CN115396514B - Resource allocation method, device and storage medium - Google Patents

Abstract

The present disclosure provides a resource allocation method, device, and storage medium, which relate to the field of communication technology and can reasonably allocate computing power resources for computing power services. The method includes: determining the service type of the target service; when the target service is a low-latency, low-computing-demand service, determining n transmission paths between the source node of the target service and at least one computing power node, where n is Positive integer; select the target transmission path that satisfies the preset delay condition from n transmission paths, and allocate the computing nodes corresponding to the target transmission path to the target business; In this case, determine the resources required by the target business; select the target computing power node from at least one computing power node, and allocate the target computing power node to the target business; the available resources of the target computing power node meet the resources required by the target business. This disclosure is used in the computing power node allocation process in business transmission.

Description

Resource allocation method, device and storage medium

technical field

The present disclosure relates to the field of communication technologies, and in particular, to a resource allocation method, device and storage medium.

Background technique

In related technologies, when computing power resources are allocated to computing power services, computing power services are usually allocated to computing power nodes with the closest distance and computing power resources that can meet computing power business needs. However, this allocation method only considers the transmission distance of the computing power business and the computing power resources of the computing power nodes, and does not consider the differences in the needs of different computing power services. As a result, it is prone to unreasonable resource allocation when allocating computing power resources for computing power business.

Contents of the invention

The present application provides a resource allocation method, device and storage medium, which are used to solve the problem of unreasonable allocation of computing power resources in the prior art.

In order to achieve the above object, the application adopts the following technical solutions:

In the first aspect, a resource allocation method is provided, the method includes: determining the service type of the target service; in the case that the target service is a service with low delay and low computing power demand, determining the source node of the target service and at least one computing power n transmission paths between nodes, where n is a positive integer; select a target transmission path that satisfies the preset delay condition from the n transmission paths, and assign the computing power node corresponding to the target transmission path to the target business ; When the target business is a business with high latency and high computing power requirements, determine the resources required by the target business; select the target computing power node from at least one computing power node; the available resources of the target computing power node meet the requirements of the target business required resources; allocate target computing power nodes to target businesses.

With reference to the first aspect above, in a possible implementation manner, the selecting a target transmission path satisfying a preset condition from the n transmission paths includes: determining n transmission paths, and bandwidth utilization of each path rate; determining that among the n paths, the path whose bandwidth utilization rate is less than or equal to the first threshold and the number of links is less than the second threshold is the target transmission path.

In combination with the first aspect above, in a possible implementation, the resources required by the target business include: computing power resources; selecting the target computing power node from at least one computing power node, including: determining the first resource required by the target business Computing power resource requirements of different types of computing power resources; the first type of computing power resources is one of the following: general computing power resources, image computing power resources, and dedicated computing power; among at least one computing power node, the first Available computing power resources of a type of computing power resources; determine the target computing power node according to the first computing power node in at least one computing power node; the available computing power resources of the first type of computing power resources of the first computing power node are greater than human resource needs.

In combination with the first aspect above, in a possible implementation manner, the resources required by the target business further include: storage resources; the method further includes: determining the storage resource requirements of the target business and the available storage resources of the first computing power node; Determine the first computing power node whose available storage resources are greater than the storage resource requirement among the first computing power nodes is the target computing power node.

In combination with the first aspect above, in a possible implementation, there are multiple target computing power nodes; assigning the target computing power nodes to the target business includes:

Determine the number of links in the transmission path from the source node to each target computing power node;

In the transmission path from the source node to each target computing power node, the target computing power node corresponding to the transmission path with the smallest number of links is assigned to the target business.

In combination with the first aspect above, in a possible implementation, the method further includes: when the target business is a low-latency, low-computing-demand business, configuring the first priority for the target business; In the case of delay and high computing power demand business, configure the second priority for the target business; where the first priority is higher than the second priority.

In the second aspect, a resource allocation device is provided, including: a processing unit; the processing unit is used to determine the type of the target service; the processing unit is also used to determine the n transmission paths between the source node of the target service and at least one computing power node, where n is a positive integer; the processing unit is also used to select a target transmission path that satisfies the preset delay condition from the n transmission paths, and The computing power node corresponding to the target transmission path is allocated to the target business; the processing unit is also used to determine the resources required by the target business when the target business is a business with high latency and high computing power requirements; the processing unit is also used to Select the target computing power node from at least one computing power node; the available resources of the target computing power node meet the resources required by the target business; the processing unit is also used to allocate the target computing power node to the target business.

In combination with the second aspect above, in a possible implementation manner, the processing unit is specifically configured to: determine n transmission paths and the bandwidth utilization rate of each path; determine that the bandwidth utilization rate of the n paths is greater than or equal to the first The path with the threshold and the number of links smaller than the second threshold is the target transmission path.

In combination with the second aspect above, in a possible implementation manner, the processing unit is specifically configured to: determine the computing power resource requirements of the first type of computing power resources required by the target business; the first type of computing power resources are one of the following : General computing power resources, image computing power resources, dedicated computing power; determine the available computing power resources of the first type of computing power resources of each computing power node in at least one computing power node; according to the first type of computing power resources in at least one computing power node A computing power node, determining the target computing power node; the available computing power resource of the first type of computing power resource of the first computing power node is greater than the computing power resource requirement.

In combination with the second aspect above, in a possible implementation manner, the processing unit is specifically configured to: determine the storage resource requirements of the target business and the available storage resources of the first computing power node; determine the available storage resources in the first computing power node The first computing power node whose resource is greater than the storage resource requirement is the target computing power node.

In combination with the second aspect above, in a possible implementation manner, the processing unit is specifically configured to: allocate the target computing power node to the target business, including: determining the link of the transmission path from the source node to each target computing power node Number; in the transmission path from the source node to each target computing power node, the target computing power node corresponding to the transmission path with the smallest number of links is assigned to the target business.

In combination with the second aspect above, in a possible implementation manner, the processing unit is specifically configured to: configure the first priority for the target service when the target service is a low-latency, low-computing-demand service; When the business is a high-latency, high-computing demand business, configure the second priority for the target business; wherein, the first priority is higher than the second priority.

In a third aspect, the present disclosure provides a resource allocation device, which includes: a processor and a memory; wherein the memory is used to store computer-executable instructions, and when the resource allocation device is running, the processor Executing the computer-executable instructions stored in the memory, so that the resource allocation apparatus executes the resource allocation method as described in the first aspect and any possible implementation manner of the first aspect.

In a fourth aspect, the present disclosure provides a computer-readable storage medium, in which instructions are stored, and when the instructions in the computer-readable storage medium are executed by a processor of a resource allocation device, the resource allocation device can Execute the resource allocation method as described in the first aspect and any possible implementation manner of the first aspect.

In the present disclosure, the names of the above-mentioned resource allocation apparatuses do not limit the devices or functional modules themselves, and in actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of the present disclosure, they fall within the scope of the claims of the present disclosure and their equivalent technologies.

These or other aspects of the present disclosure will be more clearly understood in the following description.

The technical solution provided by the present disclosure brings at least the following beneficial effects:

In this solution, the resource allocation device allocates computing power resources in different ways for services with different delays and computing power requirements. Among them, for the target business with low latency and low computing power requirements, a low-latency transmission path is selected for the target business, and the computing power nodes corresponding to the transmission path are allocated to the target business. For the target business, select a computing power node whose computing resources can meet the demand and assign it to the target business. In this way, when the application allocates resources for the computing power business, it can combine the time delay and computing power requirements of the computing power business, and select computing power nodes that meet the transmission requirements and computing power requirements for the business, thereby realizing a reasonable allocation for the computing power business Computing resources.

Description of drawings

FIG. 1 is a schematic diagram of a hardware structure of a resource allocation device provided in an embodiment of the present application;

FIG. 2 is a flowchart of a resource allocation method provided by an embodiment of the present application;

FIG. 3 is a flowchart of a resource allocation method provided by an embodiment of the present application;

FIG. 4 is a flowchart of a resource allocation method provided by an embodiment of the present application;

FIG. 5 is a flowchart of a resource allocation method provided by an embodiment of the present application;

FIG. 6 is a structural diagram of a resource allocation device provided by an embodiment of the present application.

Detailed ways

The resource allocation method, device and storage medium provided by the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

The term "and/or" in this article is just an association relationship describing associated objects, which means that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist simultaneously, and there exists alone B these three situations.

The terms “first” and “second” in the specification and drawings of the present disclosure are used to distinguish different objects, or to distinguish different processes for the same object, rather than to describe a specific order of objects.

In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present disclosure are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes other unlisted steps or units, or optionally also includes Other steps or elements inherent to the process, method, product or apparatus are included.

It should be noted that, in the embodiments of the present disclosure, words such as "exemplary" or "for example" are used as examples, illustrations or illustrations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present disclosure shall not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete manner.

FIG. 1 is a schematic structural diagram of a resource allocation device provided by an embodiment of the present disclosure. As shown in FIG. 1 , the resource allocation apparatus 100 includes at least one processor 101 , a communication line 102 , and at least one communication interface 104 , and may further include a memory 103 . Wherein, the processor 101 , the memory 103 and the communication interface 104 may be connected through a communication line 102 .

The processor 101 may be a central processing unit (central processing unit, CPU), or a specific integrated circuit (application specific integrated circuit, ASIC), or one or more integrated circuits configured to implement the embodiments of the present disclosure, For example: one or more digital signal processors (digital signal processor, DSP), or one or more field programmable gate arrays (field programmable gate array, FPGA).

Communication line 102 may include a path for communicating information between the above-described components.

The communication interface 104 is used to communicate with other devices or communication networks, and any device such as a transceiver can be used, such as Ethernet, radio access network (radio access network, RAN), wireless local area network (wireless local area networks, WLAN), etc. .

The memory 103 may be a read-only memory (read-only memory, ROM) or other types of static storage devices that can store static information and instructions, a random access memory (random access memory, RAM) or other types that can store information and instructions The dynamic storage device can also be an electrically erasable programmable read-only memory (electrically erasable programmable read-only memory, EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage ( including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to include or store desired program code in the form of instructions or data structures and can be stored by a computer Any other medium, but not limited to.

In a possible design, the memory 103 may exist independently of the processor 101, that is, the memory 103 may be a memory outside the processor 101. At this time, the memory 103 may be connected to the processor 101 through the communication line 102 for storing and executing Instructions or application program codes are executed under the control of the processor 101 to implement the method for determining resource allocation provided by the following embodiments of the present disclosure. In yet another possible design, the memory 103 can also be integrated with the processor 101, that is, the memory 103 can be an internal memory of the processor 101, for example, the memory 103 is a cache, which can be used to temporarily store some data and instructions information etc.

As an implementable manner, the processor 101 may include one or more CPUs, for example, CPU0 and CPU1 in FIG. 1 . As another implementable manner, the resource allocation apparatus 100 may include multiple processors, for example, the processor 101 and the processor 107 in FIG. 1 . As yet another implementable manner, the resource allocation apparatus 100 may further include an output device 105 and an input device 106 .

Through the description of the above embodiments, those skilled in the art can clearly understand that for the convenience and brevity of the description, only the division of the above-mentioned functional modules is used as an example for illustration. In practical applications, the above-mentioned functions can be allocated according to needs It is completed by different functional modules, that is, the internal structure of the network node is divided into different functional modules to complete all or part of the functions described above. For the specific working process of the system, modules, and network nodes described above, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the current computing power network scenario, the computing power demander that executes computing power business and the computing power provider that provides computing power resources are usually not in the same location. For example, in the currently proposed "East Data, West Computing" scenario, a new type of computing power network system integrating data centers, cloud computing, and big data is mainly established in the western region, and the intensive computing power services in the eastern region are transmitted through communication links to the In the computing power network in the western region, the computing power network in the western region processes the computing power business and returns the processing results to the eastern region. In this way, the cross-domain flow of data elements is realized, and the construction layout of the data center is optimized.

Since the transmission of computing power services from the eastern region to the western region needs to be transmitted through a long path, how to reasonably and effectively allocate computing power resources for computing power services is a technical problem that needs to be solved urgently.

In related technologies, when allocating computing power resources for computing power services, usually only the transmission distance and computing power resources of the computing power business are considered, and the computing power business is allocated to the computing power with the closest distance and the computing power resources can meet the computing power business needs. in the force node. However, the transmission delay requirements and computing performance requirements of the current computing power business are different. The resource allocation method in the prior art only considers the transmission distance of the computing power business and the computing power resources of the computing power nodes, and cannot take into account both computing power and computing power. Latency requirements and computing power requirements of power services. This leads to unreasonable resource allocation when allocating computing power resources for computing power business.

In addition, there may be problems such as unscientific business allocation, or insufficient utilization of computing power node resources, which will lead to a decline in user experience and interruption of setting services. Moreover, the resource requirements of the computing power business are not considered when the computing power nodes are currently assigned to the computing power business.

In order to reasonably allocate resources for computing power services and improve the business performance and user experience of computing power services, the embodiment of this application provides a resource allocation method. The resource allocation device adopts different methods for services with different delays and computing power requirements. Allocate computing resources. Among them, for the target business with low latency and low computing power requirements, a low-latency transmission path is selected for the target business, and the computing power nodes corresponding to the transmission path are allocated to the target business. For the target business, select a computing power node whose computing resources can meet the demand and assign it to the target business. In this way, when the application allocates resources for the computing power business, it can combine the time delay and computing power requirements of the computing power business, and select computing power nodes that meet the transmission requirements and computing power requirements for the business, so as to achieve a reasonable allocation for the computing power business The effect of computing resources.

The resource allocation method provided by the embodiment of the present disclosure can be applied to the resource allocation device shown in FIG. 1 , and as shown in FIG. 2 , the resource allocation method provided by the embodiment of the present disclosure can be implemented through the following steps 201 to 206.

Step 201, the resource allocation device determines the service type of the target service.

Wherein, the service type of the target service includes: a low-latency, low-computing-power-demanding type, and a high-latency, high-computing-power-demanding type.

In a possible implementation manner, the resource allocation device classifies the services according to the delay requirements and computing power requirements of the services, and divides the services into low-latency, low-computing-power services, and high-delay, high-computing-power services. Among them, the low-latency, low-computing-power business is a business whose required transmission delay is less than or equal to a first threshold, and required computing power resources are less than or equal to a second threshold. High-latency, high-computing-power services are services that require transmission delays greater than a first threshold and require computing resources greater than a second threshold.

Optionally, low-latency, low-computing-power services have strict requirements on transmission delay, and during allocation, the transmission delay of the service is preferentially required to be less than the first threshold.

An example, for the face recognition business, it is necessary to send the face image to the computing power node, and the computing power node will match the face image with the image in the database. Since the image recognition business usually requires high real-time performance, but requires less computing power resources, it is only the computing power resources for matching the collected face images with the images in the data. Therefore, the resource allocation device can define the face recognition service as a service with low delay and low computing power.

Optionally, high-latency, high-computing-power services have strict requirements on computing power resources, and the computing power resources of the business are preferentially required to be greater than the second threshold during allocation.

As an example, to establish a business for the face model database, the face model must first be trained for data. At this time, a large number of face photos need to be sent to the computing power node (requires a lot of storage space), and the computing power node will A large number of face photos are trained to obtain a face model, and finally a face model database is generated. In the process of model training, computing power nodes need to provide a large amount of computing power resources for model training, and model training takes a long time, so the business of establishing face model database will not have too high requirements on transmission delay. Therefore, the resource allocation device may define the face model database establishment service as a high-latency, high-computing-power service.

It can be understood that the type of the target service may also be a high-latency, low-computing-power service, or a low-latency, high-computing-power service. Among them, high-latency and low-computing-power services, since there are no requirements for computing power and time-delay, belong to the lowest priority business. According to the principle of nearest distribution, they can be allocated to the nearest computing power node to ensure the normal operation of the business. Low-latency, high-computing power services have high requirements for both delay and computing power, and are the highest priority business. In order to ensure the normal operation of the business, dedicated transmission lines can be set up for such services to ensure the normal transmission of the business.

It should be pointed out that for target services with low latency and low computing power requirements, and target services with high latency and high computing power requirements, the resource allocation device uses different methods to allocate computing power resources for the target services.

Specifically, when the target business is a low-latency, low-computing-demand business, the resource allocation device allocates computing power resources for the target business in the manner described in steps 202 and 203 below.

In the case that the target service is a service with high latency and high demand for computing power, the resource allocation device allocates computing power resources for the target service in the manner described in steps 204 to 206 below. The details will be described below.

Step 202, in the case that the target service is a low-latency, low-computing-power-demanding service, the resource allocation device determines n transmission paths between the source node of the target service and at least one computing power node.

Wherein, n is a positive integer.

In a possible implementation manner, the resource allocation device determines at least one currently available computing power node, and determines the transmission path between the target service source node and each computing power node in the at least one computing power node, and obtains the above n transfer path. Each transmission path includes at least one transmission link.

In step 203, the resource allocation device selects a target transmission path satisfying a preset delay condition from the n transmission paths, and allocates computing power nodes corresponding to the target transmission path to the target service.

An example, the preset delay condition includes: the transmission delay of the service is less than a preset delay threshold.

In a possible implementation manner, the shorter the transmission distance of the service (or the fewer links the transmission passes through), the smaller the transmission delay of the service. Therefore, the resource allocation device may sort the n transmission paths according to the number of included links in descending order, and select the path with the least number of links as the target transmission path. Wherein, the link number of the path is equal to the router hop number+1 between the source node and the destination node.

In yet another possible implementation, when the bandwidth utilization rate of the path is greater than a certain value (for example, 70%), the transmission delay of the link will be greatly affected, so the resource allocation device performs the calculation of the n transmission paths according to the included chain After sorting the number of paths in descending order, delete transmission paths including links whose bandwidth utilization rate is greater than 70%. Then select the path with the least number of links from the deleted paths as the target transmission path.

After the resource allocation device selects the target transmission path, it allocates the computing power node corresponding to the target transmission path to the target service, so that the computing power node processes the target service.

Optionally, when the bandwidth utilization of each transmission link does not meet the preset delay condition, the resource allocation device sorts the n transmission paths in ascending order according to the bandwidth utilization of the transmission links, and determines the top-ranked The transmission path is used as the target transmission path, and the computing nodes corresponding to the target transmission path are assigned to the target transmission business.

Optionally, when none of the bandwidth utilization rates determined by the resource allocation device meets the preset delay condition, and the difference between the bandwidth utilization rates of the transmission paths does not exceed 3%, the service transmission path can also be configured according to the method described in step 202. Determine the destination transfer path.

Step 204: In the case that the target service is a service with high latency and high computing power demand, the resource allocation device determines resources required by the target service.

Wherein, the resources required by the target business include: at least one of computing resources and storage resources.

Step 205, the resource allocation device selects a target computing power node from the at least one computing power node.

Among them, the available resources of the target computing power node meet the resources required by the target business.

In a possible implementation manner, the resource allocation device determines available resources of each computing power node in at least one computing power node. The resource allocation device selects a computing power node whose available resources are greater than the resources required by the target business as the target computing power node.

Optionally, when the resources required by the target business include computing power resources, the resource allocation device determines the available computing power resources of each computing power node, and selects the available computing power resources greater than the computing power resources required by the target business. The node is the target computing power node.

In the case that the resources required by the target business also include storage resources, after the resource computing power node determines that the available computing power resources are greater than the computing power resources required by the target business (referred to as the first node), determine each first The available storage resources of the nodes, and select the nodes whose available storage resources are greater than the storage resources required by the target business from the first node as the target computing power node.

Step 206, the resource allocation device allocates the target computing power node to the target business.

In a possible implementation, after the resource allocation device determines the optimal target computing power node, the resource allocation device determines the transmission path between the target computing power node and the source node of the target service, and transmits the transmission path to the target computing power node through this path. Transfer target business.

The above solution brings at least the following beneficial effects. The resource allocation device allocates computing power resources in different ways for services with different delays and computing power requirements. Among them, for the target business with low latency and low computing power requirements, a low-latency transmission path is selected for the target business, and the computing power nodes corresponding to the transmission path are allocated to the target business. For the target business, select a computing power node whose computing resources can meet the demand and assign it to the target business. In this way, when the application allocates resources for the computing power business, it can combine the time delay and computing power requirements of the computing power business, and select computing power nodes that meet the transmission requirements and computing power requirements for the business, thereby realizing a reasonable allocation for the computing power business Computing resources.

It should be pointed out that the type of the target service may also be a high-latency, low-computing-power service, or a low-latency, high-computing-power service. Among them, the high-latency and low-computing-power business, since there is no requirement for computing power and time-delay, is the lowest priority business. According to the nearest distribution principle, it can be allocated to the nearest computing power node to ensure the normal operation of the business; low Latency and high computing power services have high requirements on both time delay and computing power, and belong to the highest priority business. In order to ensure the normal operation of the business, a dedicated transmission line can be set up for this type of business to ensure the normal transmission of the business.

Referring to FIG. 2 , as shown in FIG. 3 , the above step 203 may also be specifically implemented by one or more steps in the following steps 301 - 302 .

Step 301, the resource allocation device determines n transmission paths and the bandwidth utilization rate of each path.

Specifically, after determining n paths from the source node of the target service to the target computing power node, the resource allocation device further determines the number of links in each of the n paths. The number of links for each path refers to the number of links between the source node of the target business and the target computing power node.

In step 302, the resource allocation device determines that among the n paths, a path whose bandwidth utilization rate is less than or equal to a first threshold and whose number of links is less than a second threshold is a target transmission path.

In a possible implementation manner, after determining the number of links of the n paths, the resource allocation device sorts the n paths in ascending order according to the number of links of each path. At the same time, the resource allocation device acquires the bandwidth utilization rate of each path. In this way, the resource allocation device determines the number of links of each path and the bandwidth utilization rate of each path, and then from the n paths, the path whose bandwidth utilization rate is less than or equal to the first threshold and the number of links is less than the second threshold is destination transfer path.

An example, the first threshold is 70%. The size of the second threshold may be determined according to the ranking result of the number of links between the target service and the target computing power node, which is not limited in the present disclosure.

A specific example is for face recognition and matching business. The resource allocation device determines that there are currently four available computing power nodes, namely: computing power node A, computing power node B, computing power node C, and computing power node D. First, the resource allocation device determines the number of links from the source node of the target service to the four computing power nodes. For example, the number of links to four computing power nodes are: computing power node A: 4 hops, computing power node B: 2 hops, computing power node C: 3 hops, computing power node D: 5 hops. Next, the resource allocation device sorts the four target paths in ascending order, and the sorting results are computing power node B, computing power node C, computing power node A, and computing power node D.

Further, the resource allocation device judges whether any of the four paths has a bandwidth utilization rate exceeding 70%. For example, if there are 5 links to computing power node A, the bandwidth is 100G, and the peak traffic is 50G, the bandwidth utilization rate is 50%; there are 3 links to computing power node B, the bandwidth is 50G, and the peak traffic is 50G. The traffic is 10G, 30G, and 40G respectively, and the bandwidth utilization rates are 20%, 60%, and 80% respectively; there are 4 links to computing power node C, with bandwidths of 20G, 50G, 100G, and 200G, and the peak traffic is respectively 10G, 20G, 80G, 100G, the bandwidth utilization rate is 50%, 40%, 80%, 50% respectively; there are 6 links to computing power node D, the bandwidth is 200G, and the peak traffic is 50G, then the bandwidth The utilization rate is 25%. Then, because the bandwidth utilization rate of the path to computing power node B and computing power node C exceeds 70%, the transmission delay will be relatively large and cannot meet the business needs. Therefore, the resource allocation device deletes the computing power from the above sorting results Force node B and computing power node C, get a new ranking result: computing power node A computing power node D. Since the number of links of the computing power node A is small, the ranking of the computing power node A is higher, therefore, the resource allocation device allocates the business to the computing power node A.

Optionally, among the target transmission paths determined by the resource allocation device, the bandwidth utilization rate exceeds 70% and the number of links is less than the second threshold. At this time, the bandwidth utilization rate is taken as the primary consideration condition for determining the target transmission path. For example, the resource allocation device judges that there are currently three available computing power nodes (computing power node E, computing power node F, and computing power node G), and the number of links from the target service source node to the three computing power nodes is Power node E: 2 hops, computing power node F: 3 hops, computing power node G: 2 hops. Among them, the bandwidth utilization rates of the three links to computing power node E are 20%, 50%, and 90% respectively; the bandwidth utilization rates of the four links to computing power node F are 85%, 40%, and 30% respectively. %, 20%; the bandwidth utilization rates of the three links to computing node G are 95%, 30%, and 50%, respectively.

Further, the resource allocation device determines that the bandwidth utilization ratios of the current target paths are all greater than 70%. At this time, the bandwidth utilization rate is taken as the primary judgment condition to select the path with the smallest bandwidth utilization rate. Then, since the path to computing power node F has the smallest bandwidth utilization rate, the business is allocated to computing power node F.

It should be noted that during service transmission, the transmission delay of the target service is mainly affected by the link with the highest bandwidth utilization among the transmission links, so only the link with the highest bandwidth utilization is considered when selecting the target link.

It is worth noting that when the bandwidth utilization exceeds 70%, if the bandwidth utilization difference between the target transmission paths is less than or equal to 3%, the number of links is still the primary consideration for selecting the target transmission path.

Referring to FIG. 2 , as shown in FIG. 4 , the above steps 204 - 205 may be specifically implemented by one or more of the following steps 401 - 408 .

Step 401, the resource allocation device determines the first type of computing power resource requirements of the target business.

Among them, the first type of computing power resources include: general computing power resources, image computing power resources, and dedicated computing power resources.

An example, for example, general floating-point computing services belong to general-purpose computing services, which mainly rely on the central processing unit (CPU) for calculation; image processing services belong to image computing services, mainly relying on image processors (graphics processing unit, GPU) to calculate; Field Programmable Gate Array (field programmable gate array, FPGA) business belongs to the special computing power business, which requires a special processor to operate.

Step 402, the resource allocation device determines the available computing power resources and remaining storage space of the first type of computing power resources of each computing power node in at least one computing power node.

In a possible implementation manner, the resource allocation device judges whether the remaining computing power of chip resources corresponding to the target service in each computing power node can meet the available computing power resources of the first type of computing power resources.

An example, as shown in Table 1 below, is the remaining available computing power resources of each computing power node.

Table 1. The remaining available computing power resources of each computing power node

Step 403, the resource allocation device determines a target computing power node according to the first computing power resource in at least one computing power node.

In a possible implementation manner, the resource allocation device determines a target computing power node that satisfies the target service according to the first computing power resources available in each computing power node.

It is worth noting that a target business usually only requires one type of computing resource. Therefore, in the target computing power node, it is only necessary that the corresponding first type of computing power resource can meet the computing power demand of the target business. If the target business uses multiple types of computing power resources at the same time, the target computing power node needs the computing power resources of the multiple types of computing power to meet the computing power requirements of the target business. In the embodiment of the present disclosure, the target service only needs to use one kind of computing resources as an example for illustration.

An example, as shown in Table 2 below, is the computing power requirements of different types of services in this application.

Table 2. Computing power requirements for different types of businesses

business name business type Computing power demand storage requirements business one General computing power 20TFLOPS 20T business two Image computing power 20TFLOPS 40T business three General computing power 15TFLOPS 20T

Exemplary, combined with Table 1. For target business one, first, the resource allocation device judges that the computing power resource of the first type of each computing power node meets the computing power requirement of business one. Therefore, the resource allocation device determines all computing power nodes in Table 1 as target computing power nodes.

Optional, combined with Table 1. For the target service 2, firstly, the resource allocation device judges that only two computing power nodes, computing power node A and computing power node C, can meet the computing power requirements of the second business. Therefore, the resource allocation device determines that the target computing power nodes in Table 1 are computing power node A and computing power node C.

Optional, combined with Table 1. For the target service three, firstly, the resource allocation device judges that the first type of computing power resources of the computing power nodes include computing power node A, computing power node B, and computing power node C, which can meet the computing power requirements of business three. Therefore, the resource allocation device determines that the target computing power nodes in Table 1 are computing power node A, computing power node B, and computing power node C.

In step 404, the resource allocation device determines that among the first computing power nodes, the first computing power node whose available storage resource is greater than the storage resource requirement is the target computing power node.

In a possible implementation manner, the resource allocation device further determines the storage requirements of the target business and the storage resources of the target computing power node determined in step 403 after determining the first type of computing power resources that meet the target business.

Exemplarily, Table 1 and Table 2 are combined. For the target business 1, the resource allocation device further determines the computing power nodes meeting the storage requirements of the target business 1 from the target computing power nodes determined in step 403 as two target computing power nodes, computing power node A and computing power node C .

Optionally, combine Table 1 and Table 2. For the target business 2, the resource allocation device further determines that the computing power node meeting the storage requirement of the target business 2 is a target computing power node C from the target computing power nodes determined in step 403 .

Optionally, combine Table 1 and Table 2. For the target business three, the resource allocation device further determines the computing power node meeting the storage requirement of the target business three from the target computing power nodes determined in step 403 as a target computing power node of computing power node A.

Step 405, the resource allocation device determines the number of links in the transmission path from the source node to each target computing power node.

In a possible implementation manner, the resource allocation device determines the number of router hops of the transmission path from the source node to each target computing power, and determines the number of links according to the number of router hops. Wherein, the link number of the transmission path is equal to the router hop number+1 between the source node and the destination node.

Exemplarily, for example, the number of router hops from the source node to computing power node A in step 302 is 4 hops, and the number of links from the source node to computing power node A is 5.

Step 406, the resource allocation device assigns the target computing power node corresponding to the transmission path with the smallest number of links among the transmission paths from the source node to each target computing power node to the target service.

In a possible implementation, after the resource allocation device determines at least one target computing power node that meets the computing power resource requirements and storage requirements of the target business, it selects the transmission distance according to the transmission distance between the target business source node and the target computing power node. The path with the smallest distance is used as the target transmission path.

Exemplarily, for example, there are two computing power nodes available for service one in step 404, computing power node A and computing power node C. Assume that the number of links from service 1 to computing power node A and computing power node C, two available computing power nodes, is 3 hops and 5 hops respectively. The resource allocation device determines the target transmission path as the path between the target service source node and the target computing power node A according to the principle of the nearest transmission.

It should be pointed out that, in the embodiment of the present disclosure, the resource allocation device encrypts the collected data, and encrypts the data during data transmission, so as to avoid leakage of user information caused by data theft.

In a possible implementation manner, the resource allocating device may also configure priority for the target service, so that the router preferentially transmits high-priority services according to the priority of the target service. Specifically, as shown in step 407 and step 408 in FIG. 4 .

In step 407, the resource allocation device configures the first priority for the target service when the target service is a service with low latency and low computing power demand.

One possible implementation method, low-latency, low-computing-power services have high requirements on latency. The resource allocation device marks the target service as the first priority service during transmission, and then allocates it to the computing power node with the smallest number of links and the smallest bandwidth utilization rate.

In step 408, the resource allocation device configures the second priority for the target service when the target service is a service with high latency and high computing power demand.

Wherein, the first priority is higher than the second priority.

In a possible implementation mode, high-delay and high-computing-power services do not have high requirements for time delay. The resource allocation device marks the target service as the second-priority service during transmission, and when encountering the first-priority service during transmission, Then, the transmission is suspended until the transmission of the first priority service is completed.

It should be noted that this application does not limit the order of steps 407-408 and steps 401-406. The resource allocation device may first perform steps 401-406, and then perform steps 407-408. Alternatively, the resource allocation apparatus may also perform steps 407-408 first, and then perform steps 401-406. Alternatively, the resource allocation apparatus may also execute steps 407-408, and steps 401-406 at the same time, which is not limited in this application.

Above, the resource allocation method provided by the embodiment of the present application has been described in detail.

In the following, with reference to Figure 5, the overall process of allocating computing power resources for the target business after the resource allocation device receives the target business will be described:

Step 501, the resource allocation device acquires a target service.

Step 502, the resource allocation device determines the service type of the target service.

Wherein, the specific implementation manner of step 502 is similar to the above-mentioned step 201, and the specific implementation process may refer to the above-mentioned step 201, which will not be repeated here.

Specifically, when the target business is a low-latency, low-computing-demand business, the resource allocation device allocates computing power resources for the target business in the manner described in steps 503 and 506 below.

In the case that the target service is a service with high latency and high demand for computing power, the resource allocation device allocates computing power resources for the target service in the manner described in steps 507 to 513 below.

In the case that the target service is a service with low latency and high demand for computing power, the resource allocation device allocates computing power resources for the target service in the manner described in step 514 below. The details will be described below.

It should be pointed out that, when the target service is a service with high latency and low computing power demand, the existing resource allocation method can meet the demand of the target service, which is not limited in this disclosure.

Step 503, in the case that the target service is a service with low latency and low computing power, the resource allocation device sorts in ascending order according to the number of links in the transmission path from the source node to at least one computing power node.

Wherein, the specific implementation manner of step 503 is similar to the above-mentioned step 301, and the specific implementation process may refer to the above-mentioned step 301, which will not be repeated here.

Step 504, the resource allocation device deletes the transmission path from the source node to at least one computing power node with a bandwidth utilization rate greater than a first threshold.

In a possible implementation manner, after determining the bandwidth utilization rate of the transmission path from the source node to at least one computing power node, the resource allocation device deletes the transmission path whose bandwidth utilization rate is greater than the first threshold.

Step 505, the resource allocation device allocates the target service to the transmission path with the smallest number of links.

Wherein, the specific implementation manners of the above-mentioned step 504 and step 505 are similar to the above-mentioned step 302, and the specific implementation process may refer to the above-mentioned step 302, which will not be repeated here.

Step 506, the resource allocation device configures the first priority for the target service.

In a possible implementation manner, when the resource allocation device transmits the target service, it marks the data packet of the target service with the first priority, and transmits it preferentially during the transmission process.

Wherein, the specific implementation manner of step 506 is similar to the above-mentioned step 406, and the specific implementation process may refer to the above-mentioned step 406, which will not be repeated here.

Step 507, in the case that the target service is a service with high latency and high computing power, the resource allocation device determines various resource conditions in the current computing power hub node.

Wherein, the specific implementation manner of step 507 is similar to the above-mentioned step 402, and the specific implementation process may refer to the above-mentioned step 402, which will not be repeated here.

Step 508, the resource allocation device determines the type of computing power required by the business.

Wherein, the specific implementation manner of step 508 is similar to the above-mentioned step 401, and the specific implementation process may refer to the above-mentioned step 401, which will not be repeated here.

Step 509, the resource allocation device determines whether the remaining computing power resources of the computing power nodes meet the business requirements.

Wherein, the specific implementation manner of step 509 is similar to the above-mentioned step 403, and the specific implementation process may refer to the above-mentioned step 403, which will not be repeated here.

Step 510, the resource allocation device determines whether the remaining storage space satisfies the target service requirement.

Wherein, the specific implementation manner of step 510 is similar to the above-mentioned step 404, and the specific implementation process may refer to the above-mentioned step 404, which will not be repeated here.

Step 511, the resource allocation device allocates the target service to the transmission path with the smallest number of links.

Wherein, the specific implementation manner of step 511 is similar to the above-mentioned step 405, and the specific implementation process may refer to the above-mentioned step 405, which will not be repeated here.

It should be pointed out that the resource allocation device allocates the target service to at least one computing power node whose available computing power resources and available storage resources meet the needs of the target service, and the transmission node with the smallest number of links between the source node and the target service on the path.

Step 512, the resource allocation device marks the target service as the second priority service.

In a possible implementation manner, when the resource allocation device transmits the target service, it marks the data packet of the target service with the second priority.

Wherein, the specific implementation manner of step 512 is similar to the above-mentioned step 407, and the specific implementation process may refer to the above-mentioned step 407, which will not be repeated here.

Step 513, the resource allocation device updates various resource conditions in the current computing power node.

In a possible implementation manner, the resource allocation device also updates various resource conditions of at least one available computing power node in real time while transmitting services with high latency and high computing power. To ensure that the next high-latency, high-computing power business can be allocated to the best computing power node.

Step 514, when the target service is a low-latency, high-computing service, the resource allocation device allocates the target service to the dedicated transmission line.

One possible implementation method, low-latency, high-computing-power services have high requirements for both delay and computing power, and dedicated transmission lines can be deployed for such services to ensure the normal operation of such services.

Above, the device for service transmission involved in the embodiment of the present disclosure, and the functions of each device of the resource allocation device, and the interaction between the devices are described in detail.

It can be seen that the foregoing mainly introduces the technical solutions provided by the embodiments of the present disclosure from the perspective of methods. In order to realize the above functions, it includes corresponding hardware structures and/or software modules for performing various functions. Those skilled in the art should easily realize that, in combination with the modules and algorithm steps of the examples described in the embodiments disclosed herein, the embodiments of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software drives hardware depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementation should not be considered beyond the scope of the present disclosure.

The embodiments of the present disclosure may divide the resource allocation device into functional modules according to the above method examples. For example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. Optionally, the division of modules in the embodiments of the present disclosure is schematic, and is only a logical function division, and there may be another division manner in actual implementation.

The embodiments of the present disclosure may divide the resource allocation device into functional modules according to the above method examples. For example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. Optionally, the division of modules in the embodiments of the present disclosure is schematic, and is only a logical function division, and there may be another division manner in actual implementation.

An embodiment of the present disclosure provides a resource allocation device, which is used to execute the method required by any device in the above-mentioned data integrity determination system. The resource allocation device may be the resource allocation device involved in the present disclosure, or a module in the resource allocation device; or a chip in the resource allocation device, or other devices for executing the method for determining resource allocation. No limit.

As shown in FIG. 6 , it is a schematic structural diagram of a resource allocation device provided by an embodiment of the present disclosure. The resource allocation apparatus includes: a communication unit 601 and a processing unit 602 .

The processing unit 602 is configured to determine the type of the target business; and is also used to determine n transmissions between the source node of the target business and at least one computing power node when the target business is a low-latency, low-computing-demand business Path, n is a positive integer; it is also used to select a target transmission path that meets the preset conditions from n transmission paths to transmit the target service; it is also used when the target service is a high-latency, high-computing demand service, Determine the resources required by the target business; it is also used to select the target computing power node from at least one computing power node; the available resources of the target computing power node meet the resources required by the target business; it is also used to The transmission path between them transmits the target service.

Optionally, the processing unit 602 is specifically configured to: determine the link number of each path in the n paths, and the bandwidth utilization rate of each path;

Determining a path among the n paths whose bandwidth utilization rate is greater than or equal to the first threshold and whose number of links is less than the second threshold is the target transmission path.

Optionally, the processing unit 602 is specifically configured to: determine the computing power resource requirements of the first type of computing power resources required by the target business; the first type of computing power resource is one of the following: general computing power resources, image computing power resources , dedicated computing power;

Determining available computing power resources of the first type of computing power resources of each computing power node in at least one computing power node;

A target computing power node is determined according to a first computing power node in the at least one computing power node; the available computing power resource of the first type of computing power resource of the first computing power node is greater than the computing power resource requirement.

Optionally, the processing unit 602 is specifically configured to: determine the storage resource requirements of the target business and the available storage resources of the first computing power node;

Determine the first computing power node whose available storage resources are greater than the storage resource requirement among the first computing power nodes is the target computing power node.

Optionally, the processing unit 602 is specifically configured to: transmit the target service according to the transmission path between the source node and the target computing power node, including:

Determine the number of links in the transmission path from the source node to each target computing power node;

In the transmission path from the source node to each target computing power node, select the transmission path with the smallest number of links to transmit the target service.

Optionally, the processing unit 602 is specifically configured to: configure the first priority for the target service when the target service is a low-latency, low-computing-demand service;

When the target service is a service with high latency and high computing power demand, configure the second priority for the target service; wherein, the first priority is higher than the second priority.

Optionally, the device further includes a communication unit 601; the communication unit 601 is used to obtain the network parameters of the first high-speed rail user terminal; the network parameters include: measurement reports and user bills; the processing unit 602 is also used to obtain the first high-speed rail user terminal The network parameters of the terminal determine the high-speed rail user behavior model; wherein, the high-speed rail user behavior model is used to determine the business proportion of the high-speed rail user terminal business that is greater than the preset proportion; the processing unit 602 is also used to determine according to the high-speed rail user behavior model target business.

An embodiment of the present disclosure provides a resource allocation device, which is used to execute the method required by any device in the above-mentioned data integrity determination system. The resource allocation device may be the resource allocation device involved in the present disclosure, or a module in the resource allocation device; or a chip in the resource allocation device, or other devices for executing the method for determining resource allocation. No limit.

Embodiments of the present disclosure also provide a computer-readable storage medium, in which instructions are stored. When a computer executes the instructions, the computer executes each step in the method flow shown in the above-mentioned method embodiments.

Embodiments of the present disclosure provide a computer program product containing instructions, and when the instructions are run on a computer, the computer is made to execute the resource allocation method in the above method embodiments.

Embodiments of the present disclosure provide a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run computer programs or instructions to implement the resource allocation method in the above method embodiments.

Wherein, the computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connection having one or more wires, portable computer disk, hard disk. Random Access Memory (Random Access Memory, RAM), Read-Only Memory (Read-Only Memory, ROM), Erasable Programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), registers, hard disk, optical fiber, portable compact Disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any other form of computer-readable storage medium in a suitable combination of the above, or values in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may be located in an application specific integrated circuit (Application Specific Integrated Circuit, ASIC). In the embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device.

Since the devices, equipment, computer-readable storage media, and computer program products in the embodiments of the present disclosure can be applied to the above-mentioned methods, the technical effects that can be obtained can also refer to the above-mentioned method embodiments, and the embodiments of the present disclosure are hereby No longer.

The above is only a specific implementation of the present disclosure, but the protection scope of the present disclosure is not limited thereto, and any changes or substitutions within the technical scope disclosed in the present disclosure shall be covered within the protection scope of the present disclosure . Therefore, the protection scope of the present disclosure should be determined by the protection scope of the claims.

Claims (10)

1. A resource allocation method, characterized in that, comprising: Determine the business type of the target business; When the target service is a low-latency, low-computing-demand service, determine n transmission paths between the source node of the target service and at least one computing power node, where n is a positive integer; determining the n transmission paths, and the bandwidth utilization of each transmission path; determining that among the n paths, a transmission path whose bandwidth utilization rate is less than or equal to a first threshold and whose number of links is less than a second threshold is the target transmission path; In the case that the target service is a high-delay and high-computing-demand service, determine resources required by the target service; Determine the computing power resource requirements of the first type of computing power resources required by the target business; the first type of computing power resources are one of the following: general computing power resources, image computing power resources, and dedicated computing power; determining available computing power resources of the first type of computing power resources of each computing power node in the at least one computing power node; Determine the target computing power node according to a first computing power node in the at least one computing power node; the available computing power resource of the first type of computing power resource of the first computing power node is greater than the computing power resource requirements; Allocating the target computing power node to the target service. 2. The method according to claim 1, wherein the resources required by the target business further comprise: storage resources; the method further comprises: determining the storage resource requirements of the target business and the available storage resources of the first computing power node; Determining that among the first computing power nodes, a first computing power node whose available storage resource is greater than the storage resource requirement is the target computing power node. 3. The method according to claim 2, characterized in that the number of target computing power nodes is multiple; The allocating the target computing power node to the target service includes: determining the number of links in the transmission path from the source node to each of the target computing power nodes; In the transmission path from the source node to each of the target computing power nodes, the target computing power node corresponding to the transmission path with the smallest number of links is allocated to the target service. 4. The method according to any one of claims 1-3, wherein the method further comprises: When the target service is a low-latency, low-computing-demand service, configure the first priority for the target service; If the target service is a service with high latency and high computing power demand, configure a second priority for the target service; wherein the first priority is higher than the second priority. 5. A resource allocation device, comprising: a processing unit; The processing unit is configured to determine the type of the target service; The processing unit is further configured to determine n transmission paths between the source node of the target service and at least one computing power node when the target service is a low-latency, low-computing-demand service, n is a positive integer; The processing unit is further configured to determine the n transmission paths and the bandwidth utilization rate of each transmission path; The processing unit is further configured to determine, among the n paths, a transmission path whose bandwidth utilization rate is less than or equal to a first threshold and whose number of links is less than a second threshold is the target transmission path; The processing unit is further configured to determine resources required by the target service when the target service is a service with high latency and high computing power requirements; Determine the computing power resource requirements of the first type of computing power resources required by the target business; the first type of computing power resources are one of the following: general computing power resources, image computing power resources, and dedicated computing power; The processing unit is further configured to determine the available computing power resources of the first type of computing power resources of each computing power node in the at least one computing power node; The processing unit is further configured to determine the target computing power node according to the first computing power node in the at least one computing power node; the availability of the first type of computing power resources of the first computing power node The computing power resource is greater than the computing power resource requirement; The processing unit is further configured to allocate the target computing power node to the target service. 6. The device according to claim 5, wherein the processing unit is specifically used for: determining the storage resource requirements of the target business and the available storage resources of the first computing power node; Determining that the first computing power node whose available storage resources are greater than the storage resource requirement among the first computing power nodes is the target computing power node. 7. The device according to claim 6, wherein the processing unit is specifically used for: The allocating the target computing power node to the target service includes: determining the number of links in the transmission path from the source node to each of the target computing power nodes; In the transmission path from the source node to each of the target computing power nodes, the target computing power node corresponding to the transmission path with the smallest number of links is allocated to the target service. 8. The device according to any one of claims 5-7, wherein the processing unit is further configured to: When the target service is a low-latency, low-computing-demand service, configure the first priority for the target service; If the target service is a service with high latency and high computing power demand, configure a second priority for the target service; wherein the first priority is higher than the second priority. 9. A resource allocation device, characterized by comprising: a processor and a memory; wherein the memory is used to store computer-executable instructions, and when the resource allocation device is running, the processor executes the instructions stored in the memory The computer executes instructions, so that the resource allocation device executes the resource allocation method according to any one of claims 1-4. 10. A computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when the instructions in the computer-readable storage medium are executed by the processor of the resource allocation device, the resource allocation The device executes the resource allocation method described in any one of claims 1-4.

Images