A project operation management method and system based on multi-tenant architecture

Abstract

The present application discloses a project operation management method and system based on a multi-tenant architecture. The project operation management method based on a multi-tenant architecture includes: obtaining basic information of the tenant to be evaluated; judging whether the tenant to be evaluated has tenant data volume within a preset time period before the current time point based on the basic information of the tenant to be evaluated, and if so, obtaining a trained first data volume prediction model; inputting the tenant data volume within the preset time period before the current time point into the trained data volume prediction model, thereby obtaining predicted data volume data; generating a future isolation strategy for the tenant to be evaluated based on the predicted data volume data. Through the method of the present application, different isolation strategies can be replaced by predicting the possible tenant data volume of the tenant to be evaluated within the future preset time period, so that the isolation strategy can be more adapted to the user's own situation.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a project operation management method based on a multi-tenant architecture and a project operation management system based on a multi-tenant architecture. Background Art

[0002] TSM (Information Technology Service Management) SaaS (Software as a Service) management system is a cloud-based solution designed to help organizations effectively manage and deliver IT services. Like other SaaS systems, ITSM SaaS systems require data isolation between different tenants. Furthermore, ITSM tenants have upstream and downstream relationships, requiring document exchange and data sharing.

[0003] In ITSM SaaS applications, multi-tenant architecture is a common design pattern used to serve multiple clients or "tenants" on a single SaaS software. However, existing multi-tenant implementation methods often face problems such as insufficient data isolation and high tenant operation and maintenance costs. Summary of the Invention

[0004] The purpose of the present invention is to provide a project operation management method based on a multi-tenant architecture to solve at least one of the above technical problems.

[0005] One aspect of the present invention provides a project operation management method based on a multi-tenant architecture, the project operation management method based on a multi-tenant architecture comprising:

[0006] Obtain basic information of tenants to be assessed;

[0007] According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If so, then

[0008] Obtaining a trained first data volume prediction model;

[0009] Inputting the tenant data volume within a preset time period before the current time point into the trained data volume prediction model to obtain predicted data volume data;

[0010] Generate a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

[0011] Optionally, the project operation management method based on a multi-tenant architecture further includes:

[0012] According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If not, then

[0013] Obtaining a trained second data volume prediction model;

[0014] Inputting the basic information of the tenant to be assessed into the trained prediction model to obtain predicted data volume data;

[0015] Modify or establish a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

[0016] Optionally, the project operation management method based on a multi-tenant architecture further includes:

[0017] The constructed second data volume prediction model is trained.

[0018] Optionally, the training of the constructed second data volume prediction model includes:

[0019] Acquire a historical tenant dataset, where the historical tenant dataset includes basic tenant information of each historical tenant;

[0020] Acquire a historical data volume data set, where the historical data volume data set includes historical data volume data of each historical tenant;

[0021] Constructing bipartite graph information based on the historical tenant dataset and the historical data volume dataset;

[0022] Training the second data volume prediction model using the bipartite graph information to obtain a trained association model;

[0023] Input the basic information of the tenant to be evaluated into the trained recommendation model to obtain the predicted data volume data.

[0024] Optionally, the basic information of the tenant to be assessed includes the company name, number of employees, total profit, and a textual description of the main business.

[0025] Optionally, the first data volume prediction model includes a NN convolutional neural network and a GRU network.

[0026] Optionally, modifying or establishing a future isolation strategy for the tenant to be assessed based on the predicted data volume data includes:

[0027] Obtaining a segment database, the segment database including at least two data segments and a future isolation strategy corresponding to each data segment;

[0028] Obtain a future isolation strategy corresponding to the data segment where the predicted data volume of the broken area is located.

[0029] Optionally, the project operation management method based on a multi-tenant architecture further includes:

[0030] Send the obtained future isolation strategy to the tenants to be assessed;

[0031] Obtaining response information from tenants to be assessed;

[0032] Identify the reply information and determine whether the reply information agrees with the future isolation strategy. If so,

[0033] The isolation strategy of the tenant to be assessed within a future preset time period is determined as the acquired future isolation strategy.

[0034] Optionally, the project operation management method based on a multi-tenant architecture further includes:

[0035] Identify the reply information and determine whether the reply information agrees with the future isolation strategy. If not,

[0036] Identify the reply information and determine whether other future isolation strategies are selected in the reply information. If so,

[0037] Generating strategy price difference information based on the future isolation strategy selected by the tenant to be assessed and the future isolation strategy obtained based on the predicted data volume data, and sending the strategy price difference information to the tenant to be assessed;

[0038] Obtaining price difference answer information fed back by the tenant to be assessed based on the strategic price difference information;

[0039] Identify the price difference answer information, and if the price difference answer information includes a statement of agreement to the price difference,

[0040] The isolation strategy of the tenant to be assessed within a future preset time period is determined as the future isolation strategy selected by the tenant to be assessed.

[0041] The present application also provides a project operation management system based on a multi-tenant architecture, the project operation management system based on a multi-tenant architecture comprising:

[0042] A module for acquiring basic information of tenants to be assessed, wherein the module is used to acquire basic information of tenants to be assessed;

[0043] A judgment module, which is used to judge whether the tenant to be assessed has tenant data within a preset time period before the current time point based on the basic information of the tenant to be assessed;

[0044] A model acquisition module, configured to acquire a trained first data volume prediction model when the judgment module determines that the model is yes;

[0045] A prediction module, configured to input the tenant data volume within a preset time period before a current time point into the trained data volume prediction model, thereby obtaining predicted data volume data;

[0046] A generation module is used to generate a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

[0047] Beneficial effects:

[0048] Most tenants may use different amounts of data each month or week due to business adjustments, seasonal changes, or other reasons. Through the method of this application, different isolation strategies can be replaced by predicting the amount of tenant data that the tenant to be assessed may use within a preset time in the future, so that the isolation strategy can be more suitable for the user's own situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a flowchart of a project operation management method based on a multi-tenant architecture according to an embodiment of the present application. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the implementation of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below in conjunction with the drawings in the embodiments of this application. The described embodiments are part of the embodiments of this application, rather than all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain this application, and should not be understood as limiting this application. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The embodiments of this application are described in detail below in conjunction with the drawings.

[0051] Figure 1 This is a flowchart of a project operation management method based on a multi-tenant architecture according to an embodiment of the present application.

[0052] like Figure 1 The project operation management method based on the multi-tenant architecture shown includes:

[0053] Step 1: Obtain basic information of the tenant to be assessed;

[0054] Step 2: Determine whether the tenant to be assessed has tenant data volume within the preset time period before the current time point based on the basic information of the tenant to be assessed. If so,

[0055] Step 3: Obtain the trained first data volume prediction model;

[0056] Step 4: Inputting the tenant data volume within a preset time period before the current time point into the trained data volume prediction model to obtain predicted data volume data;

[0057] Step 5: Generate a future isolation strategy for the tenant to be assessed based on the predicted data volume.

[0058] Most tenants may use different amounts of data each month or week due to business adjustments, seasonal changes, or other reasons. Through the method of this application, different isolation strategies can be replaced by predicting the amount of tenant data that the tenant to be assessed may use within a preset time in the future, so that the isolation strategy can be more suitable for the user's own situation.

[0059] In this embodiment, if the tenant has previously cooperated, its historical data exists, and the tenant data volume within the preset time period before the current time point exists. However, if the tenant is newly added, there is no historical tenant data volume. In this case, the project operation management method based on the multi-tenant architecture of the present application further includes:

[0060] According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If not, then

[0061] Obtaining a trained second data volume prediction model;

[0062] Inputting the basic information of the tenant to be assessed into the trained prediction model to obtain predicted data volume data;

[0063] Modify or establish a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

[0064] For most tenants (i.e., enterprises) with similar turnover and main business scope, the actual difference in the amount of tenant data is not large. Therefore, through the above method, the amount of tenant data that has not been cooperated with before can be predicted, thereby achieving the purpose of this application.

[0065] In this embodiment, the project operation management method based on the multi-tenant architecture further includes:

[0066] The constructed second data volume prediction model is trained.

[0067] In this embodiment, the training of the constructed second data volume prediction model includes:

[0068] Acquire a historical tenant dataset, where the historical tenant dataset includes basic tenant information of each historical tenant;

[0069] Acquire a historical data volume data set, where the historical data volume data set includes historical data volume data of each historical tenant;

[0070] Constructing bipartite graph information based on the historical tenant dataset and the historical data volume dataset;

[0071] Training the second data volume prediction model using the bipartite graph information to obtain a trained association model;

[0072] Input the basic information of the tenant to be evaluated into the trained recommendation model to obtain the predicted data volume data.

[0073] For example, a text vector training set (company name, number of employees, total profit, and textual description of main business) is used as input. A corpus is a collection of all text data used to train the model. The Word2Vec model is trained using the corpus, and the continuous bag-of-words model is used to predict target words from the context. After completing model training, the trained Word2Vec model can be used to obtain the word vector representation of each word. For each word, the corresponding vector can be obtained by querying the model.

[0074] In this embodiment, constructing a bipartite graph includes:

[0075] Get bipartite graph node information ,in, is the user dataset, To estimate the linkage dataset, is the edge between sets;

[0076] Get two row vectors according to the bipartite graph and The similarity relationship between the two parts of the graph and the edge information

[0077] After obtaining the bipartite graph, the bipartite graph is converted into a vector structure and normalized to obtain a normalized adjacency matrix.

[0078] Construct a second data volume prediction model, import the normalized adjacency matrix into its multi-layer graph convolutional network to aggregate neighborhood information, obtain the node embeddings of the main view, perform regional negative sampling on the node embeddings of the main view, and construct a Bayesian personalized ranking loss function to calculate the regional negative sampling process;

[0079] The enhanced view is obtained by performing random singular value decomposition on the normalized adjacency matrix. The node embedding of the enhanced view is obtained through a graph convolutional network. The node embedding of the enhanced view and the node embedding of the main view are used to construct a contrastive learning loss function.

[0080] Construct a total loss function to jointly train the Bayesian personalized ranking loss function and the contrastive learning loss function, and minimize the loss of the total loss function to optimize the parameters of the prediction network model;

[0081] The basic information of the tenant to be evaluated is input into the trained second data volume prediction model, and the prediction network model outputs the corresponding prediction result.

[0082] In this embodiment, the basic information of the tenant to be assessed includes the company name, number of employees, total profit, and a textual description of the main business.

[0083] In this embodiment, the first data volume prediction model includes a CNN convolutional neural network and a GRU network.

[0084] In this embodiment, modifying or establishing the future isolation strategy of the tenant to be assessed based on the predicted data volume data includes:

[0085] Obtaining a segment database, the segment database including at least two data segments and a future isolation strategy corresponding to each data segment;

[0086] Obtain a future isolation strategy corresponding to the data segment where the predicted data volume of the broken area is located.

[0087] In this embodiment, several data volume segments can be set manually, and each data volume segment corresponds to a future isolation strategy.

[0088] In this embodiment, future isolation strategies may include the following strategies:

[0089] Strategy 1: Add a column D to the tenant information table. Include this column every time you write or query data to achieve logical data isolation. Column D is a tag column used to tag tenants.

[0090] Strategy 2: When the amount of tenant data is average, use the same database for all tenants, but store the tenant data in different tables to achieve logical data isolation;

[0091] Strategy 3: Use independent databases to achieve physical isolation of tenants;

[0092] Strategy 4: Store the data of a group of tenants during a traffic surge in the same database, isolate tenants with heavy traffic, and achieve request and storage isolation.

[0093] In this embodiment, the project operation management method based on the multi-tenant architecture further includes:

[0094] Send the obtained future isolation strategy to the tenants to be assessed;

[0095] Obtaining response information from tenants to be assessed;

[0096] Identify the reply information and determine whether the reply information agrees with the future isolation strategy. If so,

[0097] The isolation strategy of the tenant to be assessed within a future preset time period is determined as the acquired future isolation strategy.

[0098] In this way, on the one hand, tenants can be informed of the isolation strategy that will be used in the future, and on the other hand, the opinions of tenants can be obtained.

[0099] In this embodiment, the project operation management method based on the multi-tenant architecture further includes:

[0100] Identify the reply information and determine whether the reply information agrees with the future isolation strategy. If not,

[0101] Identify the reply information and determine whether other future isolation strategies are selected in the reply information. If so,

[0102] Generating strategy price difference information based on the future isolation strategy selected by the tenant to be assessed and the future isolation strategy obtained based on the predicted data volume data, and sending the strategy price difference information to the tenant to be assessed;

[0103] Obtaining price difference answer information fed back by the tenant to be assessed based on the strategic price difference information;

[0104] Identify the price difference answer information, and if the price difference answer information includes a statement of agreement to the price difference,

[0105] The isolation strategy of the tenant to be assessed within a future preset time period is determined as the future isolation strategy selected by the tenant to be assessed.

[0106] It is understandable that most tenants certainly want the best isolation strategy, but it is possible that the tenant cannot reach the corresponding data volume. In this case, it is necessary to set a certain price difference so that customers have a certain sense of responsibility. If there is not a relatively large amount of data, there is no need to deliberately choose a higher isolation strategy.

[0107] The present application also provides a project operation management system based on a multi-tenant architecture, which includes a basic information acquisition module for tenants to be assessed, a judgment module, a model acquisition module, a prediction module, and a generation module, wherein:

[0108] The basic information acquisition module of the tenant to be assessed is used to obtain the basic information of the tenant to be assessed;

[0109] The judgment module is used to judge whether the tenant to be assessed has tenant data volume within a preset time period before the current time point based on the basic information of the tenant to be assessed;

[0110] The model acquisition module is used to acquire a trained first data volume prediction model when the judgment module determines that the model is yes;

[0111] The prediction module is used to input the tenant data volume within a preset time period before the current time point into the trained data volume prediction model, thereby obtaining predicted data volume data;

[0112] The generation module is used to generate a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

[0113] It should be noted that the aforementioned explanation of the method embodiment is also applicable to the system of this embodiment and will not be repeated here.

[0114] Although the present invention has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made based on the present invention. Therefore, such modifications and improvements, which do not depart from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A project operation management method based on a multi-tenant architecture, characterized in that: The project operation management method based on the multi-tenant architecture includes: Obtain basic information of tenants to be assessed; According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If so, then Obtaining a trained first data volume prediction model; Inputting the tenant data volume within a preset time period before the current time point into the trained data volume prediction model to obtain predicted data volume data; Generating a future isolation strategy for the tenant to be assessed based on the predicted data volume data; The project operation management method based on the multi-tenant architecture further includes: According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If not, then Obtaining a trained second data volume prediction model; Inputting the basic information of the tenant to be assessed into the trained prediction model to obtain predicted data volume data; Modify or establish a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

2. The project operation management method based on multi-tenant architecture according to claim 1, characterized in that: The project operation management method based on the multi-tenant architecture further includes: The constructed second data volume prediction model is trained.

3. The project operation management method based on multi-tenant architecture according to claim 2, characterized in that: The training of the constructed second data volume prediction model includes: Acquire a historical tenant dataset, where the historical tenant dataset includes basic tenant information of each historical tenant; Acquire a historical data volume data set, where the historical data volume data set includes historical data volume data of each historical tenant; Constructing bipartite graph information based on the historical tenant dataset and the historical data volume dataset; Training the second data volume prediction model using the bipartite graph information to obtain a trained association model; Input the basic information of the tenant to be evaluated into the trained recommendation model to obtain the predicted data volume data.

4. The project operation management method based on a multi-tenant architecture according to claim 3, characterized in that: The basic information of the tenant to be assessed includes the company name, number of employees, total profit, and a textual description of the main business.

5. The project operation management method based on multi-tenant architecture according to claim 4, characterized in that: The first data volume prediction model includes a CNN convolutional neural network and a GRU network.

6. A project operation management system based on a multi-tenant architecture, characterized in that: The project operation management system based on the multi-tenant architecture includes: A module for acquiring basic information of tenants to be assessed, wherein the module is used to acquire basic information of tenants to be assessed; A judgment module, which is used to judge whether the tenant to be assessed has tenant data within a preset time period before the current time point based on the basic information of the tenant to be assessed; A model acquisition module, configured to acquire a trained first data volume prediction model when the judgment module determines that the model is yes; A prediction module, configured to input the tenant data volume within a preset time period before a current time point into the trained data volume prediction model, thereby obtaining predicted data volume data; A generation module, configured to generate a future isolation strategy for the tenant to be assessed based on the predicted data volume data; According to the basic information of the tenant to be assessed, it is determined whether the tenant to be assessed has tenant data in the preset time period before the current time point. If not, then Obtaining a trained second data volume prediction model; Inputting the basic information of the tenant to be assessed into the trained prediction model to obtain predicted data volume data; Modify or establish a future isolation strategy for the tenant to be assessed based on the predicted data volume data.

Images