the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier, and the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different host access request information are arranged continuously in the order of reading.

[0060] After synchronous reading is completed, the namespace identifier, controller identifier, target port identifier, and path identifier are bound to the host access request information to maintain a continuous correspondence between the host access request information and the namespace identifier, controller identifier, target port identifier, and path identifier, and to ensure that the namespace identifier, controller identifier, target port identifier, and path identifier under different host access request information are connected in the original corresponding order.

[0061] After the corresponding binding is completed, cross-correction is performed on the host access request information, namespace identifier, controller identifier, target port identifier, and path identifier according to the host NQN and connection session identifier. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the host access request information maintain a unique correspondence under the host NQN and connection session identifier, and that the unique correspondence formed under different host NQNs and different connection session identifiers is arranged continuously in the original correspondence order.

[0062] After cross-correction, the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier are continuously associated and organized according to the host access request information. This ensures that the host access request information forms a complete access object association relationship with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier, and path identifier, and that the access object association relationships formed under different host access request information are continuously connected in the original corresponding order.

[0063] After the access object association processing is completed, the current mapping balancing process is performed on the content of the completed access object association processing according to the host access request information. During the current mapping balancing process, the execution positions of the host NQN, connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier are merged to ensure that the host NQN, connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier maintain the same balancing relationship, and the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different host access request information are arranged consecutively in balancing order.

[0064] After the location merging is completed, the connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier are arranged in the order of execution of the host NQN. This ensures that the connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to each other under the same host NQN are arranged continuously in the corresponding order of the host access request information, and that the arrangement content formed under different host NQNs is continuously connected.

[0065] After sequential alignment is completed, consistent alignment is performed on namespace identifiers, controller identifiers, target port identifiers, and path identifiers according to the connection session identifiers. This ensures that namespace identifiers, controller identifiers, target port identifiers, and path identifiers under the same connection session identifier maintain a continuous correspondence, and that the continuous correspondences formed under different connection session identifiers are connected in the original correspondence order.

[0066] After consistency reorganization is completed, the namespace identifier, controller identifier, target port identifier, and path identifier are processed according to the host NQN, connection session identifier, and host access request information. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to each other under different host NQNs are arranged in the corresponding order of the host access request information. The arranged namespace identifier, controller identifier, target port identifier, and path identifier maintain a stable correspondence with the host NQN and connection session identifier, generating dynamic mapping baseline data.

[0067] In this embodiment, after receiving the dynamic mapping baseline data, the candidate mapping generation module performs corresponding readings on the dynamic mapping baseline data according to the host NQN and host access request information, so that the namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the same host NQN and the same host access request information enter the same candidate corresponding position; when different host NQNs and different host access request information enter the corresponding readings, the namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to different host NQNs and different host access request information are arranged continuously in the reading order, and the namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the host NQN and host access request information in the dynamic mapping baseline data maintain the original continuity relationship.

[0068] After completing the corresponding read, reachability filtering is performed on the namespace identifiers and controller identifiers in the same candidate corresponding position according to the access visibility state, target publication state, and namespace attachment state. This ensures that the access visibility state, target publication state, and namespace attachment state maintain a continuous correspondence with the namespace identifiers and controller identifiers in the same candidate corresponding position, and retains the namespace identifiers and controller identifiers that satisfy the continuous correspondence. When performing reachability filtering on different candidate corresponding positions, the access visibility state, target publication state, and namespace attachment state maintain a continuous connection with the namespace identifiers and controller identifiers in different candidate corresponding positions, and the retained namespace identifiers and controller identifiers continue to maintain a continuous correspondence with the original host NQN and host access request information.

[0069] After completing the reachability filtering process, bearer conflict elimination processing is performed on the target port identifier and path identifier in the same candidate corresponding position according to the link delay, transmission retry status, port congestion status, and ANA status. This ensures that the link delay, transmission retry status, port congestion status, and ANA status maintain a continuous bearer relationship with the target port identifier and path identifier in the same candidate corresponding position, and excludes target port identifiers and path identifiers that are inconsistent with the continuous bearer relationship. When performing bearer conflict elimination processing in different candidate corresponding positions, the retained target port identifiers and path identifiers in different candidate corresponding positions are arranged continuously according to the corresponding order of host NQN and host access request information, and the namespace identifier, controller identifier, target port identifier, and path identifier maintain a continuous bearer relationship in the same candidate corresponding position.

[0070] After completing the bearer conflict resolution process, the retained namespace identifiers and controller identifiers are subjected to publication continuity verification processing according to the host access request information, access visibility status, target publication status, and namespace attachment status. This ensures that the namespace identifiers and controller identifiers corresponding to consecutive hosts under the same host NQN maintain a continuous connection at the corresponding position in the same host access request information. When performing publication continuity verification processing on different host NQNs and different host access request information, the namespace identifiers and controller identifiers maintain a continuous verification relationship with the access visibility status, target publication status, and namespace attachment status. Furthermore, the verified namespace identifiers and controller identifiers continue to maintain a continuous connection with the target port identifier and path identifier.

[0071] After completing reachability filtering, bearer conflict resolution, and publication continuity verification, candidate relationship sorting is performed according to namespace identifier, controller identifier, target port identifier, and path identifier. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier in different candidate positions are arranged consecutively according to the corresponding order of host NQN and host access request information. The consecutively arranged namespace identifier, controller identifier, target port identifier, and path identifier maintain a stable correspondence with the original host NQN and host access request information, generating candidate mapping result data.

[0072] After the candidate mapping result data is formed, the phased handover control module receives the dynamic mapping baseline data and the candidate mapping result data. It performs corresponding readings on the dynamic mapping baseline data and the candidate mapping result data according to the host NQN, connection session identifier, and host access request information, so that the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the same host NQN and the same connection session identifier enter the same handover corresponding position respectively. When performing corresponding readings on different host NQNs and different connection session identifiers, the namespace identifier, controller identifier, target port identifier, and path identifier in the dynamic mapping baseline data are arranged continuously with the namespace identifier, controller identifier, target port identifier, and path identifier in the candidate mapping result data according to the corresponding order of host NQN, connection session identifier, and host access request information, and the content entering the same handover corresponding position maintains a continuous continuity relationship.

[0073] After completing the corresponding read, the dynamic mapping baseline data and candidate mapping result data are processed for mapping version generation according to the namespace identifier, controller identifier, target port identifier and path identifier in the same handover corresponding position. This ensures that the namespace identifier, controller identifier, target port identifier and path identifier in the dynamic mapping baseline data and the namespace identifier, controller identifier, target port identifier and path identifier in the candidate mapping result data form a continuous corresponding mapping version relationship under the same host NQN and the same connection session identifier. When performing mapping version generation processing at different handover corresponding positions, the mapping version relationships formed under different host NQNs and different connection session identifiers are arranged continuously according to the corresponding order of host access request information, and the namespace identifier, controller identifier, target port identifier and path identifier in the mapping version relationship are kept in a continuous correspondence.

[0074] After the mapping version relationship is formed, shadow mapping loading is performed on the namespace identifier, controller identifier, target port identifier, and path identifier in the same handover corresponding position according to the mapping version relationship. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier in the candidate mapping result data enter the shadow mapping corresponding position without deviating from the continuous correspondence of the dynamic mapping baseline data, and that the shadow mapping corresponding position maintains a continuous connection with the current handover corresponding position. When shadow mapping loading is performed at different handover corresponding positions, the namespace identifier, controller identifier, target port identifier, and path identifier in different shadow mapping corresponding positions are arranged continuously according to the corresponding order of host NQN, connection session identifier, and host access request information, and the content in the shadow mapping corresponding position maintains a continuous connection with the original dynamic mapping baseline data.

[0075] After the shadow mapping loading process is completed, a phase confirmation and switching process is performed on the namespace identifier, controller identifier, target port identifier, and path identifier at the corresponding location of the shadow mapping according to the host access request information and connection session identifier. This process ensures that the namespace identifier, controller identifier, target port identifier, and path identifier at the corresponding location of the shadow mapping sequentially replace the corresponding content in the dynamic mapping baseline data, and that the replaced namespace identifier, controller identifier, target port identifier, and path identifier maintain a continuous correspondence with the host NQN and connection session identifier. When performing phase confirmation and switching processes at different corresponding locations of the shadow mapping, the replaced namespace identifier, controller identifier, target port identifier, and path identifier are arranged continuously according to the corresponding order of the host access request information, and the replaced content maintains a continuous connection with the unreplaced content.

[0076] After the phase confirmation switchover process is completed, the old mapping release process is performed according to the namespace identifier, controller identifier, target port identifier, and path identifier. This causes the namespace identifier, controller identifier, target port identifier, and path identifier that were replaced in the dynamic mapping baseline data to exit the current switchover corresponding position. The namespace identifier, controller identifier, target port identifier, and path identifier after replacement are arranged continuously in the corresponding order of host NQN, connection session identifier, and host access request information. After the old mapping release process is completed at all switchover corresponding positions, the remaining namespace identifier, controller identifier, target port identifier, and path identifier maintain a stable correspondence with the corresponding host NQN, connection session identifier, and host access request information, and the mapping switchover execution data is generated.

[0077] After the mapping switch execution data is formed, the access continuity verification module receives the mapping switch execution data, connection session identifier, and host access request information. It performs corresponding readings on the mapping switch execution data according to the connection session identifier and host access request information, so that the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the same connection session identifier and the same host access request information are placed in the same verification corresponding position. When performing corresponding readings with different connection session identifiers and different host access request information, the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different connection session identifiers and different host access request information are arranged continuously in the reading order, and the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier in the same verification corresponding position are kept in a continuous relationship.

[0078] After completing the corresponding read, the access continuity verification process is performed on the mapping switch execution data according to the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier in the same verification corresponding position. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the same connection session identifier and the same host access request information maintain a continuous correspondence before and after the switch, and the content that maintains a continuous correspondence is retained in the current verification corresponding position. When performing access continuity verification processing in different verification corresponding positions, the namespace identifier, controller identifier, target port identifier, and path identifier that maintain a continuous correspondence are arranged continuously according to the corresponding order of the connection session identifier and the host access request information, and the content that does not maintain a continuous correspondence continues to participate in subsequent processing.

[0079] After the access continuity verification process is completed, the namespace identifier, controller identifier, target port identifier, and path identifier stored in the current verification location are processed according to the host access request information and connection session identifier. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the same host NQN form a continuous confirmation relationship in the current host access request information. When the handover result confirmation process is performed in different verification locations, the namespace identifier, controller identifier, target port identifier, and path identifier that form a continuous confirmation relationship are arranged continuously according to the corresponding order of the host NQN and the host access request information, and the continuous confirmation relationship is kept in line with the connection session identifier.

[0080] After the handover result confirmation process is completed, based on the access continuity verification process result and the handover result confirmation process result, a failure rollback trigger process is executed for namespace identifiers, controller identifiers, target port identifiers, and path identifiers that do not form a continuous confirmation relationship. This causes the content that does not form a continuous confirmation relationship to exit the current verification corresponding position, and allows the namespace identifiers, controller identifiers, target port identifiers, and path identifiers that continuously correspond to the same connection session identifier and the same host access request information to re-enter the current verification corresponding position. When the failure rollback trigger process is executed at different verification corresponding positions, the namespace identifiers, controller identifiers, target port identifiers, and path identifiers that re-enter the current verification corresponding position are arranged continuously according to the corresponding order of the connection session identifier and the host access request information, and the content that re-enters the current verification corresponding position maintains a continuous connection relationship with the content that maintains a continuous confirmation relationship.

[0081] After completing the handover result confirmation process or failure rollback trigger process, the execution results are merged according to the host NQN, connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier. This ensures that the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different connection session identifiers are arranged consecutively in the order of the host NQN and host access request information. The arranged namespace identifier, controller identifier, target port identifier, and path identifier maintain a stable correspondence with the host NQN, connection session identifier, and host access request information, generating mapping execution result data.

[0082] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A dynamic mapping system for NVMe devices in storage area networks, characterized in that: include: The access session aggregation module performs access object merging, session correspondence organization, and access object binding on the host NQN, initiating port identifier, connection session identifier, discovery connection return information, and host access request information to obtain access relationship status data. The publishing relationship awareness module performs publishing relationship association, attachment location organization, and visibility range aggregation on subsystem identifiers, namespace identifiers, controller identifiers, namespace attachment status, access visibility status, and target publishing status to obtain publishing relationship status data. The network bearer awareness module performs path correspondence organization, bearer continuity relationship expansion, and availability merging on target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status to obtain network bearer status data. The dynamic mapping construction module extracts access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators from access relationship status data, publication relationship status data, and network bearer status data. After objective weighting by CRITIC, it constructs the corresponding cost matrix and performs Jonker-Volgenant cross-domain mapping, access object association, and current mapping adjustment based on the matrix to obtain dynamic mapping baseline data. The candidate mapping generation module performs reachability filtering, conflict resolution, publication continuity verification, and candidate relationship sorting on the dynamic mapping baseline data to obtain candidate mapping result data. The phased switching control module performs mapping version generation, shadow mapping loading, phase confirmation switching, and old mapping release on dynamic mapping baseline data and candidate mapping result data to obtain mapping switching execution data. Access the continuity verification module to perform access continuity verification, switch result confirmation, and failure rollback triggering on the mapping switch execution data, connection session identifier, and host access request information, and obtain the mapping execution result data.

2. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The access session aggregation module receives the host NQN, initiating port identifier, connection session identifier, discovery connection return information and host access request information. It performs corresponding reading of the host NQN, initiating port identifier, connection session identifier, discovery connection return information and host access request information according to the host NQN, so that the initiating port identifier, connection session identifier, discovery connection return information and host access request information corresponding to the same host NQN enter the same host access location. The access session aggregation module performs item-by-item correspondence processing on each connection session identifier under the same host NQN according to the initiating port identifier, so that each connection session identifier and its corresponding initiating port identifier are in a one-to-one correspondence, and different connection session identifiers are arranged consecutively in the access order under the same host NQN. The access session aggregation module performs synchronous correspondence and organization of the discovery connection return information and host access request information according to the connection session identifier, so that the access entry content in the discovery connection return information and the access target content in the host access request information maintain a corresponding relationship under the same connection session identifier, and the discovery connection return information and host access request information under the same connection session identifier maintain a continuous binding relationship. The access session aggregation module performs access object merging, session correspondence organization, and access object binding on the host NQN, initiating port identifier, connection session identifier, connection discovery return information, and host access request information that have been read and organized, generating access relationship status data.

3. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The publishing relationship awareness module receives the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status. It performs corresponding readings on the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status according to the subsystem identifier, so that the namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status under the same subsystem identifier are placed in the same publishing location. The release relationship awareness module performs item-by-item correspondence processing on the namespace attachment status and controller identifier according to the namespace identifier, so that each namespace identifier maintains a corresponding relationship with the corresponding namespace attachment status and controller identifier, and different namespace identifiers are arranged consecutively according to the attachment position under the same subsystem identifier. The publishing relationship awareness module performs synchronous correspondence and organization on the target publishing status according to the access visibility status, so that the access visibility status and the target publishing status maintain a corresponding relationship under the same namespace identifier and the same controller identifier, and the access visibility status and the namespace attachment status form a continuous connection relationship under the same publishing ownership location. The publishing relationship awareness module performs publishing relationship association processing, attachment location organization processing, and visibility range aggregation processing on the subsystem identifier, namespace identifier, controller identifier, namespace attachment status, access visibility status, and target publishing status that have been read and organized, generating publishing relationship status data.

4. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The network bearer awareness module receives the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status. It performs corresponding readings on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status according to the target port identifier, so that the path identifier, link delay, transmission retry status, port congestion status, and ANA status under the same target port identifier enter the same bearer location. The network bearer perception module performs item-by-item correspondence processing on link delay, transmission retry status and port congestion status according to path identifier, so that each path identifier is associated with the corresponding link delay, transmission retry status and port congestion status, and different path identifiers are arranged continuously according to path position under the same target port identifier. The network bearer awareness module performs synchronous correspondence processing on the path identifier and the target port identifier according to the ANA status, so that the ANA status is consistent with the corresponding path identifier and the target port identifier, and the ANA status forms a continuous bearer relationship with link delay, transmission retry status and port congestion status under the same path identifier. The network bearer awareness module performs path correspondence processing, bearer continuity relationship expansion processing, and availability merging processing on the target port identifier, path identifier, link delay, transmission retry status, port congestion status, and ANA status that have been read and processed, generating network bearer status data.

5. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The dynamic mapping construction module receives access relationship status data, publication relationship status data, and network bearer status data. It performs corresponding readings on the access relationship status data, publication relationship status data, and network bearer status data according to the host NQN and connection session identifier, so that the subsystem identifier, namespace identifier, controller identifier, target port identifier, and path identifier that are consecutively corresponding to the same host NQN and the same connection session identifier enter the same corresponding position. Based on host access request information, namespace attachment status, access visibility status, target publication status, link latency, transmission retry status, port congestion status, and ANA status, extract access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators from access relationship status data, publication relationship status data, and network bearer status data, and ensure that access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators maintain a continuous correspondence in the same corresponding position; Objective weighting based on the CRITIC algorithm is performed according to access continuity index, publication consistency index, bearer stability index and handover disturbance index to construct the corresponding cost matrix, and to ensure that each corresponding cost value in the corresponding cost matrix has a continuous correspondence with the host NQN, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding location. According to the corresponding cost matrix, the Jonker-Volgenant algorithm is used to perform cross-domain correspondence processing on the access relationship status data, the publication relationship status data and the network bearer status data, so that the host NQN, namespace identifier, controller identifier, target port identifier and path identifier form a continuous correspondence. Based on the host access request information, the content that has completed the cross-domain correspondence processing of the Jonker-Volgenant algorithm is processed to perform access object association processing and current mapping reorganization processing, so that the namespace identifier, controller identifier, target port identifier and path identifier that are continuously corresponding under different host NQN are arranged continuously in the corresponding order of the host access request information, and dynamic mapping baseline data is generated.

6. The NVMe device dynamic mapping system for storage area networks according to claim 5, characterized in that: The objective weighting process based on the CRITIC algorithm is as follows: Synchronous reading of access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators is performed according to the same corresponding location, so that access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators maintain a continuous correspondence under the same host NQN, the same connection session identifier, the same namespace identifier, the same controller identifier, the same target port identifier, and the same path identifier; Perform consistency processing according to the value direction of access continuity indicators, publication consistency indicators, bearer stability indicators and handover disturbance indicators, so that access continuity indicators, publication consistency indicators, bearer stability indicators and handover disturbance indicators maintain a corresponding relationship in the same direction before entering the subsequent weighting process; Discreteness calculations are performed on the access continuity index, publication consistency index, bearer stability index, and handover disturbance index under the same corresponding location to maintain a continuous distinguishing relationship between the value differences of the access continuity index, publication consistency index, bearer stability index, and handover disturbance index. Perform correlation-based processing on access continuity indicators, publication consistency indicators, bearer stability indicators, and handover disturbance indicators under the same corresponding location, so that access continuity indicators maintain a correlation with publication consistency indicators, bearer stability indicators, and handover disturbance indicators; publication consistency indicators maintain a correlation with bearer stability indicators and handover disturbance indicators; and bearer stability indicators maintain a correlation with handover disturbance indicators. Based on the results of the discreteness calculation and the results of the correlation, objective weighting is applied to the access continuity index, the publication consistency index, the bearer stability index, and the handover disturbance index, so that the access continuity index, the publication consistency index, the bearer stability index, and the handover disturbance index form a weight relationship that is continuously corresponding to the same corresponding position. According to the weight relationship, the access continuity index, publication consistency index, bearer stability index and handover disturbance index are combined and mapped so that the access continuity index, publication consistency index, bearer stability index and handover disturbance index under the same corresponding position jointly form the corresponding cost value in the corresponding cost matrix, and the corresponding cost value maintains a continuous correspondence with the host NQN, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding position.

7. The NVMe device dynamic mapping system for storage area networks according to claim 5, characterized in that: The steps for handling cross-domain mappings in the Jonker-Volgenant algorithm are as follows: The corresponding cost values ​​in the corresponding cost matrix are read sequentially according to the same corresponding position, and each corresponding cost value is synchronized with the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier and path identifier under the same corresponding position, so that each corresponding cost value has a complete correspondence when entering the cross-domain correspondence processing of the Jonker-Volgenant algorithm. According to the host NQN and connection session identifier, the namespace identifier, controller identifier, target port identifier and path identifier in each corresponding position are initially mapped and sorted, so that each host NQN and each connection session identifier is first mapped into a current corresponding position, and the remaining namespace identifier, controller identifier, target port identifier and path identifier are mapped into the corresponding position to be adjusted in the order of their corresponding values; Based on the corresponding value in the current corresponding position, the namespace identifier, controller identifier, target port identifier and path identifier under the same host NQN and the same connection session identifier are compared item by item. The content with the smaller corresponding value is retained in the current corresponding position, and the content with the larger corresponding value is removed from the current corresponding position and moved to the corresponding position to be adjusted. Based on the namespace identifier, controller identifier, target port identifier, and path identifier of the exiting current corresponding position, perform a continuous search on the remaining corresponding positions to be adjusted, so that the content of the exiting current corresponding position is compared with the corresponding value in different corresponding positions to be adjusted in turn, and a new continuous correspondence adjustment direction is formed in the comparison process; According to the adjustment direction formed by continuous search, the namespace identifier, controller identifier, target port identifier and path identifier in the current corresponding position and the corresponding position to be adjusted are replaced. The content that exits the current corresponding position and has a small corresponding value enters the new current corresponding position, and the replaced content continues to enter the remaining corresponding positions to be adjusted for subsequent comparison. Based on the corresponding generation value after position replacement, the namespace identifier, controller identifier, target port identifier, and path identifier under different host NQNs and different connection session identifiers are repeatedly adjusted so that each host NQN and each connection session identifier gradually forms a namespace identifier, controller identifier, target port identifier, and path identifier that are continuously corresponding to the smaller corresponding generation value. Perform conflict elimination and sorting according to the current corresponding positions after repeated adjustments, so that the same namespace identifier, the same controller identifier, the same target port identifier, or the same path identifier no longer occupies the same corresponding position between different host NQNs and different connection session identifiers, and return the conflicting content to the corresponding position to be adjusted to continue to perform continuous search and position replacement. The results of the current corresponding positions are merged according to the results of the conflict resolution and sorting, so that the host NQN, connection session identifier, namespace identifier, controller identifier, target port identifier and path identifier form a stable and continuous correspondence.

8. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The candidate mapping generation module receives dynamic mapping baseline data and performs corresponding reading on the dynamic mapping baseline data according to the host NQN and host access request information, so that the namespace identifier, controller identifier, target port identifier and path identifier that are consecutively corresponding to the same host NQN and the same host access request information enter the same candidate corresponding position. Based on the access visibility status, target publication status, and namespace attachment status, reachability filtering is performed on the namespace identifiers and controller identifiers in the same candidate corresponding location to ensure that the access visibility status, target publication status, and namespace attachment status maintain a continuous correspondence with the namespace identifiers and controller identifiers in the same candidate corresponding location, and namespace identifiers and controller identifiers that satisfy the continuous correspondence are retained. Based on link delay, transmission retry status, port congestion status, and ANA status, bearer conflict elimination processing is performed on the target port identifier and path identifier in the same candidate corresponding location to ensure that the link delay, transmission retry status, port congestion status, and ANA status maintain a continuous bearer relationship with the target port identifier and path identifier in the same candidate corresponding location, and target port identifiers and path identifiers that are inconsistent with the continuous bearer relationship are excluded. Based on the host access request information, access visibility status, target publication status, and namespace attachment status, the retained namespace identifier and controller identifier are subjected to publication continuity verification processing to ensure that the namespace identifiers and controller identifiers corresponding to consecutively under the same host NQN maintain a continuous connection relationship at the corresponding position in the same host access request information. After completing reachability filtering, bearer conflict resolution, and publication continuity verification, the namespace identifier, controller identifier, target port identifier, and path identifier are sorted according to their corresponding order in the host NQN and host access request information, thus generating candidate mapping result data.

9. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The phased switching control module receives dynamic mapping baseline data and candidate mapping result data, and performs corresponding reading on the dynamic mapping baseline data and candidate mapping result data according to the host NQN, connection session identifier and host access request information, so that the namespace identifier, controller identifier, target port identifier and path identifier corresponding to the same host NQN and the same connection session identifier respectively enter the same switching corresponding position. Mapping version generation processing is performed on the dynamic mapping baseline data and candidate mapping result data according to the namespace identifier, controller identifier, target port identifier and path identifier in the same switching location, so that the namespace identifier, controller identifier, target port identifier and path identifier in the dynamic mapping baseline data and the namespace identifier, controller identifier, target port identifier and path identifier in the candidate mapping result data form a continuous corresponding mapping version relationship under the same host NQN and the same connection session identifier; According to the mapping version relationship, the namespace identifier, controller identifier, target port identifier and path identifier in the same handover corresponding position are subjected to shadow mapping loading process, so that the namespace identifier, controller identifier, target port identifier and path identifier in the candidate mapping result data enter the shadow mapping corresponding position without deviating from the continuous correspondence of the dynamic mapping baseline data, and the shadow mapping corresponding position maintains a continuous connection relationship with the current handover corresponding position. According to the host access request information and connection session identifier, the namespace identifier, controller identifier, target port identifier and path identifier entering the corresponding position of the shadow mapping are processed by phase confirmation and switching, so that the namespace identifier, controller identifier, target port identifier and path identifier in the corresponding position of the shadow mapping replace the corresponding content in the dynamic mapping baseline data in sequence, and the replaced namespace identifier, controller identifier, target port identifier and path identifier maintain a continuous correspondence with the host NQN and connection session identifier. Based on the namespace identifier, controller identifier, target port identifier, and path identifier after the completion phase confirmation switchover process, the old mapping release process is performed, so that the namespace identifier, controller identifier, target port identifier, and path identifier that were replaced in the dynamic mapping baseline data are removed from the current switchover corresponding position, and the namespace identifier, controller identifier, target port identifier, and path identifier after the replacement are arranged continuously in the corresponding order of host NQN, connection session identifier, and host access request information, generating mapping switchover execution data.

10. The NVMe device dynamic mapping system for storage area networks according to claim 1, characterized in that: The access continuous verification module receives the mapping switch execution data, connection session identifier, and host access request information. It performs corresponding reading on the mapping switch execution data according to the connection session identifier and host access request information, so that the host NQN, namespace identifier, controller identifier, target port identifier, and path identifier corresponding to the same connection session identifier and the same host access request information enter the same verification corresponding position. The access continuity verification process is performed on the mapping switch execution data according to the host NQN, namespace identifier, controller identifier, target port identifier and path identifier in the same verification corresponding position, so that the namespace identifier, controller identifier, target port identifier and path identifier corresponding to the same connection session identifier and the same host access request information maintain a continuous correspondence before and after the switch, and the content that maintains the continuous correspondence is retained in the current verification corresponding position; Based on the host access request information and connection session identifier, perform a handover result confirmation process on the namespace identifier, controller identifier, target port identifier, and path identifier stored in the current verification corresponding position, so that the namespace identifier, controller identifier, target port identifier, and path identifier continuously corresponding to the same host NQN form a continuous confirmation relationship in the corresponding position of the current host access request information; Based on the access continuity verification processing result and the switching result confirmation processing result, the namespace identifier, controller identifier, target port identifier and path identifier that have not formed a continuous confirmation relationship are subjected to failure rollback trigger processing, so that the content that has not formed a continuous confirmation relationship is removed from the current verification corresponding position, and the namespace identifier, controller identifier, target port identifier and path identifier that are continuously corresponding to the same connection session identifier and the same host access request information are re-entered into the current verification corresponding position. The execution results of the host NQN, connection session identifier, host access request information, namespace identifier, controller identifier, target port identifier, and path identifier after the completion of the handover result confirmation process or the failure rollback trigger process are merged. This process arranges the namespace identifier, controller identifier, target port identifier, and path identifier corresponding to different connection session identifiers in a continuous order according to the corresponding order of the host NQN and host access request information, generating mapping execution result data.

Images