Heterogeneous resource scheduling method and apparatus, MEC device, and storage medium

Abstract

The application provides a heterogeneous resource scheduling method and device, a MEC equipment and a storage medium, wherein the method comprises the following steps: dividing a heterogeneous resource pool according to a heterogeneous resource type, the heterogeneous resource type comprising an information technology (IT) resource and a communication technology (CT) resource, the heterogeneous resource pool comprising a first resource pool for providing a basic resource to service data with an IT resource requirement and a second resource pool for providing a basic resource to service data with a CT resource requirement; monitoring a first resource occupation condition of the first resource pool and a second resource occupation condition of the second resource pool in real time; and adjusting the basic resource in the first resource pool and the second resource pool according to the first resource occupation condition and the second resource occupation condition. Through the above method, the rational allocation of the IT resource and the CT resource of the MEC equipment can be realized, and the utilization rate of the resource is effectively improved.

Description

Technical Field

[0001] This application relates to the field of edge computing technology, and in particular to a heterogeneous resource scheduling method, apparatus, MEC device and storage medium. Background Technology

[0002] With the continuous development of technologies such as the Industrial Internet and artificial intelligence, various types of businesses are increasingly demanding basic resources such as computing and network resources, leading to the emergence of Mobile Edge Computing (MEC).

[0003] In a MEC (Multi-access Edge Computing) system, communication equipment such as User Plane Functions (UPF) and Broadband Remote Access Servers (BRAS) are deployed, along with business processing systems and intelligent applications such as business applications (APPs). Therefore, MEC equipment integrates both Communication Technology (CT) and Information Technology (IT) resources. In a cloud-based environment, they share basic resources of various types.

[0004] However, IT resources and CT resources belong to heterogeneous businesses, and their business logic and operating mechanisms are different. However, MEC equipment does not have a coordination mechanism or related functions for heterogeneous resources, which leads to the inefficient use of IT resources and CT resources, thereby affecting the network service capabilities and computing capabilities of MEC equipment. Summary of the Invention

[0005] This application provides a heterogeneous resource scheduling method, apparatus, MEC device, and storage medium to solve the problem of low resource utilization caused by the unreasonable allocation of IT and CT resources.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] This application provides a heterogeneous resource scheduling method applied to mobile edge computing (MEC) devices, the method comprising:

[0008] Heterogeneous resource pools are divided according to heterogeneous resource types, including information technology (IT) resources and communication technology (CT) resources. The heterogeneous resource pools include a first resource pool for providing basic resources to business data with IT resource requirements and a second resource pool for providing basic resources to business data with CT resource requirements.

[0009] The system monitors the first resource occupancy of the first resource pool and the second resource occupancy of the second resource pool in real time; and adjusts the basic resources in the first resource pool and the second resource pool based on the first resource occupancy and the second resource occupancy.

[0010] In one implementation, the basic resources include at least one of the following types of resources: computing resources, storage resources, or network resources.

[0011] In one implementation, the step of dividing the heterogeneous resource pool according to the heterogeneous resource type includes:

[0012] Obtain historical resource usage information for resources corresponding to heterogeneous resource types;

[0013] The initial resource allocation ratio between the first resource pool and the second resource pool is determined based on the historical resource occupancy; and heterogeneous resource pools are divided based on the total basic resources of the MEC devices and the initial resource allocation ratio.

[0014] In one implementation, adjusting the basic resources in the first resource pool and the second resource pool based on the first resource occupancy and the second resource occupancy includes:

[0015] If the resources occupied by the first resource pool or the second resource pool reach a preset proportion of the total basic resources of the MEC device, the basic resources in the first resource pool and the second resource pool are adjusted according to the resource occupancy status of the first resource pool and the second resource pool.

[0016] In one implementation, after dividing the heterogeneous resource pools according to the heterogeneous resource type, the method further includes:

[0017] In response to a first service data packet initiated by a target service terminal, identify the resources corresponding to the heterogeneous resource types required by the first service data packet;

[0018] When the resource corresponding to the heterogeneous resource type is an IT resource, resources are allocated to the first service data packet based on the first resource pool to process the first service data packet.

[0019] When the resource corresponding to the heterogeneous resource type is a CT resource, resources are allocated to the first service data packet based on the second resource pool to process the first service data packet.

[0020] In one implementation, the basic resources include two or more types of basic resources, and after dividing the heterogeneous resource pools according to the heterogeneous resource types, it further includes:

[0021] Configure the scheduling priority of each type of basic resource in the first resource pool and the second resource pool respectively;

[0022] When there are multiple second service data packets that simultaneously require IT resources and / or CT resources, resources are allocated to the multiple second service data packets based on the first resource pool and / or the second resource pool and the scheduling priority.

[0023] In one implementation, the basic resources include computing resources, storage resources, and network resources.

[0024] The step of configuring the scheduling priorities of each type of basic resource in the first resource pool and the second resource pool respectively includes:

[0025] The scheduling priority of each type of basic resource in the first resource pool is configured as computing resources > storage resources > network resources, and the scheduling priority of each type of basic resource in the second resource pool is configured as network resources > storage resources > computing resources.

[0026] According to another aspect of this application, a heterogeneous resource scheduling device is provided, applied to MEC equipment, the device comprising:

[0027] The resource partitioning module is configured to partition heterogeneous resource pools according to heterogeneous resource types, including information technology (IT) resources and communication technology (CT) resources. The heterogeneous resource pools include a first resource pool for providing basic resources to business data with IT resource requirements and a second resource pool for providing basic resources to business data with CT resource requirements.

[0028] The monitoring module is configured to monitor the first resource occupancy of the first resource pool and the second resource occupancy of the second resource pool in real time.

[0029] The resource scheduling module is configured to adjust the basic resources in the first resource pool and the second resource pool based on the first resource occupancy status and the second resource occupancy status.

[0030] According to another aspect of this application, an MEC device is provided, comprising: a processor, and a memory communicatively connected to the processor;

[0031] The memory stores computer-executed instructions;

[0032] The processor executes computer execution instructions stored in the memory to implement the heterogeneous resource scheduling method.

[0033] According to another aspect of this application, a computer-readable storage medium is provided, wherein computer-executable instructions are stored therein, which, when executed by a processor, are used to implement the heterogeneous resource scheduling method described above.

[0034] The heterogeneous resource scheduling method, apparatus, MEC device, and storage medium provided in this application divide heterogeneous resource pools according to heterogeneous resource types, including IT resources (information technology) and CT resources (communication technology). Each heterogeneous resource pool includes a first resource pool for providing basic resources to business data requiring IT resources and a second resource pool for providing basic resources to business data requiring CT resources. The system monitors the first resource occupancy of the first resource pool and the second resource occupancy of the second resource pool in real time, and adjusts the basic resources in the first and second resource pools based on these occupancy statuses. This process, by dividing the heterogeneous resource pools and utilizing different pools for separate scheduling of heterogeneous resources, and by dynamically adjusting the resources within each pool by monitoring their occupancy, effectively achieves a reasonable allocation of IT and CT resources within the MEC device, improving resource utilization. Attached Figure Description

[0035] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0036] Figure 1 A possible network architecture diagram provided for an embodiment of this application;

[0037] Figure 2 A flowchart illustrating a heterogeneous resource scheduling method provided in an embodiment of this application;

[0038] Figure 3 A flowchart illustrating another heterogeneous resource scheduling method provided in this application embodiment;

[0039] Figure 4 A flowchart illustrating another heterogeneous resource scheduling method provided in this application embodiment;

[0040] Figure 5 This is one of the structural schematic diagrams of a heterogeneous resource scheduling device provided in an embodiment of this application;

[0041] Figure 6 This is a second schematic diagram of a heterogeneous resource scheduling device provided in an embodiment of this application;

[0042] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0043] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0044] MEC is a concept that emerged alongside 5G technology, and is sometimes referred to as fog computing.

[0045] 5G technology applications will be applied to industries such as manufacturing and healthcare, which have extremely strict requirements for network bandwidth and latency. MEC is a computing platform deployed close to the user side. If enterprise-side application systems are deployed on MEC, lower end-to-end service latency can be provided. This deployment method can reduce the transmission pressure on the backbone network.

[0046] Taking the industrial internet platform as an example, its task is to provide enterprises with computing resources. In addition, on this basis, it provides various application services around the production business process, such as product design, production, management and service.

[0047] Large enterprises may build their own cloud platform, while small and medium-sized enterprises (SMEs) will place their business on public clouds, such as Alibaba Cloud or carrier clouds. However, whether it's a self-built cloud or a public cloud, if the centralized cloud is far from the production site, it will be difficult to meet the needs of businesses that involve production processes, require rapid response, or have high real-time requirements. Therefore, such businesses need edge computing, or edge cloud.

[0048] Collaboration is needed between the cloud and the edge. For example, MEC devices can collect a large amount of field data, perform filtering on the edge cloud, and then send it to the central cloud. The central cloud uses this data as a benchmark to train and propose an optimized model, which is then sent back to the edge cloud, where it can also perform localized data processing.

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar components or components having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0050] Figure 1 This application provides a possible network architecture diagram, illustrating a 5G access network architecture and the network location of the MEC (Multi-access Edge Computing) device. A service terminal (supporting 5G functionality) wirelessly transmits information to a 5G base station. The 5G base station receives the wireless signal, converts it into a wired optical signal, and transmits it to the MEC device. The MEC device processes this signal in three ways: First, the MEC device itself processes the service signal. Second, the MEC device transmits the signal to the core network, and then through the core network equipment to the central cloud for processing. The central cloud, also known as a public cloud (e.g., Alibaba Cloud), has the advantage of strong computing power but the disadvantage of being physically far from the service side. Third, the MEC device transmits the signal to the enterprise's private cloud. Enterprise private clouds are often deployed within the enterprise's own campus and are generally self-built by the enterprise. Compared to public clouds, enterprises maintain their own private cloud infrastructure. As a relay node in the entire network, the MEC plays a crucial role, combining IT (computing) and CT (service data offloading or signaling interaction in communication networks).

[0051] Among them, the business terminals may include, but are not limited to, computers, smartphones, tablets, e-book readers, Moving Picture Experts Group Audio Layer III (MP3) players, Moving Picture Experts Group Audio Layer IV (MP4) players, portable computers, in-vehicle computers, wearable devices, desktop computers, set-top boxes, smart TVs, etc.

[0052] Currently, MEC devices lack internal mechanisms and functions for allocating and coordinating heterogeneous resources. In related technologies, MEC devices handle IT and CT services by directly scheduling resources based on their availability. However, IT and CT are heterogeneous resource types, and their business logic and operating mechanisms differ. For example, IT services require higher computing resources, while CT services require higher network resources. Unified scheduling of heterogeneous services would inevitably lead to a waste or shortage of infrastructure resources, resulting in inefficient utilization of these resources.

[0053] To address the aforementioned technical issues, this application provides a heterogeneous resource scheduling method, apparatus, MEC device, and storage medium. By dividing heterogeneous resource pools according to their types, a first resource pool provides basic resources for business data requiring IT resources, and a second resource pool provides basic resources for business data requiring CT resources. This enables separate scheduling of resources for heterogeneous business data and real-time monitoring of resource occupancy in both the first and second resource pools. Based on the resource occupancy of each pool, resources are adjusted to achieve dynamic allocation of basic resources within the heterogeneous resource pools. This effectively improves the utilization rate of basic resources, thereby enhancing the network service capabilities and computing power of the MEC device.

[0054] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0055] The network architecture diagram of this application has been briefly described above. The following section describes its application in... Figure 1 Taking a mobile edge computing (MEC) device as an example, this application will explain in detail the heterogeneous resource scheduling method provided in its embodiments.

[0056] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating a heterogeneous resource scheduling method provided in an embodiment of this application, the method including steps S201-S203.

[0057] Step S201: Divide the heterogeneous resource pools according to the heterogeneous resource types. The heterogeneous resource types include information technology (IT) resources and communication technology (CT) resources. The heterogeneous resource pools include a first resource pool for providing basic resources to business data with IT resource requirements and a second resource pool for providing basic resources to business data with CT resource requirements.

[0058] In this embodiment, the heterogeneous resource types include IT resources and CT resources, namely, resources for processing IT services and resources for processing CT services. IT services and CT services share the underlying basic resources (such as computing, storage, and image processing). This embodiment divides the heterogeneous resources into pools and uses different pools to provide basic resources for the business data that require heterogeneous resources, thereby realizing the separate scheduling of heterogeneous resources.

[0059] Specifically, the basic resources in this embodiment include at least one of the following types of resources: computing resources, storage resources, or network resources.

[0060] It is understood that the basic resources in this embodiment can be one or any combination of the above-mentioned resource types. As technology continues to develop, the basic resources may also be limited to the above-mentioned resources.

[0061] In one implementation, different types of basic resources can be virtualized using a virtualization module for resource access. Specifically, the virtualization module virtualizes each type of basic resource according to a unified interface and standard, thereby forming a resource pool within the device. This resource pool can be divided into a first resource pool and a second resource pool, and corresponding resource access is performed when business data is received.

[0062] In one implementation, in order to achieve reasonable allocation of resources in the heterogeneous resource pool and improve resource utilization, step S201 of this embodiment divides the heterogeneous resource pool according to the type of heterogeneous resources, which may specifically include the following steps:

[0063] Obtain historical resource usage information for resources corresponding to heterogeneous resource types;

[0064] The initial resource allocation ratio between the first resource pool and the second resource pool is determined based on the historical resource usage.

[0065] Heterogeneous resource pools are divided based on the total basic resources of the MEC devices and the initial resource allocation.

[0066] Specifically, historical resource usage can be the resource usage when processing business using existing technologies without heterogeneous resource pool partitioning. For example, if the resource usage ratio of MEC processing IT business and processing CT business is 1:3 in a given period, this resource usage ratio can be determined as the initial resource allocation ratio of the first resource pool and the second resource pool. In some embodiments, the initial resource allocation ratio can also be adjusted according to the actual situation.

[0067] Step S202: Monitor the first resource occupancy status of the first resource pool and the second resource occupancy status of the second resource pool in real time.

[0068] In one implementation, the remaining resource status of the first resource pool and the remaining resource status of the second resource pool can be directly obtained to monitor the resource occupancy of the first and second resource pools.

[0069] In another possible implementation, the MEC can be divided into business function modules and network function modules. The business function modules handle IT services, while the network function modules handle CT services. These modules interface with the first and second resource pools, respectively. A monitoring program is created to obtain the usage of basic resources by the business and network function modules through internal signaling interaction; that is, the first resource occupancy in the first resource pool and the second resource occupancy in the second resource pool.

[0070] Step S203: Adjust the basic resources in the first resource pool and the second resource pool according to the first resource occupancy and the second resource occupancy.

[0071] In this embodiment, although the heterogeneous resource pool is divided based on historical resource usage, which improves resource utilization to some extent, in practical applications, there may be uneven business operations, leading to significant differences in resource usage between the first and second resource pools. This could still result in resource waste or insufficient resources. Therefore, this embodiment monitors the usage of the first and second resource pools in real time and dynamically adjusts them to further improve the utilization of MEC resources.

[0072] In one implementation, in order to effectively improve resource utilization while reducing process complexity, step S203 of this embodiment adjusts the basic resources in the first resource pool and the second resource pool according to the first resource occupancy and the second resource occupancy, which may include the following steps:

[0073] If the resources occupied by the first resource pool or the second resource pool reach a preset proportion of the total basic resources of the MEC device, the basic resources in the first resource pool and the second resource pool are adjusted according to the resource occupancy status of the first resource pool and the second resource pool.

[0074] In this embodiment, the resource occupancy of the first resource pool and the second resource pool can be monitored in real time. When either of them reaches a preset ratio Y of the total resources, the basic resources in the resource pool are adjusted. It should be noted that those skilled in the art can adapt the preset ratio by combining practical applications and existing technologies, and the preset ratio of the first resource pool and the second resource pool can be the same or different. This embodiment does not make a specific setting in this regard.

[0075] In one implementation, a utilization ratio can be set. When the utilization of the first resource pool and the utilization of the second resource pool reach this ratio, the basic resources in the first resource pool and the second resource pool are adjusted. The adjustment process can be to schedule the basic resources of the resource pool with the smaller utilization ratio to the resource pool with the larger utilization ratio. In some embodiments, the basic resources in the first resource pool and the second resource pool can also be adjusted according to existing technology. This embodiment does not specifically limit the specific adjustment method.

[0076] In one implementation, a token mechanism can be introduced to schedule resources in the first and second resource pools based on a token control algorithm. For example, the number of tokens is set to N. The size of N corresponds to the total remaining amount of basic resources; assuming the total remaining amount of basic resources is quantified as 100, then N is set to 100. In subsequent processes, tokens are allocated to the first or second resource pool based on the real-time occupancy status of the first and second resources (the amount of remaining resources). Only after obtaining a token can one apply for the corresponding basic resource. Furthermore, since the types of basic resources (e.g., computing resources, storage resources, network resources) differ, the actual tokens can also be divided into N1, N2, ..., Ni, etc., depending on the resource type such as CPU, NPU, and storage.

[0077] In some embodiments, the two implementations of Y and N mentioned above can be combined. It is understood that, unlike the setting of N, Y is a total value and a unified metric; in other words, Y can be a weighted value of the utilization rates of various basic resources. Specifically, Y = y1*Y1 + y2*Y2 + ... + yi*Yi. Where Y1, Y2, ..., Yi represent the utilization rate of each basic resource, and y1, y2, ..., yi are pre-defined weights. When the token mechanism is enabled, MEC will allocate resources according to the needs of IT and / or CT services by distributing tokens.

[0078] As can be seen, this embodiment divides heterogeneous resource pools according to heterogeneous resource types, using the first resource pool to provide basic resources for business data requiring IT resources, and the second resource pool to provide basic resources for business data requiring CT resources. This enables separate scheduling of resources for heterogeneous business data, effectively improving resource utilization. Furthermore, by monitoring the resource occupancy of the first and second resource pools in real time and dynamically adjusting the resources in each pool based on their resource occupancy, flexible scheduling of heterogeneous resources can be achieved. This effectively avoids resource waste and shortages, improving the network service capabilities, computing power, and storage capacity of the MEC.

[0079] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating another heterogeneous resource scheduling method provided in the application embodiment. Based on the above embodiment, this embodiment provides a resource scheduling method when receiving service data packets, which can realize flexible resource scheduling for different service data. Specifically, in addition to the above steps S201-S203, after dividing the heterogeneous resource pool according to the heterogeneous resource type in step 201, the following steps S301-S303 are also included.

[0080] In this embodiment, steps S301-S303 are executed before steps S202 and S203. In some embodiments, steps S301-S303 may also be executed after steps S202 or S203. This embodiment does not specifically limit the execution order.

[0081] Step S301: In response to the first service data packet initiated by the target service terminal, identify the resources corresponding to the heterogeneous resource type required by the first service data packet.

[0082] In one implementation, the resources required for the first service data packet can be identified based on the service type. For example, if the service type is IT service, such as the service corresponding to network devices or functions like Broadband Access Server (BRAS) and Service Router (SR), the heterogeneous resource type required is IT resource.

[0083] Understandably, in response to, is used to indicate the conditions or states on which the operation being performed depends. When the conditions or states on which it depends are met, one or more operations being performed can be performed in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations being performed are executed.

[0084] Step S302: If the resource corresponding to the heterogeneous resource type is an IT resource, allocate resources to the first service data packet based on the first resource pool to process the first service data packet.

[0085] Step S303: If the resource corresponding to the heterogeneous resource type is a CT resource, allocate resources to the first service data packet based on the second resource pool to process the first service data packet.

[0086] Compared to related technologies, which process business packets without differentiating the resource types required by the packets and allocate resources based on predetermined resource scheduling methods, leading to either waste or insufficiency of basic resources and inefficient utilization, this embodiment, upon receiving a business data packet, first identifies the resource type required by the packet and then allocates resources to the corresponding resource type according to a pre-defined heterogeneous resource pool. This enables the rational allocation of IT and CT resources, improving MEC resource utilization.

[0087] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating another heterogeneous resource scheduling method provided in this application. Based on the above embodiments, this embodiment considers that the basic resources include multiple types of resources. To further improve the utilization rate of each type of resource, this embodiment configures the scheduling priority of each basic resource, and allocates resources according to the scheduling priority during resource allocation. Specifically, in addition to the above steps S201-S203, the basic resources include two or more types of basic resources. After dividing the heterogeneous resource pool according to the heterogeneous resource type in step S201, steps S401 and S402 are also included.

[0088] In this embodiment, steps S401 and S402 are executed before steps S202 and S203. In some embodiments, steps S301-S303 may also be executed after steps S202 or S203. This embodiment does not specifically limit the execution order.

[0089] Step S401: Configure the scheduling priority of each type of basic resource in the first resource pool and the second resource pool respectively.

[0090] In one implementation, the basic resources include computing resources, storage resources, and network resources. Step S401, configuring the scheduling priorities of each type of basic resource in the first resource pool and the second resource pool respectively, may include the following steps:

[0091] The scheduling priority of each type of basic resource in the first resource pool is configured as computing resources > storage resources > network resources, and the scheduling priority of each type of basic resource in the second resource pool is configured as network resources > storage resources > computing resources.

[0092] This embodiment takes into account that at the same time, the MEC may have different requirements for different types of basic resources when receiving service messages. For example, IT services often involve cloud computing, so the demand for CPU, NPU, and GPU is high. Therefore, this embodiment configures the scheduling priority in the first resource pool as computing resources > storage resources > network resources. That is, the MEC gives higher priority to basic resources such as CPU, NPU, and GPU computing resources for IT services, while giving lower priority to resources such as FPGA and storage. On the other hand, CT services are mainly communication services, so the demand for resources such as transmission, FPGA, and storage is high. This embodiment configures the scheduling priority in the second resource pool as network resources > storage resources > computing resources. The MEC gives high priority to resources such as FPGA and storage for CT services, while giving low priority to basic resources such as CPU, NPU, and GPU.

[0093] It should be noted that in some embodiments, the scheduling priority can be adjusted in real time according to the actual resource requirements of IT and CT services. For example, if the CPU requirement of CT services is high priority at a certain moment, the MEC can choose to change the CPU requirement of IT services to low priority based on the resource usage of IT services. This can improve the flexibility of the overall system and achieve better utilization of basic resources.

[0094] Step S402: When there are multiple second service data packets that simultaneously require IT resources and / or CT resources, allocate resources to the multiple second service data packets based on the first resource pool and / or the second resource pool and the scheduling priority.

[0095] Under the priority scheduling mechanism adopted in this embodiment, when IT and CT services simultaneously initiate requests for the use of a certain type of basic resource, the MEC prioritizes the service with the higher priority. For example, when a service function module and a network function module simultaneously initiate CPU requests, and the service function module has a higher priority for CPU, the resource adjustment module will prioritize allocating tokens to the service function module to meet its access to and use of basic resources, and then meet the network function module's request. Under this mechanism, when basic resources are scarce in the short term, the needs of important or urgent function modules for accessing basic resources are prioritized.

[0096] Please refer to Figure 5 , Figure 5 This is one of the structural schematic diagrams of a heterogeneous resource scheduling device provided in an embodiment of this application, applied to an MEC device. The device includes a resource partitioning module 51, a monitoring module 52, and a resource scheduling module 53, wherein...

[0097] The resource partitioning module 51 is configured to partition heterogeneous resource pools according to heterogeneous resource types. The heterogeneous resource types include information technology (IT) resources and communication technology (CT) resources. The heterogeneous resource pools include a first resource pool for providing basic resources to business data with IT resource requirements and a second resource pool for providing basic resources to business data with CT resource requirements.

[0098] Monitoring module 52 is configured to monitor the first resource occupancy of the first resource pool and the second resource occupancy of the second resource pool in real time.

[0099] The resource scheduling module 53 is configured to adjust the basic resources in the first resource pool and the second resource pool according to the first resource occupancy status and the second resource occupancy status.

[0100] In one implementation, the basic resources include at least one of the following types of resources: computing resources, storage resources, or network resources.

[0101] In one embodiment, the resource allocation module 51 includes:

[0102] The acquisition unit is configured to acquire the historical resource usage of resources corresponding to heterogeneous resource types.

[0103] The determination unit is configured to determine the initial resource allocation ratio between the first resource pool and the second resource pool based on the historical resource occupancy; and the partitioning unit is configured to partition the heterogeneous resource pool based on the total basic resources of the MEC device and the initial resource allocation ratio.

[0104] In one embodiment, the adjustment and allocation module 53 is specifically configured to adjust the basic resources in the first resource pool and the second resource pool according to the resource occupancy status of the first resource pool and the second resource pool when the resources occupied by the first resource pool or the second resource pool reach a preset proportion of the total basic resources of the MEC device.

[0105] In one embodiment, the device further includes:

[0106] The identification module is configured to identify the resources corresponding to the heterogeneous resource types required by the first service data packet in response to the first service data packet initiated by the target service terminal.

[0107] The first allocation module is configured to allocate resources to the first service data packet based on the first resource pool when the resource corresponding to the heterogeneous resource type is an IT resource, so as to process the first service data packet.

[0108] The second allocation module is configured to allocate resources to the first service data packet based on the second resource pool when the resource corresponding to the heterogeneous resource type is a CT resource, so as to process the first service data packet.

[0109] In one embodiment, the device further includes:

[0110] The priority configuration module is configured to configure the scheduling priority of each type of basic resource in the first resource pool and the second resource pool respectively.

[0111] The resource scheduling module is configured to allocate resources to the multiple second service data packets that simultaneously require IT resources and / or CT resources, based on the first resource pool and / or the second resource pool and the scheduling priority.

[0112] In one implementation, the basic resources include computing resources, storage resources, and network resources.

[0113] The priority configuration module is specifically configured to configure the scheduling priority of each type of basic resource in the first resource pool as computing resources > storage resources > network resources, and to configure the scheduling priority of each type of basic resource in the second resource pool as network resources > storage resources > computing resources.

[0114] In an exemplary embodiment of this application, combined with Figure 6 As shown, the heterogeneous resource scheduling device comprises a basic resource module 61 (i.e., resource allocation module 51), a virtualization module 62, an IT monitoring module 63 (i.e., monitoring module 52), a CT monitoring module 64 (i.e., monitoring module 52), a service function module 65, a network function module 66, and a resource adjustment module 67 (i.e., resource scheduling module 53).

[0115] The basic resource module 61 stores basic resources, including computing resources, storage resources, and network transmission resources. These basic resources are homogeneous resources that every upper-layer application needs to access. The virtualization module 62 virtualizes each resource in the basic resource module according to a unified interface and standard, thereby forming a resource pool within the device. This resource pool can be accessed by the business function module 65 and the network function module 66. In this embodiment, the resource pool is divided into a first resource pool and a second resource pool, which are accessed by the business function module 65 and the network function module 66, respectively.

[0116] Business function module 65 corresponds to IT resources, which can be presented in the form of a specific business system; network function module 66 corresponds to CT resources, which can be presented in the form of network devices or functions such as Broadband Access Server (BRAS) and Service Router (SR); business function modules and network function modules can have built-in business analysis functions, which can identify the types of IT business data and CT business data, and extract the corresponding (scheduling) priority information from the packets.

[0117] The IT monitoring module 63 is connected to the business function module 65 and the virtualization module 62 respectively. It obtains the usage of basic resources by the business function module 65 through internal signaling interaction. At the same time, the IT monitoring module 63 is connected to the resource adjustment module 67 and shares the resource usage information with the resource adjustment module 67.

[0118] The CT monitoring module 64 is connected to the network function module 66 and the virtualization module 62 respectively. It obtains the usage of basic resources by the network function module 66 through internal signaling interaction. At the same time, the IT monitoring module 64 is connected to the resource adjustment module 67 and shares the resource usage information with the resource adjustment module 67.

[0119] The resource adjustment module 67 is connected to both the business function module 65 and the network function module 66, and obtains the priority requirements of IT business data and CT business data through internal signaling interaction. Simultaneously, as mentioned above, the resource adjustment module 67 is also connected to the IT monitoring module 63 and the CT monitoring module 64, enabling real-time monitoring of the basic resource usage by the business function module 65 and the network function module 66. Based on all input information and a pre-deployed preset mechanism (such as the token mechanism described below), the resource adjustment module 67 dynamically allocates basic resources for the business function module 65 and the network function module 66, and sends the relevant decisions to both modules. Subsequently, during business processing, the business function module 65 and the network function module 66, constrained by this strategy, utilize the virtualization module to access basic resources, thereby improving the utilization rate of IT and CT resources.

[0120] The preset mechanism can be as follows: A token mechanism is introduced within the resource adjustment module. The number of tokens is set to N. The size of N corresponds to the total remaining amount of basic resources. Simply put, assuming the total remaining amount of basic resources is quantified as 100, then N is set to 100. In subsequent processes, the resource adjustment module will allocate tokens to the business function modules or network function modules based on the real-time remaining amount of basic resources. Only with a token can one apply for the corresponding basic resources. Since the types of basic resources differ, the actual tokens are also divided into N1, N2, ..., Ni, etc., depending on the resource type such as CPU, NPU, and storage.

[0121] To minimize the complexity of internal session flows, a threshold Y is set for basic resources, for example, Y = 60%. When the total utilization of basic resources by the business function module or network function module exceeds Y, the resource adjustment module will activate the token mechanism; otherwise, the resource adjustment module will not activate this mechanism. Since the IT detection module and CT detection module can detect the occupancy of basic resources in real time and send it to the resource adjustment module, the resource adjustment module can know whether the current resource occupancy exceeds the threshold Y.

[0122] It's important to note that, unlike the setting of N, Y is a total value and a unified indicator. In other words, Y is a weighted average of the utilization rates of various basic resources. Specifically, Y = y1*Y1 + y2*Y2 + ... + yi*Yi. Here, Y1, Y2, ..., Yi represent the utilization rate of each basic resource, and y1, y2, ..., yi are pre-defined weights.

[0123] Once the token mechanism is enabled, the resource adjustment module will allocate resources by distributing tokens according to the needs of the business function module or network function module.

[0124] Specifically, at any given time, the needs of business function modules and network function modules for different basic resources may differ. For example, business function modules often involve cloud computing services, thus having high demands for CPU, NPU, and GPU. In terms of priority matching, this means that business function modules assign high priority to basic resources like CPU, NPU, and GPU, while assigning low priority to resources such as FPGA and storage. Similarly, network function modules involve communication services, thus having high demands for FPGA, storage, and transmission resources. Again, in terms of priority matching, network function modules assign high priority to resources like FPGA and storage, while assigning low priority to basic resources like CPU, NPU, and GPU. The key point of this scheduling mechanism is that, in this situation, when both business function modules and network function modules simultaneously request the use of a certain type of basic resource, the resource adjustment module will prioritize satisfying the corresponding high-priority function module. For example, when both business function modules and network function modules simultaneously request CPU, and the business function module has a higher priority for CPU, the resource adjustment module will prioritize allocating tokens to the business function module to satisfy its access to and use of basic resources, and then satisfy the network function module's request. Under this mechanism, when basic resources are in short supply, priority is given to meeting the needs of important or urgent functional modules for accessing basic resources.

[0125] In actual implementation, the following two modes can be selected regarding the priority of upper-level functional modules:

[0126] The first approach involves configuring the priorities of basic resources for both business function modules and network function modules at the initial system design stage, and storing this information in the resource adjustment module. Subsequent business processing will then follow this configuration.

[0127] The second mode allows business function modules and network function modules to adjust their priority in consuming basic resources in real time, based on their own needs, to better adapt to each other's or the overall requirements. For example, if the network function module has a high priority for CPU usage at a certain moment, the business function module can choose to lower its CPU usage priority based on its own business needs. This improves the overall system's flexibility and achieves better utilization of basic resources.

[0128] This application also provides a corresponding MEC device, such as... Figure 7 As shown, it includes: a processor 72, and a memory 71 communicatively connected to the processor 72;

[0129] The memory 71 stores computer-executed instructions;

[0130] The processor 72 executes the computer execution instructions stored in the memory 71 to implement the heterogeneous resource scheduling method.

[0131] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the heterogeneous resource scheduling method described above.

[0132] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0133] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer.

[0134] Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0135] In the description of the embodiments of this application, the term "and / or" merely indicates a relationship describing the associated objects, meaning that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the term "at least one" indicates any combination of at least two of a plurality of options, for example, including at least one of A, B, and C, which can represent any one or more elements selected from a set including communication between A, B, and C. Furthermore, the term "multiple" means two or more, unless otherwise precisely specified.

[0136] In the description of the embodiments of this application, the terms "first," "second," "third," "fourth," etc. (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A heterogeneous resource scheduling method, characterized in that, The method, applied to mobile edge computing (MEC) devices, includes: Obtain historical resource usage information for resources corresponding to heterogeneous resource types, including IT resources (information technology) and CT resources (communication technology). The initial resource allocation ratio between the first resource pool and the second resource pool is determined based on the historical resource usage; wherein, the first resource pool is used to provide basic resources for business data with IT resource requirements, and the second resource pool is used to provide basic resources for business data with CT resource requirements; Heterogeneous resource pools are divided according to the total basic resources of the MEC devices and the initial resource allocation ratio. The heterogeneous resource pools include the first resource pool and the second resource pool. Real-time monitoring of the first resource occupancy status of the first resource pool and the second resource occupancy status of the second resource pool; When the resources occupied by the first resource pool or the second resource pool reach a preset proportion of the total basic resources of the MEC device, the basic resources in the first resource pool and the second resource pool are dynamically adjusted according to the resource occupancy status of the first resource pool and the second resource pool. The basic resources include at least two of computing resources, storage resources, and network resources. After dividing the heterogeneous resource pools according to the type of heterogeneous resources, it also includes: Configure the scheduling priority of each type of basic resource in the first resource pool and the second resource pool respectively, and the scheduling priority of each type of basic resource in the first resource pool and the second resource pool is different; When there are multiple second service data packets that simultaneously require IT resources and CT resources, resources are allocated to the multiple second service data packets based on the first resource pool, the second resource pool, and the scheduling priority.

2. The method according to claim 1, characterized in that, After dividing the heterogeneous resource pools according to the type of heterogeneous resources, it also includes: In response to a first service data packet initiated by a target service terminal, identify the resources corresponding to the heterogeneous resource types required by the first service data packet; When the resource corresponding to the heterogeneous resource type is an IT resource, resources are allocated to the first service data packet based on the first resource pool to process the first service data packet. When the resource corresponding to the heterogeneous resource type is a CT resource, resources are allocated to the first service data packet based on the second resource pool to process the first service data packet.

3. The method according to claim 1, characterized in that, The basic resources include computing resources, storage resources, and network resources. The step of configuring the scheduling priorities of each type of basic resource in the first resource pool and the second resource pool respectively includes: The scheduling priority of each type of basic resource in the first resource pool is configured as computing resources > storage resources > network resources, and the scheduling priority of each type of basic resource in the second resource pool is configured as network resources > storage resources > computing resources.

4. A heterogeneous resource scheduling device, characterized in that, Applied to MEC devices, the device includes: The resource allocation module is configured to obtain historical resource occupancy information for heterogeneous resource types, including IT resources (information technology) and CT resources (communication technology); determine an initial resource allocation ratio between a first resource pool and a second resource pool based on the historical resource occupancy information; wherein the first resource pool is used to provide basic resources for business data with IT resource requirements, and the second resource pool is used to provide basic resources for business data with CT resource requirements; and allocate heterogeneous resource pools based on the total basic resources of the MEC devices and the initial resource allocation ratio, wherein the heterogeneous resource pools include the first resource pool and the second resource pool. The monitoring module is configured to monitor the first resource occupancy of the first resource pool and the second resource occupancy of the second resource pool in real time. The resource scheduling module is configured to dynamically adjust the basic resources in the first resource pool and the second resource pool according to the resource occupancy status of the first resource pool and the second resource pool when the resources occupied by the first resource pool or the second resource pool reach a preset proportion of the total basic resources of the MEC device. The priority configuration module is configured to configure the scheduling priority of each type of basic resource in the first resource pool and the second resource pool respectively, and the scheduling priorities of each type of basic resource in the first resource pool and the second resource pool are different; The resource scheduling module is further configured to allocate resources to the multiple second service data packets that simultaneously require IT resources and CT resources, based on the first resource pool, the second resource pool, and the scheduling priority.

5. An MEC device, comprising: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the heterogeneous resource scheduling method as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the heterogeneous resource scheduling method as described in any one of claims 1 to 3.

Images