A resource sharing method and apparatus, a storage medium, and a computer program product

Abstract

The application provides a resource sharing method and device, a storage medium and a computer program product. The method comprises the following steps: in the case that an alliance application initiated by a first service server to a second service server is received, multi-dimensional feature extraction is performed on a first service of the first service server to obtain multi-dimensional service feature data; cross fusion is performed between the multi-dimensional service feature data to obtain cross feature data; the resource matching degree corresponding to the first service server is determined according to the multi-dimensional service feature data and the cross feature data; in the case that the first service server and the second service server are determined to be successfully allied according to the resource matching degree, resource sharing is performed on a first resource instance provided by the first service server and a second resource instance provided by the second service server, so that the first service server and / or the second service server schedules the shared resource instance.

Description

Technical Field

[0001] This application relates to electronic application technology, and more particularly to a resource sharing method and apparatus, storage medium, and computer program product. Background Technology

[0002] Currently, there are generally two approaches to running games during peak hours: one is to adopt a waiting strategy, allowing users to play only when there are available resource instances; the other is to use redundant resource instances, purchasing or renting multiple instance resources. However, the first approach leads to high latency for cloud gaming access, while the second approach results in wasted hardware resources. Summary of the Invention

[0003] This application provides a resource sharing method and apparatus, a storage medium, and a computer program product.

[0004] The technical solution of this application is implemented as follows: Firstly, this application proposes a resource sharing method, the method comprising: Upon receiving an alliance request sent by the first business server in response to the alliance application initiated by the second business server, multi-dimensional feature extraction is performed on the first business of the first business server to obtain multi-dimensional business feature data. Cross-fusion of multi-dimensional business feature data yields cross-feature data. Based on the multi-dimensional business feature data and the cross-feature data, the resource matching degree corresponding to the first business server is determined; If the first business server and the second business server are successfully allied based on the resource matching degree, the first resource instance provided by the first business server and the second resource instance provided by the second business server are shared so that the first business server and / or the second business server can schedule the shared resource instance.

[0005] Optionally, the cross-fusion includes at least one of the following: second-order cross-fusion, higher-order cross-fusion.

[0006] Optionally, determining the resource matching degree corresponding to the first business server based on the multi-dimensional business feature data and the cross-feature data includes: Obtain the feature weights set by the second service server; The feature weights, the multi-dimensional business feature data, and the cross-feature data are input into the factorization machine model to obtain the resource matching degree.

[0007] Optionally, the method further includes: Receive an alliance application initiated by the second service server; the alliance application includes at least one of the following: resource information of the second service, peak and off-peak time periods and corresponding resource demand information, off-peak time periods and corresponding resource availability information, joining conditions, and service feature weights; Upon receiving the alliance joining request sent by the first business server, and determining that the first business server meets the joining conditions based on the joining conditions, multi-dimensional feature extraction is performed on the first business to obtain multi-dimensional business feature data.

[0008] Optionally, the resource sharing between the first resource instance provided by the first service server and the second resource instance provided by the second service server includes: Create a shared resource pool based on the first resource instance and the second resource instance; Set scheduling priorities for the first resource instance and the second resource instance so that the first service server and / or the second service server can perform scheduling according to the scheduling priorities; Obtain the service package body of the first service and the service package body of the second service; and cross-install the service package body of the first service and the service package body of the second service in the shared resource pool.

[0009] Optionally, the method further includes: Receive resource adjustment instructions from the first service server and / or the second service server, and adjust resources according to the corresponding resource instance in the shared resource pool based on the resource adjustment instructions.

[0010] Optionally, the multi-dimensional business feature data includes, but is not limited to, at least one of the following: static business feature data, dynamic business feature data, and user behavior feature data.

[0011] Secondly, this application proposes a resource sharing device, which includes a processor, a memory, and a communication bus; the processor implements the above-mentioned resource sharing method when executing a running program stored in the memory.

[0012] Thirdly, this application proposes a storage medium on which a computer program is stored, which, when executed by a processor, implements the above-described resource sharing method.

[0013] Fourthly, this application proposes a computer program product, including a computer program that implements the above-mentioned resource sharing method when executed by a processor.

[0014] The embodiments of this application have the following beneficial effects: When the first service server detects that more resource instances are needed to run services during peak periods, it can send an alliance request to the second service server. By performing multi-dimensional feature extraction and cross-fusion on the first service, the multi-dimensional service feature data and cross-feature data of the first service are determined, thereby accurately determining the resource matching degree of the first service server. This allows the first service server and the second service server with higher matching degree to form an alliance. The allied service servers share resource instances, enabling each allied service server to schedule shared resource instances to run services during peak periods, thereby reducing service access latency and avoiding waste of hardware resources. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a resource sharing method provided in an embodiment of this application; Figure 2 A schematic diagram illustrating an exemplary resource matching process provided in this application embodiment; Figure 3 This is a schematic diagram illustrating the alliance process and resource sharing among allies in an exemplary cloud gaming platform, provided as an embodiment of this application. Figure 4 A schematic diagram illustrating an exemplary ally resource sharing method provided in this application embodiment; Figure 5 A flowchart illustrating an exemplary resource sharing method provided in this application embodiment; Figure 6 A schematic diagram of the structure of a resource sharing device provided in this application embodiment. Figure 1 ; Figure 7 A schematic diagram of the structure of a resource sharing device provided in this application embodiment. Figure 2 .

[0016] It should be noted that the terms "first" and "second" mentioned above are only used to distinguish between different options and do not represent the degree of superiority or inferiority of the options or their priority in the implementation process. Detailed Implementation

[0017] In order to gain a more detailed understanding of the features and technical content of the embodiments of this application, the implementation of the embodiments of this application will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for reference and illustration only and are not intended to limit the embodiments of this application.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0019] In the following description, references to "some embodiments" refer to a subset of all possible embodiments. It is understood that "some embodiments" may be the same or different subsets of all possible embodiments and may be combined with each other without conflict. It should also be noted that the terms "first, second, third" used in the embodiments of this application are merely for distinguishing similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0020] This application provides a resource sharing method, such as... Figure 1 As shown, the method may include: S101. Upon receiving an alliance request sent by the first business server in response to the alliance application initiated by the second business server, perform multi-dimensional feature extraction on the first business of the first business server to obtain multi-dimensional business feature data.

[0021] The resource sharing method proposed in this application can be applied to scenarios where multiple cloud games are running on a cloud gaming platform.

[0022] In a cloud gaming scenario, the first service server can be the first cloud gaming server on the cloud gaming platform, and the second service server can be the second cloud gaming server on the same platform. Correspondingly, the first service is the first cloud game, the second service is the second cloud game, and the service feature data is cloud gaming feature data.

[0023] In this embodiment, an alliance application initiated by a second service server is received. The alliance application includes at least one of the following: resource information of the second service, peak and valley time periods and corresponding resource demand information, off-peak and valley time periods and corresponding resource availability information, joining conditions, and service feature weights. Upon receiving an alliance joining request sent by a first service server, and determining that the first service server meets the joining conditions based on the joining conditions, multi-dimensional feature extraction is performed on the first service to obtain multi-dimensional service feature data.

[0024] In one embodiment, the resource information of the second service includes, but is not limited to, at least one of the following: service type, host configuration, number of concurrent paths, geographical distribution, network bandwidth, etc. The specific information can be selected according to the actual situation, and this application embodiment does not make specific limitations.

[0025] In one embodiment, the joining conditions include, but are not limited to, at least one of the following information: the number of business servers applying to join, the business type, the geographical distribution, the network bandwidth, etc. The specific selection can be made according to the actual situation, and this application embodiment does not impose specific limitations.

[0026] In this embodiment, the second service server can initiate an alliance request when it detects that more resource instances are needed to run services during peak periods. The first service server, upon detecting that more resource instances are needed to run services during peak periods, sends an alliance request to the second service server in response to the alliance request.

[0027] In one embodiment, the second business server that initiates the alliance application can be called the alliance leader, while the first business server that sends the joining request in response to the alliance application can become an alliance member.

[0028] In one embodiment, the number of second service servers is one or more, and the specific number can be selected according to the actual situation. This application embodiment does not make specific limitations.

[0029] It should be noted that the alliance leader can set a first validity period and a second validity period. Only during the first validity period can a second business server initiate a membership request in response to an alliance application; after the first validity period ends, no more membership requests will be accepted. During the second validity period, resource matching will be performed on the resources of second business servers applying to join the alliance. At the end of the second validity period, cloud gaming providers that meet the requirements will be included as alliance members.

[0030] It should be noted that the business server collects the number of concurrent online users during the historical time period of its business operation, and determines the daily changes in the number of online users in terms of daily, weekly, and monthly dimensions. This includes, but is not limited to, peak start and end times, peak number of users, peak steepness, peak and trough start and end times, peak and trough number of users, trough steepness, normal start and end times, normal number of users, overall CPU utilization of cloud gaming instances, and peak discount thresholds. Based on the changing data, business operation resources can be analyzed. The business server determines whether resource instances are sufficient, especially during peak periods. If the business server determines that resource instances are insufficient during peak periods, it can choose to form alliances.

[0031] It should be noted that the aforementioned business server can be either the first business server or the second business server in this application embodiment. The specific choice can be made according to the actual situation, and this application embodiment does not impose any specific limitations. When choosing to form an alliance, one can choose to initiate an alliance application (as executed by the second business server) or choose to send an alliance request in response to the alliance application (as executed by the first business server). The specific alliance method to be executed can be selected according to the actual situation. For example, if there is already a relevant alliance application on the resource sharing device, the alliance method of requesting an alliance can be executed; if there is no relevant alliance application on the resource sharing device, the alliance method of applying for an alliance can be executed. This application embodiment does not impose any specific limitations.

[0032] In one embodiment, multi-dimensional features of the first service are extracted, and the resulting multi-dimensional service feature data includes, but is not limited to, at least one of the following: static dimension service feature data, dynamic dimension service feature data, and user behavior feature dimension service feature data.

[0033] In one embodiment, static-dimensional business feature data can be extracted by analyzing the packet body using binary analysis.

[0034] In one embodiment, dynamic business feature data can be extracted through sandbox runtime monitoring.

[0035] In one embodiment, business feature data of the user behavior feature dimension can be extracted through SDK tracking and stream processing technology.

[0036] For example, in cloud gaming scenarios, static business characteristic data could include triangle density, material complexity, etc. Dynamic business characteristic data could include GPU memory usage change rate, network input / output (IO) request frequency, rendering thread queue length, etc. User behavior characteristic data could include user combo complexity, standard deviation of camera turning angle, etc.

[0037] S102. Cross-integrate the multi-dimensional business feature data to obtain cross-feature data.

[0038] In this embodiment, cross-fusion includes at least one of the following: second-order cross-fusion and higher-order cross-fusion. The specific choice can be made according to the actual situation, and this embodiment does not impose any specific limitations.

[0039] In this embodiment, second-order cross-fusion can include: first-level cross-fusion, second-level cross-fusion, and third-level cross-fusion. First-level cross-fusion can correspond to the cross-fusion between static-dimensional business feature data and dynamic-dimensional business feature data; second-level cross-fusion can correspond to the cross-fusion between dynamic-dimensional business feature data and user behavior feature data; and third-level cross-fusion can correspond to the cross-fusion between static-dimensional business feature data and user behavior feature data. The specific choice can be made according to actual circumstances, and this embodiment does not impose specific limitations.

[0040] In one embodiment, the first-level cross-validation involves constructing a matrix from static and dynamic business feature data to calculate the correlation between hardware requirements and real-time load.

[0041] Optionally, in cloud gaming scenarios, the first-level cross can be (number of triangles × memory rate of change). Weight 1 + (material complexity × rendering thread queue length) Weight 2. This allows us to determine the dynamic impact of geometric complexity on video memory pressure. This cross-feature is the complexity rendering index, which is very important. The weights of its higher-order cross-features can be set according to the game characteristics and the features of the cloud gaming platform.

[0042] In one embodiment, the second-level cross-validation involves constructing a matrix from the business feature data of the dynamic dimension and the business feature data of the user behavior dimension to calculate the causal relationship between system load and user behavior.

[0043] Optionally, in cloud gaming scenarios, the second-level crossover can be (network I / O request frequency × complexity of user combos). With a weight of 3, we can determine the rate of change in network transmission caused by a player's combo. This cross-feature is the behavioral network index, which reflects the impact of a player's operating habits and characteristics on network bandwidth. It can be used to analyze the matchmaking of games under weak network conditions or those sensitive to network latency.

[0044] In one embodiment, the three-level cross-validation involves constructing a matrix from static business feature data and user behavior feature data to calculate the implicit relationship between business design and user experience.

[0045] Optionally, in cloud gaming scenarios, the three-level intersection can be (material complexity × camera turning angle). Weight 4, this cross feature is the index of game material's influence on player operation. This index can reflect the degree of influence of different game materials on player operation and turning, and can also reflect the indirect influence on the rendered screen. The similarity of user profiles can be deduced from this index.

[0046] In this embodiment of the application, a higher-order cross is a combination of the original features and the second-order cross features into a larger higher-order cross.

[0047] It should be noted that this process can lead to an explosion of feature dimensions, so higher-order crosses require dimensionality reduction and fusion.

[0048] Optionally, in cloud gaming scenarios, higher-order cross features can be found in Table 1.

[0049] Table 1

[0050] S103. Based on multi-dimensional business characteristic data and cross-characteristic data, determine the resource matching degree corresponding to the first business server.

[0051] In this embodiment, the feature weights set by the second business server are obtained; the feature weights, multi-dimensional business feature data, and cross-feature data are input into the factorization machine model to obtain the resource matching degree.

[0052] It should be noted that by using factorization machines to model and reduce the dimensionality of cross-feature data, high-dimensional sparse features can be effectively handled, and the latent vectors of cross-features can be learned.

[0053] In one embodiment, the model algorithm for the factorization machine is shown in Equation (1).

[0054] (1) in, This represents the resource matching degree, which is the output of the model. The bias term is a constant used to adjust the baseline values ​​of the model. The linear feature part consists of the feature values ​​and weights of the static dimension feature part, the feature values ​​and weights of the dynamic dimension feature part, and the feature values ​​and weights of the user behavior dimension feature part. It is the weight of the i-th feature. is the value of the i-th feature. N is the total number of features. The second-order crossover portion represents the crossover eigenvalues ​​between features, i.e., the crossover eigenvalues ​​obtained after the first-order, second-order, and third-order crossovers mentioned above. It is a vector and The inner product represents the interaction weights between feature i and feature j. and These are the values ​​of the i-th and j-th features, respectively. The inner summation ensures that each feature pair is calculated only once, avoiding duplicate calculations.

[0055] It should be noted that, and Feature weights can be set for the first business server.

[0056] It should be noted that the above factorization machine model algorithm includes second-order cross-fusion. If it is necessary to include higher-order cross-fusion, higher-order cross-fusion terms can be added to the above factorization machine model algorithm. The specific choice can be made according to the actual situation. This application embodiment does not make specific limitations.

[0057] In one alternative embodiment, the resource matching process can be exemplified by referring to... Figure 2The process includes: extracting static features (SF), dynamic features (DSF), and user features (PBF); using a three-level crossover mechanism to obtain first-level crossover (SF×DSF), second-level crossover (DSF×PBF), and third-level crossover (SF×PBF); and then performing higher-order crossover to output the final resource matching degree.

[0058] It should be noted that the model is built based on the information from the first business server's application for alliance registration. First, multi-dimensional feature extraction is performed on the business characteristics of the first business server. These features are then analyzed and labeled. Finally, a resource matching algorithm is built based on these features and labels to achieve high-precision matching.

[0059] S104. If the alliance between the first business server and the second business server is successfully determined based on the resource matching degree, the first resource instance provided by the first business server and the second resource instance provided by the second business server shall be shared so that the first business server and / or the second business server can schedule the shared resource instance.

[0060] In this embodiment, the first service server can set a minimum resource matching degree for joining the alliance. If the resource matching degree of the first service server is greater than the minimum resource matching degree, it is determined that the first service server and the second service server have successfully formed an alliance, that is, the first service server has successfully joined the alliance. At this time, the alliance contract information is signed, and the first resource instance and the second resource instance within the alliance are integrated. The integration process includes defining the shared resource pool, setting scheduling priorities, and cross-installing service packages.

[0061] In this embodiment, a shared resource pool is created based on a first resource instance and a second resource instance; a scheduling priority is set for the first resource instance and the second resource instance so that the first service server and / or the second service server can perform scheduling according to the scheduling priority; the service package body of the first service and the service package body of the second service are obtained; and the service package bodies of the first service and the second service are cross-installed in the shared resource pool.

[0062] In one embodiment, a shared resource pool is created based on the first resource instance and the second resource instance as the business server joining the alliance. According to the signed alliance contract information, matching resource information is provided. The business platform virtually selects the resource instances provided by all alliance members. The selected resource instances form the shared resource pool. The resource instances in the shared resource pool are public resources of the alliance members. When the number of online users in a certain business operation of a member reaches a set threshold, in order to avoid the experience problem of users queuing, the business server can use the resource instances in the shared resource pool.

[0063] In one embodiment, setting scheduling priorities for the first and second resource instances involves assigning priority values ​​to the cloud resource instances running services, thus classifying the resource instances within the shared resource pool. For example, the service operation level of this service server can be set to level 1, and the service operation level of other service servers can be set to level 2.

[0064] In one embodiment, the cross-installation of service packets involves resource instances within a shared resource pool. When a resource instance is initially identified within the shared resource pool, it only contains the service packets of its own service server and needs to install the service packets of other service servers within the alliance. This installation process can include packet distribution and packet installation. Packet distribution refers to the initial identification of resource instances within the shared resource pool, transmitting the service packets of this service server to other service servers within the shared resource pool, and not receiving service packets already received by the machine. Packet installation refers to the installation of other received service packets according to a certain sorting rule after receiving them. Examples of such sorting rules include the time of packet reception, ascending order of packet name, etc. Specific choices can be made based on actual circumstances, and this embodiment does not impose specific limitations.

[0065] It should be noted that the process of sharing resources between the first resource instance and the second resource instance may also only include the creation of the shared resource pool and the cross-installation of the service package. Whether or not to include the setting of scheduling priority can be selected according to the actual situation. This application embodiment does not make specific limitations.

[0066] For example, the alliance process and resource sharing among allies in cloud gaming platforms can be seen in [reference needed]. Figure 3 Cloud Game A initiates an alliance application, and Cloud Game B and Cloud Game C join the alliance. The three exchange game packages and share information 1, information 2, and information 3 in the alliance square. Information 1, information 2, and information 3 are provided by Cloud Game A, Cloud Game B, and Cloud Game C, respectively.

[0067] In this embodiment of the application, a resource adjustment instruction issued by the first service server and / or the second service server can also be received, and the corresponding resource instance in the shared resource pool can be adjusted according to the resource adjustment instruction.

[0068] In one embodiment, the dynamic adjustment of resource instances within the shared resource pool refers to the dynamic inclusion and release of resource instances in the shared resource pool based on peak and off-peak periods. Dynamic inclusion can involve adding a specific number of idle instance resources from the resource instances of the business server to the shared resource pool during off-peak periods. The specific number added is based on the shareable quantity during that business period, which is calculated based on the current number of online users for that business period, combined with the historical online user counts and their weights across three dimensions: daily, weekly, and monthly. The aforementioned weights are optional, representing the degree of influence of the weighting of the three dimensions on the prediction of the current number of online users. Dynamic release can involve releasing a specific number of idle instance resources from the shared resource pool back to the business server during peak periods.

[0069] In one embodiment, the process involves scheduling resource instances from a shared resource pool when one of the allies experiences resource shortages. Taking cloud gaming as an example, the scheduling process refers to searching the shared resource pool based on the game identifier of a specific cloud game. Prioritizing resource instances with a game run level of 1, if no such instances are found, then searching resource instances with a run level of 2, and finally sorting the retrieved instances in descending order of machine performance. The resource instances at the top of the list are then scheduled to run the corresponding cloud game. Machine performance can include factors such as CPU, memory, disk space, and network bandwidth, and appropriate weights can be set for performance comparison.

[0070] For example, see Figure 4 (Step ①) The user logs into the cloud gaming platform to play cloud game A. At this time, the game gradually enters the peak period of online users. The normal cloud gaming instance resources of cloud game A have been used up and new resource instances are needed. Since cloud game A has formed an alliance with cloud game B and cloud game C, and the alliance is in a non-peak gaming period, the cloud gaming platform will schedule the shared resource instances of other members in the shared resource pool to run cloud game A (Step ②), which solves the problem of insufficient resources for cloud game A and improves the resource utilization rate within the alliance.

[0071] In one embodiment, if a business server within the alliance (i.e., the first business server and / or the second business server in this application embodiment) violates or terminates the signed contract, causing the resources in the shared resource pool to fail to operate according to the alliance success rules set at the time of alliance formation, the alliance is terminated.

[0072] It should be noted that, to ensure the stability of the alliance, the alliance formation period can be set, and penalties can be set for early withdrawal from the alliance or breach of contract. A threshold before the alliance expires can be set to determine whether to extend the period for another cycle. If the alliance withdraws before the expiration date or breaches the contract, corresponding penalties will be imposed.

[0073] Furthermore, if the conditions for initiating an alliance application are not met within the set conditions, the alliance application is considered to have failed and cannot be joined by other business servers. Members who have already joined will also be unable to perform mutual assistance operations.

[0074] Understandably, when the first business server detects that more resource instances are needed to run services during peak periods, it can send an alliance request to the second business server. By performing multi-dimensional feature extraction and cross-fusion on the first business, the multi-dimensional business feature data and cross-feature data of the first business are determined, thereby accurately determining the resource matching degree of the first business server. This allows the first business server and the second business server with higher matching degree to form an alliance. The allied business servers share resource instances, enabling each allied business server to schedule shared resource instances to run services during peak periods, thereby reducing service access latency and avoiding waste of hardware resources.

[0075] Based on the above embodiments, an exemplary resource sharing process is proposed, such as... Figure 5 As shown, it includes: 1. The game server collects historical concurrent online game user counts, and analyzes them in groups by day, week, and month to summarize information such as peak and trough periods of online user counts.

[0076] 2. The game server predicts the number of online users and plans resources accordingly.

[0077] 3. The game server initiates an alliance request.

[0078] 4. Wait for the alliance to be formed successfully. Each time a game server applies to join the alliance and passes the verification, the conditions for a successful alliance will be triggered.

[0079] 5. If the conditions for initiating an alliance application are not met within the set conditions, the alliance application will fail and will be cancelled.

[0080] 6. Once an alliance is successfully formed, configure resource sharing information for allies.

[0081] 7. Allies can dynamically adjust the resource instances they share in the alliance's shared resource pool based on historical predictions.

[0082] 8. Dispatch resources from the alliance's shared resource pool to cope with peak periods for alliance members.

[0083] This application provides a resource sharing device. For example... Figure 6 As shown, the resource sharing device 1 includes: The extraction unit 10 is used to extract multi-dimensional features of the first business of the first business server when it receives an alliance request sent by the first business server in response to the alliance application initiated by the second business server, so as to obtain multi-dimensional business feature data. The fusion unit 11 is used to cross-fuse multi-dimensional business feature data to obtain cross-feature data. The determining unit 12 is used to determine the resource matching degree corresponding to the first business server based on the multi-dimensional business feature data and the cross-feature data. The sharing unit 13 is used to share resources between the first resource instance provided by the first business server and the second resource instance provided by the second business server when the alliance between the first business server and the second business server is successfully determined according to the resource matching degree, so that the first business server and / or the second business server can schedule the shared resource instance.

[0084] Optionally, the cross-fusion includes at least one of the following: second-order cross-fusion, higher-order cross-fusion.

[0085] Optionally, the resource sharing device further includes: an acquisition unit; The acquisition unit is used to acquire the feature weights set by the second service server; The determining unit 12 is further configured to input the feature weights, the multi-dimensional business feature data, and the cross-feature data into the factorization machine model to obtain the resource matching degree.

[0086] Optionally, the resource sharing device further includes: a receiving unit; The receiving unit is used to receive an alliance application initiated by the second service server; the alliance application includes at least one of the following: resource information of the second service, peak and off-peak time periods and corresponding resource demand information, off-peak time periods and corresponding resource availability information, joining conditions, and service feature weights; The extraction unit 10 is further configured to, upon receiving the alliance joining request sent by the first business server and determining, based on the joining conditions, that the first business server meets the joining conditions, perform multi-dimensional feature extraction on the first business to obtain multi-dimensional business feature data.

[0087] Optionally, the resource sharing device further includes: a creation unit, a setting unit, and an installation unit; The creation unit is used to create a shared resource pool based on the first resource instance and the second resource instance; The setting unit is used to set scheduling priorities for the first resource instance and the second resource instance, so that the first service server and / or the second service server can perform scheduling according to the scheduling priorities; The acquisition unit is further configured to acquire the service packet body of the first service and the service packet body of the second service; The installation unit is used to cross-install the service package of the first service and the service package of the second service in the shared resource pool.

[0088] Optionally, the receiving unit is further configured to receive resource adjustment instructions issued by the first service server and / or the second service server, and adjust resources according to the corresponding resource instance in the shared resource pool based on the resource adjustment instructions.

[0089] Optionally, the multi-dimensional business feature data includes, but is not limited to, at least one of the following: static business feature data, dynamic business feature data, and user behavior feature data.

[0090] This application provides a resource sharing device that, upon receiving an alliance request from a first service server in response to an alliance application initiated by a second service server, performs multi-dimensional feature extraction on the first service of the first service server to obtain multi-dimensional service feature data; cross-fuses the multi-dimensional service feature data to obtain cross-feature data; determines the resource matching degree corresponding to the first service server based on the multi-dimensional service feature data and the cross-feature data; and, if the alliance between the first service server and the second service server is successfully established based on the resource matching degree, shares the first resource instance provided by the first service server and the second resource instance provided by the second service server, so that the first service server and / or the second service server can schedule the shared resource instance. Therefore, the resource sharing device proposed in this embodiment allows the first service server to send an alliance request to the second service server when it detects that more resource instances are needed to run services during peak periods. By performing multi-dimensional feature extraction and cross-fusion on the first service, the multi-dimensional service feature data and cross-feature data of the first service are determined, thereby accurately determining the resource matching degree of the first service server. This allows the first service server and the second service server with higher matching degree to form an alliance. The allied service servers share resource instances, enabling each allied service server to schedule shared resource instances to run services during peak periods, thereby reducing service access latency and avoiding waste of hardware resources.

[0091] Figure 7This is a schematic diagram of the composition structure of a resource sharing device 1 provided in an embodiment of this application. In practical applications, based on the same disclosed concept of the above embodiments, such as... Figure 7 As shown, the resource sharing device 1 in this embodiment includes: a processor 14, a memory 15, and a communication bus 16.

[0092] The processor 14 described above can be at least one of the following: Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), CPU, controller, microcontroller, and microprocessor. It is understood that, for different devices, the electronic device used to implement the above processor function can also be other types, and this embodiment does not impose specific limitations.

[0093] In this embodiment, the communication bus 16 is used to realize the connection communication between the processor 14 and the memory 15; when the processor 14 executes the running program stored in the memory 15, it implements the following resource sharing method: Upon receiving an alliance request from the first business server in response to an alliance application initiated by the second business server, the system performs multi-dimensional feature extraction on the first business of the first business server to obtain multi-dimensional business feature data; it then cross-fuses the multi-dimensional business feature data to obtain cross-feature data; based on the multi-dimensional business feature data and the cross-feature data, it determines the resource matching degree corresponding to the first business server; and if the alliance between the first business server and the second business server is successfully established based on the resource matching degree, it shares resources between the first resource instance provided by the first business server and the second resource instance provided by the second business server, so that the first business server and / or the second business server can schedule the shared resource instance.

[0094] Furthermore, the cross-fusion includes at least one of the following: second-order cross-fusion and higher-order cross-fusion.

[0095] Furthermore, the processor 14 is also used to obtain the feature weights set by the second business server; and input the feature weights, the multi-dimensional business feature data and the cross feature data into the factorization machine model to obtain the resource matching degree.

[0096] Furthermore, the processor 14 is also configured to receive an alliance application initiated by the second service server; the alliance application includes at least one of the following: resource information of the second service, peak and off-peak periods and corresponding resource demand information, off-peak periods and corresponding resource availability information, joining conditions, and service feature weights; upon receiving the alliance joining request sent by the first service server, and determining that the first service server meets the joining conditions based on the joining conditions, multi-dimensional feature extraction is performed on the first service to obtain multi-dimensional service feature data.

[0097] Furthermore, the processor 14 is also configured to create a shared resource pool based on the first resource instance and the second resource instance; set scheduling priorities for the first resource instance and the second resource instance so that the first service server and / or the second service server can perform scheduling according to the scheduling priorities; obtain the service package body of the first service and the service package body of the second service; and cross-install the service package body of the first service and the service package body of the second service in the shared resource pool.

[0098] Furthermore, the processor 14 is also configured to receive resource adjustment instructions issued by the first service server and / or the second service server, and adjust resources according to the resource adjustment instructions for the corresponding resource instance in the shared resource pool.

[0099] Furthermore, the multi-dimensional business feature data includes, but is not limited to, at least one of the following: static dimension business feature data, dynamic dimension business feature data, and user behavior feature dimension business feature data.

[0100] This application provides a storage medium storing a computer program thereon. The computer-readable storage medium stores one or more programs, which can be executed by one or more processors and applied in a resource sharing device. The computer program implements the resource sharing method described above.

[0101] Based on the above embodiments, this application provides a computer program product, including a computer program that can be executed by one or more processors, and the computer program implements the resource sharing method described above.

[0102] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause an image display device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this disclosure.

[0104] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application.

Claims

1. A resource sharing method, characterized in that, The method includes: Upon receiving an alliance request sent by the first business server in response to the alliance application initiated by the second business server, multi-dimensional feature extraction is performed on the first business of the first business server to obtain multi-dimensional business feature data. Cross-fusion of multi-dimensional business feature data yields cross-feature data. Based on the multi-dimensional business feature data and the cross-feature data, the resource matching degree corresponding to the first business server is determined; If the first business server and the second business server are successfully allied based on the resource matching degree, the first resource instance provided by the first business server and the second resource instance provided by the second business server are shared so that the first business server and / or the second business server can schedule the shared resource instance.

2. The method according to claim 1, characterized in that, The cross-fusion includes at least one of the following: second-order cross-fusion and higher-order cross-fusion.

3. The method according to claim 1, characterized in that, The step of determining the resource matching degree corresponding to the first business server based on the multi-dimensional business feature data and the cross-feature data includes: Obtain the feature weights set by the second service server; The feature weights, the multi-dimensional business feature data, and the cross-feature data are input into the factorization machine model to obtain the resource matching degree.

4. The method according to claim 1, characterized in that, The method further includes: Receive an alliance application initiated by the second service server; the alliance application includes at least one of the following: resource information of the second service, peak and off-peak time periods and corresponding resource demand information, off-peak time periods and corresponding resource availability information, joining conditions, and service feature weights; Upon receiving the alliance joining request sent by the first business server, and determining that the first business server meets the joining conditions based on the joining conditions, multi-dimensional feature extraction is performed on the first business to obtain multi-dimensional business feature data.

5. The method according to claim 1, characterized in that, The resource sharing between the first resource instance provided by the first service server and the second resource instance provided by the second service server includes: Create a shared resource pool based on the first resource instance and the second resource instance; Set scheduling priorities for the first resource instance and the second resource instance so that the first service server and / or the second service server can perform scheduling according to the scheduling priorities; Obtain the service package body of the first service and the service package body of the second service; and cross-install the service package body of the first service and the service package body of the second service in the shared resource pool.

6. The method according to claim 5, characterized in that, The method further includes: Receive resource adjustment instructions from the first service server and / or the second service server, and adjust resources according to the corresponding resource instance in the shared resource pool based on the resource adjustment instructions.

7. The method according to claim 1, characterized in that, The multi-dimensional business feature data includes, but is not limited to, at least one of the following: static business feature data, dynamic business feature data, and user behavior feature data.

8. A resource sharing device, characterized in that, The resource sharing device includes: a processor, a memory, and a communication bus; when the processor executes the running program stored in the memory, it implements the method as described in any one of claims 1-7.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images