CN113190518B - Consistency maintenance method in collaborative editing based on mobile cloud computing - Google Patents

Abstract

The invention discloses a method for maintaining consistency in collaborative editing based on mobile cloud computing. integrated operations. The invention makes full use of the computing and storage advantages of the cloud, effectively breaks through the resource limitation of the mobile terminal, allows the mobile terminal to perform lightweight collaborative editing operations, and the cloud performs heavyweight collaborative editing operations, thereby providing mobile users with better The interactivity and responsiveness of mobile co-editing maintains the consistency of shared document copies for mobile collaborative editing.

Description

Consistency maintenance method in collaborative editing based on mobile cloud computing

technical field

The invention relates to the technical field oriented to cloud computing and mobile computing, in particular to a method for maintaining consistency in collaborative editing based on mobile cloud computing.

Background technique

With the rapid development of mobile computing and collaborative computing, more and more mobile collaborative applications have been successfully applied to various fields of production and life, such as mobile collaborative office, mobile medical care, and mobile commerce. As a typical representative of mobile collaborative office, mobile collaborative editing allows geographically dispersed mobile users to collaborate with each other through the mobile Internet to complete the editing or design of shared objects, which is conducive to the sharing of group wisdom and improves collaborative work efficiency. In recent years, mobile collaborative editing has gradually become a hot issue in the CSCW field.

As the scale of mobile users continues to increase, mobile terminals face resource constraints such as computing, storage, and battery capacity. Therefore, introducing cloud computing into mobile terminals can break through the resource limitations of mobile terminals and bring better user experience to mobile collaborative users. In order to provide mobile users with good interactivity and responsiveness, both the cloud and the mobile terminal use a fully replicated data structure to store copies of shared objects, which brings great challenges to the consistency maintenance of shared objects.

Typical consistency maintenance methods are OT (Operation Transformation), AST (Address SpaceTransformation) and CRDT (CRDT, Commutative Replicated Data Type). Relevant studies have shown that, compared with the traditional consistency maintenance method, the CRDT method can ensure high responsiveness and high scalability under higher network latency, and is suitable for large-scale dynamic collaborative editing. In the collaborative editing research of mobile cloud computing, some research achievements have been made on mobile collaboration based on OT and AST technology, such as Google Docs, OptiCloud and Hermes. However, the research on CRDT-based consistency maintenance methods mostly focuses on traditional desktop collaborative applications. There are still many technical challenges in extending and applying CRDT technology to collaborative editing under the mobile cloud. On the one hand, in the mobile collaborative work environment, the mobility of users brings unstable networks, which leads to the accumulation of a large number of local operation sequences for mobile users, and the traditional CRDT technology cannot directly handle a large number of offline operation sequences; on the other hand, , there is a big gap between the computing and storage resources of the mobile terminal and the desktop PC, and the traditional CRDT technology does not take into account the factors of the mobile terminal. Therefore, applying CRDT technology to collaborative editing under the mobile cloud requires further improvement and expansion in the collaborative editing framework, support for offline operations, mobile terminal resources, and latency.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for maintaining consistency in collaborative editing based on mobile cloud computing. The present invention breaks through the resource limitation of mobile terminals, stores mobile terminal copies (cloud clones) in the cloud, and designs a resource allocation strategy, Let the cloud perform heavyweight collaborative editing operations, and mobile terminals perform lightweight collaborative editing operations, improve the responsiveness of collaborative editing on the mobile terminal, and further maintain the consistency of cloud clones and document copies shared by all mobile terminals.

To achieve this purpose, a method for maintaining consistency in collaborative editing based on mobile cloud computing designed by the present invention is characterized in that it comprises the following steps:

Step 1: Clear the operation log locallog', remotelog', remotelog", log', c2mremotelog of the cloud clone, and clear the operation log locallog, remotelog and log of the mobile terminal; initialize the synchronization semaphore c of the cloud clone to 0, and set the The synchronization semaphore m of the mobile terminal is initialized to 1;

The operation log locallog' is used to store the local operations received from the mobile terminal; the operation log remotelog' is used to store the remote operations sent by other clones in the cloud; the operation log remotelog' is used to store the collection of remote operations performed by the cloud clone; operations The log log' is used to store the set of operations and remote operations that have been performed by the cloud clone; the operation log c2mremotelog is used to store the remote operations generated by the cloud that can be directly performed by the mobile terminal;

The operation log locallog is used to store the local operations generated by the mobile terminal; the operation log remotelog is used to store the directly executable remote operations from the cloud clone transfer; the operation log log is used to store all the local operations and remote operations performed by the mobile terminal;

Step 2: Mobile terminal integration operation. First, generate a local operation o, add the local operation o to the operation log locallog, call the function IntegrateLocal(o) to integrate the local operation, and add the executed local operation o to the operation log log ; Then, set the synchronization semaphore m to 0, that is, synchronize the operation with the cloud clone, send the operation log locallog to the cloud clone, receive the directly executable remote operations sent by the cloud clone, and store them in the operation log remotelog; finally execute Directly executable remote operations in the operation log remotelog;

Step 3: Cloud clone integration operation. First, the cloud clone receives the local operation set sent from the mobile terminal and stores it in the operation log locallog'. Next, the function LocalExecution(o) is called to execute the local operation in the operation log locallog', and call The function Update(o) updates the remote operations performed by the cloud clone, generates remote operations that can be directly performed by the mobile terminal, stores it in the remote operation set c2mremotelog, and then sends the remote operation set c2mremotelog to the mobile terminal; finally, set the synchronization semaphore c as 1. Receive and integrate the remote operation set sent from other cloud clones, call the function LocalExecution(o) to execute the remote operation in the operation log remotelog', and add the executed remote operation to the operation log remotelog".

The invention makes full use of the computing and storage advantages of the cloud, effectively breaks through the resource limitation of the mobile terminal, allows the mobile terminal to perform lightweight collaborative editing operations, and the cloud performs heavyweight collaborative editing operations, thereby providing mobile users with better The interactivity and responsiveness of mobile co-editing maintains the consistency of shared document copies for mobile collaborative editing.

Description of drawings

1 is a schematic diagram of the collaborative editing architecture of the cloud and a mobile terminal according to the present invention;

In Figure ₁ , all cloud clones _C1 ', _C2 ', C3'... _Cn ' maintain the same data structure DS', DS' is the data structure DS of the shared document stored by the mobile terminal ( data structure), DS includes a hash table and a singly linked list;

Fig. 2 is the DS structure diagram that each mobile terminal maintains;

Figure 3 is a task assignment model for collaborative editing between the cloud and the mobile terminal;

FIG. 4 is a schematic diagram of the transformation and integration functions of the cloud clone local operation and the generation of directly executable remote operation functions.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings and specific embodiments:

As shown in Figure 1, a method for maintaining consistency in collaborative editing based on mobile cloud computing, the method includes a collaborative editing architecture between mobile terminals and cloud, cloud clone and mobile terminal full replication data structure, cloud clone and mobile terminal synchronization strategy, wherein the collaborative editing architecture of the mobile terminal and the cloud establishes a centralized collaborative editing architecture for the cloud clone and the mobile terminal, and the fully replicated data structure of the cloud clone and the mobile terminal is used to store each cloud clone, mobile terminal The shared document copy of the terminal and the generated operation log set, the synchronization strategy of cloud clone and mobile terminal includes operation incremental synchronization method and cloud clone, the consistency maintenance method of mobile terminal, and the consistency in collaborative editing based on mobile cloud computing The maintenance method includes the following steps:

Step 1: Initialization of cloud clone and mobile terminal full replication data structure, the cloud clone (cloud server stores multiple mobile terminal copies, each cloud clone stores the same data structure as the mobile terminal, which is equivalent to the mobile terminal data structure) copy) operation log locallog', remotelog', remotelog", log', c2mremotelog, clear the operation log locallog, remotelog and log of the mobile terminal; initialize the synchronization semaphore c of the cloud clone to 0, and synchronize the mobile terminal The semaphore m is initialized to 1;

The operation log locallog' is used to store the local operations (local operations, that is, the user generates operations on the mobile phone) received from the mobile terminal (the mobile terminal is a mobile phone or a pad, etc., the user can generate operations on the mobile phone, such as inserting a string in a document). operation, including insert operation and delete operation in the document); the operation log remotelog' is used to store the remote operations sent by other clones in the cloud (the remote operation refers to the local operation sent by a certain mobile terminal from other mobile terminals, That is, the operations generated by other mobile terminals); the operation log "remotelog" is used to store the remote operation set that the cloud clone has performed; the operation log' is used to store the operation and the remote operation set that the cloud clone has performed (there is a server in the cloud). , pass the operation to the cloud clone corresponding to the mobile terminal, and let the cloud clone execute); the operation log c2mremotelog is used to store the remote operations generated by the cloud that can be directly executed by the mobile terminal;

The operation log locallog is used to store the local operations generated by the mobile terminal; the operation log remotelog is used to store the directly executable remote operations from the cloud clone transfer; the operation log log is used to store all the local operations and remote operations performed by the mobile terminal;

Step 2: Mobile terminal integration operation. First, generate a local operation o, add the local operation o to the operation log locallog, call the function IntegrateLocal(o) to integrate the local operation, and add the executed local operation o to the operation log log ; Then, set the synchronization semaphore m to 0, that is, synchronize the operation with the cloud clone, send the operation log locallog to the cloud clone, receive the directly executable remote operations sent by the cloud clone, and store them in the operation log remotelog; finally execute Directly executable remote operations in the operation log remotelog;

Step 3: Cloud clone integration operation. First, the cloud clone receives the local operation set sent from the mobile terminal and stores it in the operation log locallog'. Next, calls the function LocalExecution(o) (local execution function, which is used for cloud clone execution. Local operation and remote operation call) execute the local operation in the operation log locallog', and call the function Update(o) to update the remote operation performed by the cloud clone, generate the remote operation that the mobile terminal can directly perform, and store it in the remote operation collection c2mremotelog, Then send the remote operation set c2mremotelog to the mobile terminal; finally, set the synchronization semaphore c to 1, receive and integrate the remote operation set sent from other cloud clones, and call the function LocalExecution(o) to execute the remote operation in the operation log remotelog' , and add the executed remote operation to the operation log remotelog".

In the above technical solution, each cloud clone maintains the same data structure DS' (including operations such as insertion and deletion, traversal, etc. of the nodes in the hash table and the singly linked list in the data structure), and the data structure DS' includes: A hash table and a singly linked list, the hash table of the data structure DS' is used to store the operation object that has performed local operations and remote operations (the operation object can be understood as node, and the operation object of the insertion operation is a node in the linked list. , which stores information such as inserted strings), the singly linked list of the data structure DS' is used to link all the executed operation objects in total order.

Each cloud clone maintains a synchronization semaphore c. When the value of the synchronization semaphore c is 1, the cloud clone and other clones perform remote operation synchronization; when the value of the synchronization semaphore c is 0, the cloud clone and the mobile terminal Synchronization of local operations and directly executable remote operations.

Each mobile terminal maintains the same data structure DS. The data structure DS includes a hash table and a singly linked list. The hash table of the data structure DS is used to store operation objects that have performed local operations and remote operations. The data structure DS The singly linked list is used to link all executed operation objects in total order.

Each mobile terminal maintains a synchronization semaphore m. When the value of the synchronization semaphore m is 1, the mobile terminal generates and integrates local operations; when the value of the synchronization semaphore m is 0, the mobile terminal is cloned with the cloud, and the local Actions are passed to the cloud clone and receive remote actions passed by the cloud clone that can be executed directly.

In the above technical solution, the specific method of calling the function LocalExecution(o) to perform the local operation in the operation log locallog' is: when the local operation o is an insertion operation, the target operation object of the local operation o is found in the singly linked list through a hash function Tar_node, due to the concurrent insertion operation, multiple new objects may be inserted after the target position of the target operation object Tar_node (mobile terminal 1 and mobile terminal 2 generate insert operations at the same time, and it is possible that they insert characters in the same position of the document, which is a conflict. operations. need to compare each other's identifier IDs to determine the final position), so the local operation o needs to do a size comparison with the identifier IDs of multiple concurrently inserted objects, when the first operation object that is smaller than the identifier ID of the local operation o is found When next_node, update the target object of local operation o to the previous object pre_node of next_node, and insert the operation object of local operation o between pre_node and next_node; when the local operation o is a deletion operation, its target operation object Tar_node The visible value of is minus 1, that is, the operation object of the local operation o is set to the tombstone, and the visible value is the visibility of the node. If the visible value is <= 0, the node is invisible, and if the visible value is > 0, the node is visible.

In the above technical solution, the specific method of calling the function LocalExecution(o) to execute the remote operation in the operation log remotelog' is: when the remote operation o is an insertion operation, the target operation object of the remote operation o is found in the singly linked list through a hash function Tar_node, since the concurrent insertion operation may insert multiple new objects after the target position of the target operation object Tar_node, the remote operation o needs to compare the size with the identifier IDs of multiple concurrently inserted objects, and when the first one is found that is smaller than the remote operation o When the operation object of the identifier ID is next_node, the target object of the remote operation o is updated to the previous object pre_node of the next_node, and the operation object of the remote operation o is inserted between pre_node and next_node; when the remote operation o is a delete operation , decrement the visible value of the target operation object Tar_node of the remote operation o by 1, that is, set the operation object of the remote operation o to the tombstone.

In the above technical solution, the specific method of calling the function Update(o) to update the remote operation performed by the cloud clone is: when a remote operation o is integrated, the operation object cur_node of the remote operation o needs to be found in the singly linked list. When the object linked behind the operation object of o is the remote_node object of the remote insertion operation, the target object of the remote_node object of the remote insertion operation needs to be set to the operation object cur_node, that is, the target identifier ID of the remote_node is updated to the identifier ID of the remote operation o .

In the above technical solution, the specific execution process of calling the function IntegrateLocal(o) to integrate the local operation is: first, through the hash table, find the target operation object Tar_node of the local operation o; if the operation type of the local operation o is an insert operation, directly in the target operation After the operation object Tar_node, insert the object New_node of the local operation o; if the operation type of the local operation o is a delete operation, directly reduce the value of the visibile of the target operation object Tar_node by 1, that is, set the Tar_node as a tombstone.

The operations o (including local operations and remote operations) involved in the above scheme and the related definitions of the node node are as follows:

Definition 1: An operation o is a six-tuple <type, l, tar_key, cur_key, con, tar_o>, (1) type is the operation type of o; (2) l is a boolean variable with a value of true or false ; (3) tar_key is the identifier of the target operation of o; (4) cur_key is the identifier of o; (5) con is the character inserted or deleted; (6) tar_o is the target operation to be undone.

Definition 2: A node is a five-tuple <cur_key, visible, con, next, link>, (1) cur_key is the identifier of the node; (2) visible is an integer variable used to represent the visibility of the node; (3) con is the content stored by node; (4) next is a pointer to the next node in the List; (5) link is a pointer to link Hash Table (Hash Table) and List (Single Linked List) The corresponding node in .

Definition 3: Identifier (ID, identifier)

The identifier of operation o consists of three elements, namely <s, site, ssv>, (1) s is the unique collaborative session number; (2) site is the unique mobile site number; (3) ssv is the state of the operation vector sum.

Definition 4: Given any two operations o _i and o _j , their IDs are denoted as ID(o _i ) and ID(o _j ), if ID(o _i )<ID(o _j ), if and only if (1) ID(o _i )[s]<ID(o _j )[s] (compare the size of the collaborative session number s in the IDs of two operations o _i and o _j ), or (2) ID(o _i ) [s]=ID(o _j )[s], ID(o _i )[ssv]<ID(o _j )[ssv] (compare the size of state vector and ssv in ID of two operations o _i and o _j ) ; or (3) ID(o _i )[s]=ID(o _j )[s],ID(o _i )[ssv]=ID(o _j )[ssv],ID(o _i )[site]< ID(o _j )[site] (compare the size of the site number site in the IDs of the two operations o _i and o _j ).

The framework of collaborative editing between cloud and mobile terminals is shown in Figure 1. The cloud stores copies of each mobile terminal. For example, cloud copy _C1 ' is a clone of mobile terminal _C1 , cloud copy _C2 ' is a clone of mobile terminal _C2 , and cloud copy _Cn ' is a clone of mobile terminal _Cn . A plurality of geographically dispersed mobile terminals participate in the same collaborative session through the mobile Internet. When the synchronization semaphore m of the mobile terminal is 1, the mobile terminal generates and integrates local operations; when the synchronization semaphore m of the mobile terminal is 0, the local operation is sent to the cloud clone, and the directly executable remote operation transmitted by the cloud clone is received . When the synchronous semaphore c of the cloud clone is 1, it receives and integrates remote operations from other clones; when the synchronous semaphore of the cloud clone is 0, it receives and integrates the local operation transmitted by the mobile terminal, and generates a remote operation that can be directly executed. to the mobile terminal. The mobile terminal directly executes the directly executable remote operation after receiving it.

The DS data structure is described in detail with reference to FIG. 2 . When multiple mobile users start collaborative editing, DSs of all mobile terminals are initialized, HT (hash table) and List (single linked list) are initialized to be empty, and the single linked list Lview of the view layer is also empty. When a mobile user generates a local insert operation, insert the newly inserted node into HT and List. For example, a mobile user can insert four nodes node1, node2, node3, node4 in a row. When a mobile user deletes, the corresponding node is set as a tombstone. For example, delete the already inserted node node4. Lview links all visible nodes in the List in total order. For example, all visible nodes node1, node2, node3, node5, node6 in the total order link List.

With reference to FIG. 3 , the task allocation model for collaborative editing between the cloud and the mobile terminal is further described in detail: the mobile terminal performs a lightweight collaborative editing task. First, initialize the synchronization semaphore m=1, generate and execute local operations; then update the synchronization semaphore m=0, pass the local operation to be synchronized to the cloud clone; then, receive and execute the directly executable sent by the cloud clone The remote operation is performed, and the value of the synchronization signal m is updated to 1, that is, the mobile terminal can continue to generate and perform the remote operation. Cloud clone performs heavyweight editing tasks. First, initialize the synchronization semaphore c=0 of the cloud clone, receive and integrate the local operations transmitted by the mobile terminal; then, update the synchronization semaphore c=1, and propagate the received local operations to the cloud clone; next, convert and integrate the remote operation from the cloud clone; finally, update the synchronization signal c=0, generate a remote operation that can be directly executed by the mobile terminal and send it to the mobile terminal.

Figure 4 shows the basic idea of the transformation and integration function (LocalExecution(o _fl )) of cloud clone local operation and the generation of directly executable remote operation function (Update(o _frj )). Suppose the initial state of the DS is "aa-bb-cc-dd-ee", and "aa-bb-cc-dd-ee" is the existing string of the document DS. The remote operation o _fri has inserted "dd" after "cc", and the remote operation o _frj has inserted "ee" after "dd". Suppose the local operation to be performed o _fli inserts "ff" after "cc". Since there may be texts "dd" and "ee" inserted by concurrent operations after "cc", call LocalExecution(o _fli ) to generate directly executable local operations of o _fli '. After ID comparison, o _fli 'inserts "ff" between "dd" and "ee". Therefore, the target ID of o _fli ' is updated from ID _cc to ID _dd . Since the execution of o _fli ' affects the execution effect of o _frj , call Update(o _frj ) to generate a directly executable remote operation o _frj '. The target ID of o _frj ' is updated from ID _dd to ID _ff , and finally o _frj in remotelog" is updated to o _frj '.

The invention breaks through the resource limitation of the mobile terminal, allows the cloud to perform heavyweight collaborative editing operations, and the mobile terminal performs lightweight collaborative editing operations, improves the responsiveness of collaborative editing on the mobile terminal, and further maintains cloud clones and all mobile terminals share the same copy of documents sex.

The content not described in detail in this specification belongs to the prior art known to those skilled in the art.

Claims (9)

1. a method for maintaining consistency in the collaborative editing based on mobile cloud computing, is characterized in that, it comprises the steps: Step 1: Clear the operation log locallog', remotelog', remotelog", log', c2mremotelog of the cloud clone, and clear the operation log locallog, remotelog and log of the mobile terminal; initialize the synchronization semaphore c of the cloud clone to 0, and set the The synchronization semaphore m of the mobile terminal is initialized to 1; The operation log locallog' is used to store the local operations received from the mobile terminal; the operation log remotelog' is used to store the remote operations sent by other clones in the cloud; the operation log remotelog' is used to store the collection of remote operations performed by the cloud clone; operations The log log' is used to store the set of operations and remote operations that have been performed by the cloud clone; the operation log c2mremotelog is used to store the remote operations generated by the cloud that can be directly performed by the mobile terminal; The operation log locallog is used to store the local operations generated by the mobile terminal; the operation log remotelog is used to store the directly executable remote operations transmitted from the cloud clone; the operation log log is used to store all the local operations and remote operations performed by the cloud clone; Step 2: Mobile terminal integration operation. First, generate a local operation o, add the local operation o to the operation log locallog, call the function IntegrateLocal(o) to integrate the local operation, and add the executed local operation o to the operation log log ; Then, set the synchronization semaphore m to 0, that is, synchronize the operation with the cloud clone, send the operation log locallog to the cloud clone, receive the directly executable remote operations sent by the cloud clone, and store them in the operation log remotelog; finally execute Directly executable remote operations in the operation log remotelog; Step 3: Cloud clone integration operation. First, the cloud clone receives the local operation set sent from the mobile terminal and stores it in the operation log locallog'. Next, the function LocalExecution(o) is called to execute the local operation in the operation log locallog', and call The function Update(o) updates the remote operations performed by the cloud clone, generates remote operations that can be directly performed by the mobile terminal, stores it in the remote operation set c2mremotelog, and then sends the remote operation set c2mremotelog to the mobile terminal; finally, set the synchronization semaphore c as 1. Receive and integrate the remote operation set sent from other cloud clones, call the function LocalExecution(o) to execute the remote operation in the operation log remotelog', and add the executed remote operation to the operation log remotelog". 2. The method for maintaining consistency in collaborative editing based on mobile cloud computing according to claim 1, wherein each cloud clone maintains the same data structure DS', and the data structure DS' includes a hash table And a singly linked list, the hash table of the data structure DS' is used to store the operation objects that have performed local operations and remote operations, and the singly linked list of the data structure DS' is used to link all the performed operation objects in total order. 3. The method for maintaining consistency in collaborative editing based on mobile cloud computing according to claim 1, wherein each cloud clone maintains a synchronization semaphore c, and when the value of the synchronization semaphore c is 1, the cloud The clone and other clones perform synchronization of remote operations; when the value of the synchronization semaphore c is 0, the cloud clone and the mobile terminal perform synchronization of local operations and directly executable remote operations. 4. the method for maintaining consistency in the collaborative editing based on mobile cloud computing according to claim 1, is characterized in that: each mobile terminal maintains identical data structure DS, and data structure DS comprises a hash table and a singly linked list , the hash table of the data structure DS is used to store the operation objects that have performed local operations and remote operations, and the singly linked list of the data structure DS is used to link all the performed operation objects in total order. 5. The method for maintaining consistency in collaborative editing based on mobile cloud computing according to claim 1, characterized in that: each mobile terminal maintains a synchronizing semaphore m, and when the value of the synchronizing semaphore m is 1, the mobile terminal Generate and integrate local operations; when the value of the synchronization semaphore m is 0, the mobile terminal synchronizes with the cloud clone, transmits local operations to the cloud clone, and receives remote operations that can be directly performed by the cloud clone. 6. the method for maintaining consistency in the collaborative editing based on mobile cloud computing according to claim 1, is characterized in that: the concrete method of calling function LocalExecution (o) to execute the local operation in operation log locallog ' is: when local operation o For the insertion operation, the target operation object Tar_node of the local operation o is found in the singly linked list through the hash function. Since the concurrent insertion operation may insert multiple new objects after the target position of the target operation object Tar_node, the local operation o needs to be combined with the target operation object Tar_node. The identifier IDs of multiple concurrently inserted objects are compared in size. When the first operation object next_node smaller than the identifier ID of the local operation o is found, the target object of the local operation o is updated to the previous object pre_node of the next_node, and the The operation object of the local operation o is inserted between pre_node and next_node; when the local operation o is a deletion operation, the visible value of the target operation object Tar_node is decremented by 1. 7. the method for maintaining consistency in the collaborative editing based on mobile cloud computing according to claim 1, is characterized in that: the concrete method of calling function Update (o) to update the remote operation that cloud clone has performed is: when a remote operation o When integrated, the operation object cur_node of the remote operation o needs to be found in the singly linked list. When the object linked behind the operation object of the remote operation o is the remote_node object of the remote insertion operation, the target object of the remote_node object of the remote insertion operation needs to be set For the operation object cur_node. 8. the method for maintaining consistency in the collaborative editing based on mobile cloud computing according to claim 1, is characterized in that: the concrete execution process of calling function IntegrateLocal(o) integrated local operation is: at first by hash table, find local operation The target operation object Tar_node of o; if the operation type of the local operation o is an insert operation, insert the object New_node of the local operation o directly after the target operation object Tar_node; if the operation type of the local operation o is a delete operation, directly insert the target operation object The value of visibile of Tar_node is decremented by 1. 9. the method for maintaining consistency in the collaborative editing based on mobile cloud computing according to claim 1, is characterized in that: the concrete method of the remote operation in the call function LocalExecution(o) to execute the operation log remotelog' is: when the remote operation o For the insertion operation, the target operation object Tar_node of the remote operation o is found in the singly linked list through the hash function, and the size of the remote operation o is compared with the identifier IDs of multiple concurrent insertion objects. When the first one smaller than the remote operation o is found When the operation object of the identifier ID is next_node, update the target object of the remote operation o to the previous object pre_node of the next_node, and insert the operation object of the remote operation o between pre_node and next_node; when the remote operation o is a delete operation, Decrease the visible value of the target operation object Tar_node of the remote operation o by 1.

Images