Assessment scheduling method, system and device for computing power resources and medium

Abstract

The invention discloses a computing power resource evaluation scheduling method, system and device and a medium, and the method comprises the steps: obtaining business computing power demand data, a demand business type of the business computing power demand data, and processing computing power information, network computing power information, memory computing power information and storage computing power information of a data center; according to the demand service type, a center weight set is obtained, and the center weight set comprises a target processing weight, a target network weight, a target memory weight and a target storage weight; according to the target processing weight, the target network weight, the target memory weight and the target storage weight, performing resource evaluation on the processing computing power information, the network computing power information, the memory computing power information and the storage computing power information to obtain resource evaluation information; and according to the resource evaluation information, carrying out computing power scheduling distribution on the business computing power demand data. The method can effectively improve the accuracy and practicability of computing power resource scheduling. The invention relates to the technical field of data center operation and maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of data center operation and maintenance, and in particular to an evaluation and scheduling method, system, device and medium for computing power resources. Background Art

[0002] With the rapid development of information technology, as an important infrastructure to support various applications and services, the efficient management of the computing power of data centers has become an important task for optimizing resource utilization, reducing operating costs and improving service quality.

[0003] Currently, the existing evaluation and scheduling methods for data center computing power resources usually allocate and schedule corresponding computing power according to the computing power requirements of business applications. The accuracy of the computing power resources provided to business applications in this way is relatively low and the practicability is not high.

[0004] Therefore, the problems existing in the prior art still need to be solved and optimized urgently. Summary of the Invention

[0005] An object of the present invention is to solve at least to some extent one of the technical problems existing in the related art.

[0006] To this end, an object of an embodiment of the present invention is to provide an evaluation and scheduling method, system, device and medium for computing power resources, wherein the method can effectively improve the accuracy and practicability of computing power resource scheduling.

[0007] In order to achieve the above technical object, the technical solutions adopted in the embodiments of the present application include:

[0008] In a first aspect, an embodiment of the present application provides an evaluation and scheduling method for computing power resources, including:

[0009] Obtain business computing power demand data and the required business type of the business computing power demand data, as well as the processing computing power information, network computing power information, memory computing power information and storage computing power information of the data center;

[0010] According to the required business type, obtain a central weight set, the central weight set includes a target processing weight, a target network weight, a target memory weight and a target storage weight, and the object weight is negatively correlated with the usage of the corresponding resources of the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight or the target storage weight;

[0011] According to the target processing weight, the target network weight, the target memory weight and the target storage weight, perform resource evaluation on the processing computing power information, the network computing power information, the memory computing power information and the storage computing power information to obtain resource evaluation information;

[0012] Perform computing power scheduling and allocation on the service computing power demand data according to the resource evaluation information.

[0013] In addition, according to the method of the above embodiments of the present application, the following additional technical features may also be included:

[0014] Further, in an embodiment of the present application, obtaining processing computing power information includes:

[0015] Obtain the processing frequency, the number of processing cores, and the current processing load of the data center;

[0016] Perform processing computing power calculation on the current processing load according to the processing frequency and the number of processing cores to obtain the processing computing power information;

[0017] Obtain network computing power information, including:

[0018] Obtain the network bandwidth, network utilization rate, and current network latency of the data center;

[0019] Perform network computing power calculation on the current network latency according to the network bandwidth and the network utilization rate to obtain the network computing power information.

[0020] Further, in an embodiment of the present application, obtaining memory computing power information includes:

[0021] Obtain the memory frequency, bus bandwidth, and current memory load of the data center;

[0022] Perform memory computing power calculation on the current memory load according to the memory frequency and the bus bandwidth to obtain the memory computing power information;

[0023] Obtain storage computing power information, including:

[0024] Obtain the storage throughput, storage response time, and number of operations per second of the data center;

[0025] Perform storage computing power calculation on the number of operations per second according to the storage throughput and the storage response time to obtain the storage computing power information.

[0026] Further, in an embodiment of the present application, the obtaining the central weight set according to the required service type includes:

[0027] Obtain the original weight set, resource usage data set, and historical resource data set of the data center at the current moment;

[0028] Perform constraint update on the original weight set according to the resource usage data set and the historical resource data set to obtain the central weight set.

[0029] Further, in an embodiment of the present application, the step of constraining and updating the original weight set according to the resource usage data set and the historical resource data set to obtain the central weight set includes:

[0030] Performing data analysis and load prediction on the historical resource data set to obtain a load prediction data set;

[0031] According to the load prediction data set, classifying the resource usage data set to obtain a resource classification data set, where the resource classification data set includes processing classification information, memory classification information, storage classification information, and network classification information. The target classification information is used to indicate whether the corresponding computing power resource is a bottleneck resource or an idle resource, and the target classification information is any one of the processing classification information, the memory classification information, the storage classification information, or the network classification information;

[0032] According to the processing classification information, the memory classification information, the storage classification information, and the network classification information, performing relevant updates on the original weight set to obtain the central weight set.

[0033] Further, in an embodiment of the present application, the step of classifying the resource usage data set according to the load prediction data set to obtain a resource classification data set includes:

[0034] Obtaining the processing resource prediction information, memory resource prediction information, storage resource prediction information, and network resource prediction information of the load prediction data set, as well as the processing resource usage information, memory resource usage information, storage resource usage information, and network resource usage information in the resource usage data set;

[0035] Performing first information classification on the processing resource prediction information according to the processing resource usage information to obtain the processing classification information;

[0036] Performing second information classification on the memory resource prediction information according to the memory resource usage information to obtain the memory classification information;

[0037] Performing third information classification on the storage resource prediction information according to the storage resource usage information to obtain the storage classification information;

[0038] Performing fourth information classification on the network resource prediction information according to the network resource usage information to obtain the network classification information.

[0039] Further, in the embodiments of the present application, the relevant update of the original weight set according to the processing classification information, the memory classification information, the storage classification information, and the network classification information to obtain the central weight set includes:

[0040] Obtain the original processing weight, original network weight, original memory weight, and original storage weight of the original weight set;

[0041] If the processing classification information indicates that the processing resources are bottleneck resources, then according to the processing classification information, perform a weight negative correlation constraint update on the original processing weight to obtain the target processing weight; or, if the processing classification information indicates that the processing resources are idle resources, then according to the processing classification information, perform a weight positive correlation constraint update on the original processing weight to obtain the target processing weight;

[0042] If the network classification information indicates that the network resources are bottleneck resources, then according to the network classification information, perform a weight negative correlation constraint update on the original network weight to obtain the target network weight; or, if the network classification information indicates that the network resources are idle resources, then according to the network classification information, perform a weight positive correlation constraint update on the original network weight to obtain the target network weight;

[0043] If the memory classification information indicates that the memory resources are bottleneck resources, then according to the memory classification information, perform a weight negative correlation constraint update on the original memory weight to obtain the target memory weight; or, if the memory classification information indicates that the memory resources are idle resources, then according to the memory classification information, perform a weight positive correlation constraint update on the original memory weight to obtain the target memory weight;

[0044] If the storage classification information indicates that the storage resources are bottleneck resources, then according to the storage classification information, perform a weight negative correlation constraint update on the original storage weight to obtain the target storage weight; or, if the storage classification information indicates that the storage resources are idle resources, then according to the storage classification information, perform a weight positive correlation constraint update on the original storage weight to obtain the target storage weight.

[0045] In a second aspect, the embodiments of the present application provide an evaluation and scheduling system for computing power resources, including:

[0046] A first processing unit, configured to obtain business computing power demand data and the required business type of the business computing power demand data, as well as the processing computing power information, network computing power information, memory computing power information, and storage computing power information of the data center;

[0047] A second processing unit, configured to obtain a central weight set according to the required service type, where the central weight set includes a target processing weight, a target network weight, a target memory weight, and a target storage weight, and the object weight is negatively correlated with the usage of the corresponding resources of the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight, or the target storage weight;

[0048] A third processing unit, configured to perform resource evaluation on the processing computing power information, the network computing power information, the memory computing power information, and the storage computing power information according to the target processing weight, the target network weight, the target memory weight, and the target storage weight, so as to obtain resource evaluation information;

[0049] A fourth processing unit, configured to perform computing power scheduling and allocation on the service computing power demand data according to the resource evaluation information.

[0050] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0051] At least one processor;

[0052] At least one memory, configured to store at least one program;

[0053] When the at least one program is executed by the at least one processor, the at least one processor implements the method in the first aspect above.

[0054] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, in which a program executable by a processor is stored, and the program executable by the processor is used to implement the method in the first aspect above when executed by the processor.

[0055] The advantages and beneficial effects of the present application will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present application:

[0056] An evaluation and scheduling method, system, device and medium for computing power resources disclosed in an embodiment of the present application. In this method, business computing power demand data, the required business type of the business computing power demand data, as well as the processing computing power information, network computing power information, memory computing power information and storage computing power information of the data center are obtained; according to the required business type, a central weight set is obtained, and the central weight set includes a target processing weight, a target network weight, a target memory weight and a target storage weight. The object weight is negatively correlated with the usage of the corresponding resources in the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight or the target storage weight; according to the target processing weight, the target network weight, the target memory weight and the target storage weight, resource evaluation is performed on the processing computing power information, the network computing power information, the memory computing power information and the storage computing power information to obtain resource evaluation information; according to the resource evaluation information, computing power scheduling and allocation are performed on the business computing power demand data. This method obtains the central weight set through the required business type, and performs resource evaluation on the computing power resources (such as processing resources, network resources, memory resources and storage resources) of the data center through the central weight set based on negative correlation constraints, which can effectively improve the accuracy of the resource evaluation information of the data center and is beneficial to improving the accuracy and practicability of computing power resource scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following introduces the accompanying drawings related to the technical solutions in the embodiments of the present application or the prior art. It should be understood that the accompanying drawings in the following introduction are only for conveniently and clearly expressing some embodiments of the technical solutions in the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative efforts.

[0058] Figure 1 It is a flowchart showing the evaluation and scheduling method for computing power resources provided by an embodiment of the present application;

[0059] Figure 2 It is a structural diagram showing the evaluation and scheduling system for computing power resources provided by an embodiment of the present application;

[0060] Figure 3 It is a structural diagram showing an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] Embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application, and should not be construed as a limitation of the present application. For the step numbers in the following embodiments, they are only set for the convenience of description and explanation, and no limitation is imposed on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0063] Currently, the existing methods for evaluating and scheduling computing power resources in data centers usually allocate and schedule corresponding computing power according to the computing power requirements of business applications. However, since the computing power of a data center is jointly provided by multiple different types of resources (such as CPU resources, memory resources, storage resources, and network bandwidth resources), and different types of resources contribute differently to the computing power of the data center, the computing power resources provided in this way often do not match well with business applications, cannot meet the computing power requirements of business applications well, have low accuracy, and low practicality. In addition, this method easily causes some types of resources in the data center to run at full load while some other resources are in an idle state. It cannot adjust timely and flexibly according to the actual operating conditions of the data center and the requirements of different workloads. The computing power resources of the data center are not fully utilized, there is a waste of computing power resources, and the efficiency and accuracy of resource allocation in the data center are not satisfactory.

[0064] In view of this, embodiments of the present invention provide a method, system, device, and medium for evaluating and scheduling computing power resources. Among them, the method obtains the central weight set of the data center for the current demand business type, and evaluates the computing power resources (such as processing resources, network resources, memory resources, and storage resources) of the data center based on the central weight set with negative correlation constraints. It can not only effectively improve the accuracy of data center resource evaluation information, which is beneficial to improving the accuracy and practicality of computing power resource scheduling; at the same time, it can also effectively improve the rationality of allocating various types of resources in the data center, which is beneficial to avoiding waste of computing power resources, improving the utilization rate of computing power resources in the data center, enabling the data center to adjust timely and flexibly based on the actual operating conditions and different workload requirements, and thus being beneficial to improving the efficiency and accuracy of data center resource allocation.

[0065] Referring to Figure 1 , in the embodiments of the present application, a method for evaluating and scheduling computing power resources includes:

[0066] Step 110: Obtain the business computing power demand data and the required business type of the business computing power demand data, as well as the processing computing power information, network computing power information, memory computing power information, and storage computing power information of the data center;

[0067] In the embodiments of the present application, the business computing power demand data may be the computing power demand of a business application that requires the data center to provide computing power resources, and the required business type may be the device usage of the business application. The specific device usage may be divided into general-purpose, compute-intensive, memory-intensive, storage-intensive, etc.

[0068] It can be understood that the processing computing power information of the data center may be the CPU computing power at the current moment of the data center, the network computing power information may be the network computing power at the current moment of the data center, and the memory computing power information and the storage computing power information are similar to the processing computing power information and can be simply inferred.

[0069] In some embodiments, obtaining the processing computing power information includes:

[0070] A1: Obtain the processing frequency, number of processing cores, and current processing load of the data center;

[0071] A2: Calculate the processing computing power of the current processing load according to the processing frequency and the number of processing cores to obtain the processing computing power information;

[0072] Obtaining the network computing power information includes:

[0073] B1: Obtain the network bandwidth, network utilization rate, and current network latency of the data center;

[0074] B2: Calculate the network computing power of the current network latency according to the network bandwidth and the network utilization rate to obtain the network computing power information.

[0075] In the embodiments of the present application, the processing frequency, number of processing cores, CPU computing power coefficient, and current processing load of the CPU can be obtained through the monitoring module of the data center, and then the processing computing power information of the data center regarding processing resources at the current moment is determined based on the obtained processing frequency, number of processing cores, CPU computing power coefficient, and current processing load. The processing computing power information can be expressed as:

[0076] comp1 = fre1 * N c * load1 * β1

[0077] where comp1 is the processing computing power information; fre1 is the processing frequency; load1 is the current processing load; N c is the number of processing cores; β1 is the CPU computing power coefficient.

[0078] It is understandable that the overall computing power of the data center is also affected by network bandwidth and latency. Therefore, the network bandwidth, network utilization rate, network computing power coefficient, and current network latency of the data center can also be obtained through the monitoring module of the data center, and the network computing power information of the data center at the current moment regarding the network can be calculated based on the obtained network bandwidth, network utilization rate, network computing power coefficient, and current network latency. The network computing power information can be expressed as:

[0079] comp2 = BW * N u * fn * β2

[0080] Wherein, comp2 is the network computing power information; BW is the network bandwidth; N u is the network utilization rate; fn is the current network latency; β2 is the network computing power coefficient.

[0081] In some embodiments, obtaining the memory computing power information includes:

[0082] C1. Obtaining the memory frequency, bus bandwidth, and current memory load of the data center;

[0083] C2. Performing memory computing power calculation on the current memory load according to the memory frequency and the bus bandwidth to obtain the memory computing power information;

[0084] Obtaining the storage computing power information includes:

[0085] D1. Obtaining the storage throughput, storage response time, and number of operations per second of the data center;

[0086] D2. Performing storage computing power calculation on the number of operations per second according to the storage throughput and the storage response time to obtain the storage computing power information.

[0087] It is understandable that the memory and storage performance of the data center affect the speed of data processing in the data center. Therefore, in the embodiments of the present application, the memory frequency, bus bandwidth, current memory load, and memory computing power coefficient of the data center memory can be obtained through the monitoring module of the data center. Among them, the current memory load can specifically be the memory utilization rate of the data center at the current moment; then, the memory computing power information is determined based on the obtained memory frequency, bus bandwidth, current memory load, and memory computing power coefficient. The memory computing power information can be expressed as:

[0088] comp3 = fre2 * BB * load2 * β3

[0089] Wherein, comp3 is the memory computing power information; fre2 is the memory frequency; load2 is the current memory load; BB is the bus bandwidth; β3 is the memory computing power coefficient.

[0090] It should be noted that the storage throughput, storage response time, storage computing power coefficient, and number of operations per second of the data center can be obtained through the monitoring module of the data center. Among them, the number of operations per second can be the number of input / output operations per second of the data center storage system; then, based on the obtained storage throughput, storage response time, and number of operations per second, storage computing power information is determined, and the storage computing power information can be expressed as:

[0091]

[0092] Among them, comp4 is the storage computing power information; TP is the storage throughput; RT is the storage response time; IOPS is the number of operations per second; β4 is the storage computing power coefficient.

[0093] Step 120: According to the required service type, obtain a set of central weights. The object weight is negatively correlated with the usage of the corresponding resources of the data center. The object weight can be any one of the target processing weight, the target network weight, the memory weight, or the target storage weight.

[0094] In the embodiments of the present application, a set of central weights corresponding to the data center can be obtained based on the required service type indicated by the service application. The set of central weights includes a target processing weight, a target network weight, a target memory weight, and a target storage weight. The target processing weight is the weight ratio of the processing resources of the data center at the current moment, and the target processing weight is negatively correlated with the usage of the processing resources of the data center; the target network weight is the weight ratio of the network resources of the data center at the current moment, and the target network weight is negatively correlated with the usage of the network resources of the data center; the target memory weight is the weight ratio of the memory resources of the data center at the current moment, and the target memory weight is negatively correlated with the usage of the memory resources of the data center; the target storage weight is the weight ratio of the storage resources of the data center at the current moment, and the target storage weight is negatively correlated with the usage of the storage resources of the data center.

[0095] In some embodiments, step 120: According to the required service type, obtain a set of central weights, including:

[0096] E1: Obtain the original weight set, resource usage data set, and historical resource data set of the data center at the current moment.

[0097] In the embodiments of the present application, the original weight set may be a set of initial weights preset based on the demand business type. The original weight set includes an original processing weight, an original network weight, an original memory weight, an original storage weight, etc. Exemplarily, if the demand business type is general, its original processing weight, original network weight, original memory weight, and original storage weight are all 0.25; or, if the demand business type is computationally intensive, its original processing weight may be 0.4, while the original network weight, original memory weight, and original storage weight may be 0.2. The weight values for the demand business types of memory-intensive or storage-intensive are similar to those of the computationally intensive demand business type mentioned above and can be simply deduced by analogy. Additionally, in practical applications, the original weight set may be the central weight set obtained during the computing power scheduling and allocation process for the previous same demand business type.

[0098] It should be noted that the resource usage data set may be a set of resource usage data of various types at the current moment in the data center, which specifically includes CPU usage data, memory usage data, storage usage data, network usage data, etc. Among them, the CPU usage data may include CPU processing load, CPU processing load fluctuation, etc.; the memory usage data may include the current memory occupancy, remaining memory space, etc.; the storage usage data may include the occupancy rate of the storage device, I / O performance, etc.; the network usage data may include network traffic, bandwidth utilization, etc. Additionally, the historical resource data may be a set of resource usage data of various types in the data center during a historical time period, and its content is similar to that of the resource usage data set, which will not be elaborated herein in the present application.

[0099] E2. Constraint-update the original weight set according to the resource usage data set and the historical resource data set to obtain the central weight set.

[0100] Further, the step E2. Constraint-update the original weight set according to the resource usage data set and the historical resource data set to obtain the central weight set includes:

[0101] E21. Perform data analysis and load prediction on the historical resource data set to obtain a load prediction data set;

[0102] In the embodiments of the present application, the periodic changes and long-term change trends of the resource usage data of various types in the data center during a historical time period may be analyzed to predict the resource requirements that the data center needs to meet for various types at a future moment, and the predicted resource requirements are determined as the load prediction data set. The specific data analysis and load prediction methods may be implemented through machine learning or prediction algorithms (such as regression analysis, time series analysis, etc.), which will not be elaborated herein in the present application.

[0103] E22. Classify the resource usage dataset according to the load prediction dataset to obtain a resource classification dataset, where the resource classification dataset includes processing classification information, memory classification information, storage classification information, and network classification information. The target classification information is used to indicate whether the corresponding computing power resource is a bottleneck resource or an idle resource, and the target classification information is any one of the processing classification information, the memory classification information, the storage classification information, or the network classification information;

[0104] Further, the step E22. Classify the resource usage dataset according to the load prediction dataset to obtain a resource classification dataset, including:

[0105] E221. Obtain the processing resource prediction information, memory resource prediction information, storage resource prediction information, and network resource prediction information of the load prediction dataset, as well as the processing resource usage information, memory resource usage information, storage resource usage information, and network resource usage information in the resource usage dataset;

[0106] E222. Perform a first information classification on the processing resource prediction information according to the processing resource usage information to obtain the processing classification information;

[0107] E223. Perform a second information classification on the memory resource prediction information according to the memory resource usage information to obtain the memory classification information;

[0108] E224. Perform a third information classification on the storage resource prediction information according to the storage resource usage information to obtain the storage classification information;

[0109] E225. Perform a fourth information classification on the network resource prediction information according to the network resource usage information to obtain the network classification information.

[0110] In the embodiments of the present application, the resource requirements of the data center at a future moment can be obtained according to the load prediction dataset, which can be specifically divided into processing resource prediction information, memory resource prediction information, storage resource prediction information, and network resource prediction information; and according to the resource usage dataset, the resource usage situation of the data center at the current moment can be obtained, which can be specifically divided into processing resource usage information, memory resource usage information, storage resource usage information, and network resource usage information.

[0111] It can be understood that for the processing resources in the data center, information analysis and classification can be performed on the processing resource usage information and the processing resource prediction information. It can analyze whether the data center based on the current processing resource usage information can meet the processing resource requirements indicated by the processing resource prediction information, and when the data center meets the processing resource requirements indicated by the processing resource prediction information, generate processing classification information indicating that the processing resources are idle resources; or, when the data center does not meet the processing resource requirements indicated by the processing resource prediction information, generate processing classification information indicating that the processing resources are bottleneck resources.

[0112] Specifically, in one feasible implementation, it can be determined whether the sum of the processing resource requirements of the processing resource usage information at the current moment and the processing resource prediction information at a future moment is within the load range of the data center's processing resources, and when the sum of the processing resource requirements of the processing resource usage information at the current moment and the processing resource prediction information at a future moment is within the load range of the data center's processing resources, generate processing classification information indicating that the processing resources are idle resources; or, if the sum of the processing resource requirements of the processing resource usage information at the current moment and the processing resource prediction information at a future moment is outside the load range of the data center's processing resources, generate processing classification information indicating that the processing resources are bottleneck resources.

[0113] In another feasible implementation, based on the periodic changes and long-term change trends of the resource usage data of various types in the data center during the historical time period, the processing resource usage information at a future moment can be calculated in combination with the processing resource usage information at the current moment, and then it can be determined whether the total processing resource requirements are within the load range of the data center's processing resources based on the processing resource usage information and the processing resource prediction information at the future moment, so as to obtain the processing classification information.

[0114] It should be noted that for the memory classification information, the storage classification information, and the network classification information, their contents are similar to the foregoing processing classification information and can be simply analogized.

[0115] E23. Update the original weight set according to the processing classification information, the memory classification information, the storage classification information, and the network classification information to obtain the central weight set.

[0116] Further, the step E23 of updating the original weight set according to the processing classification information, the memory classification information, the storage classification information, and the network classification information to obtain the central weight set includes:

[0117] E231. Obtain the original processing weight, the original network weight, the original memory weight, and the original storage weight of the original weight set;

[0118] E232. If the processing classification information indicates that the processing resource is a bottleneck resource, update the original processing weight with a negative weight correlation constraint according to the processing classification information to obtain the target processing weight; or, if the processing classification information indicates that the processing resource is an idle resource, update the original processing weight with a positive weight correlation constraint according to the processing classification information to obtain the target processing weight.

[0119] E233. If the network classification information indicates that the network resource is a bottleneck resource, update the original network weight with a negative weight correlation constraint according to the network classification information to obtain the target network weight; or, if the network classification information indicates that the network resource is an idle resource, update the original network weight with a positive weight correlation constraint according to the network classification information to obtain the target network weight.

[0120] E234. If the memory classification information indicates that the memory resource is a bottleneck resource, update the original memory weight with a negative weight correlation constraint according to the memory classification information to obtain the target memory weight; or, if the memory classification information indicates that the memory resource is an idle resource, update the original memory weight with a positive weight correlation constraint according to the memory classification information to obtain the target memory weight.

[0121] E235. If the storage classification information indicates that the storage resource is a bottleneck resource, update the original storage weight with a negative weight correlation constraint according to the storage classification information to obtain the target storage weight; or, if the storage classification information indicates that the storage resource is an idle resource, update the original storage weight with a positive weight correlation constraint according to the storage classification information to obtain the target storage weight.

[0122] In the embodiments of the present application, the relevant update of the original object weight can be performed respectively based on the classification information of the corresponding resources. Specifically, the relevant update of the original processing weight can be performed based on the processing classification information, the relevant update of the original network weight can be performed based on the network classification information, the relevant update of the original memory weight can be performed based on the memory classification information, and the relevant update of the original storage weight can be performed based on the storage classification information. The obtained target processing weight, target network weight, target memory weight, and target storage weight are determined as the central weight set.

[0123] It can be understood that for processing resources, if the processing classification information indicates that the processing resources belong to bottleneck resources, it means that the load of the processing resources in the data center is too high and has become a bottleneck. At this time, based on the processing classification information, the original processing weight can be updated with a negative weight correlation constraint to obtain a target processing weight smaller than the original processing weight, so as to avoid the deviation of the computing power resource evaluation from the real situation due to the too high load in the data center; or, if the processing classification information indicates that the processing resources belong to idle resources, it means that the processing resources in the data center are idle. At this time, based on the processing classification information and the load range of the processing resources, the original processing weight can be updated with a positive weight correlation constraint to obtain a target processing weight greater than the original processing weight and within the load range of the processing resources.

[0124] It should be noted that the content of steps E233 to E235 is similar to the content of the foregoing step E232 and can be simply analogized. Therefore, this application will not elaborate here.

[0125] Step 130: According to the target processing weight, the target network weight, the target memory weight, and the target storage weight, perform a resource evaluation on the processing computing power information, the network computing power information, the memory computing power information, and the storage computing power information to obtain resource evaluation information;

[0126] In the embodiments of the present application, based on the target processing weight, the target network weight, the target memory weight, and the target storage weight, the corresponding computing power information can be weighted and summed to obtain resource evaluation information based on weighted summation. Alternatively, based on the target processing weight, the target network weight, the target memory weight, and the target storage weight, the corresponding computing power information can be weighted and averaged to obtain resource evaluation information based on weighted averaging.

[0127] Exemplarily, the obtained resource evaluation information based on weighted summation can be expressed as:

[0128] RCP = W CPU *comp1 + W NetWork *comp2 + W Memory *comp3 + W Storage *comp4

[0129] Wherein, RCP is the resource evaluation information; W CPU is the target processing weight; W NetWork is the target network weight; W Memory is the target memory weight; W Storage is the target storage weight.

[0130] It should be noted that in the embodiments of the present application, the object weight is negatively correlated with the corresponding resources in the data center. It can indicate that the data center restricts the corresponding resources within the load range, and the obtained resource evaluation information can fully consider the contributions of different types of resources to the data center. As a result, the data center can be adjusted in a timely and flexible manner based on the actual operating conditions and different workload requirements, which is beneficial to improving the rationality of resource allocation for various types of resources in the data center, and further beneficial to improving the efficiency and accuracy of resource allocation in the data center. Specifically, for example, if the object weight is the target processing weight, it indicates that the data center restricts its processing resources within the processing load range, and this processing load range can be determined by the parameters of the processing resources in the data center or obtained through pre-setting. The same applies to other types of object weights and can be simply deduced by analogy.

[0131] Step 140: According to the resource evaluation information, perform computing power scheduling and allocation on the service computing power demand data.

[0132] In the embodiments of the present application, after the data center obtains resource evaluation information that better fits the actual operating conditions of the data center and different workload requirements, it can perform allocation and scheduling on the computing power requirements of the service computing power demand data based on the obtained resource evaluation information, thereby improving the efficiency and accuracy of resource allocation in the data center.

[0133] Next, a computing power resource evaluation and scheduling system proposed according to an embodiment of the present application will be described in detail with reference to the accompanying drawings.

[0134] Refer to Figure 2 , a computing power resource evaluation and scheduling system proposed in the embodiments of the present application includes:

[0135] The first processing unit 101 is configured to obtain service computing power demand data and the required service type of the service computing power demand data, as well as the processing computing power information, network computing power information, memory computing power information, and storage computing power information of the data center;

[0136] The second processing unit 102 is configured to obtain a central weight set according to the required service type. The central weight set includes a target processing weight, a target network weight, a target memory weight, and a target storage weight. The object weight is negatively correlated with the usage of the corresponding resources in the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight, or the target storage weight;

[0137] The third processing unit 103 is configured to perform resource evaluation on the processing computing power information, the network computing power information, the memory computing power information, and the storage computing power information according to the target processing weight, the target network weight, the target memory weight, and the target storage weight to obtain resource evaluation information;

[0138] The fourth processing unit 104 is configured to perform computing power scheduling and allocation on the service computing power demand data according to the resource evaluation information.

[0139] It can be understood that the content in the above method embodiments is applicable to the system embodiments of the present application. The functions specifically implemented by the system embodiments of the present application are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0140] Referring to Figure 3 , an embodiment of the present application further provides an electronic device, including:

[0141] At least one processor 201;

[0142] At least one memory 202, configured to store at least one program;

[0143] When the at least one program is executed by the at least one processor 201, the at least one processor 201 implements the above method embodiments.

[0144] Similarly, it can be understood that the content in the above method embodiments is applicable to the device embodiments of the present application. The functions specifically implemented by the device embodiments of the present application are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0145] An embodiment of the present application further provides a computer-readable storage medium, in which a program executable by a processor 201 is stored, and the program executable by the processor 201 is used to implement the above method embodiments when executed by the processor 201.

[0146] Similarly, the content in the above method embodiments is applicable to the computer-readable storage medium embodiments of the present application. The functions specifically implemented by the computer-readable storage medium embodiments of the present application are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0147] In some alternative embodiments, the functions / operations mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the functions / operations involved, two consecutive blocks shown may actually be executed substantially simultaneously or the blocks can sometimes be executed in the reverse order. In addition, the embodiments presented and described in the flowcharts of the present application are provided by way of example for a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logical flows presented herein. Alternative embodiments are expected, in which the order of various operations is changed and the sub-operations described as part of a larger operation are executed independently.

[0148] In addition, although the present application has been described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the functions and / or features may be integrated in a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It should also be understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present application. Rather, given the attributes, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the modules will be understood within the ordinary skills of an engineer. Thus, those skilled in the art can implement the present application as set forth in the claims without undue experimentation. It should also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the appended claims and their equivalents.

[0149] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of the embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

[0151] More specific examples (nonexhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing it as appropriate, and then storing it in a computer memory.

[0152] It should be understood that various parts of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0153] In the above description of this specification, the descriptions referring to the terms "one embodiment / example", "another embodiment / example", or "certain embodiments / examples", etc., mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0154] Although the embodiments of the present application have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present application, and the scope of the present application is defined by the claims and their equivalents.

[0155] The above has specifically described the preferred embodiments of the present application, but the present application is not limited to the embodiments. Those skilled in the art can make various equivalent deformations or substitutions without departing from the spirit of the present application, and these equivalent deformations or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A method for evaluating and scheduling computing resources, characterized in that: include: Obtaining business computing power demand data and the business type of the business computing power demand data, as well as processing computing power information, network computing power information, memory computing power information, and storage computing power information of the data center; According to the demand business type, a center weight set is obtained, where the center weight set includes a target processing weight, a target network weight, a target memory weight, and a target storage weight, where an object weight is negatively correlated with usage of corresponding resources of the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight, or the target storage weight; Performing resource evaluation on the processing computing power information, the network computing power information, the memory computing power information, and the storage computing power information according to the target processing weight, the target network weight, the target memory weight, and the target storage weight to obtain resource evaluation information; The computing power demand data of the business is scheduled and allocated based on the resource evaluation information.

2. The method according to claim 1, characterized in that Get processing power information, including: Obtaining a processing frequency, a number of processing cores, and a current processing load of the data center; Calculating the processing power of the current processing load according to the processing frequency and the number of processing cores to obtain the processing power information; Get network computing power information, including: Obtain the network bandwidth, network usage and current network latency of the data center; The network computing power is calculated based on the network bandwidth and the network usage rate to obtain the network computing power information.

3. The method according to claim 1, characterized in that Get memory computing power information, including: Obtaining a memory frequency, a bus bandwidth, and a current memory load of the data center; Calculating the memory computing power of the current memory load according to the memory frequency and the bus bandwidth to obtain the memory computing power information; Get storage computing power information, including: Obtaining storage throughput, storage response time, and number of operations per second of the data center; The storage computing power is calculated for the number of operations per second according to the storage throughput and the storage response time to obtain the storage computing power information.

4. The method according to claim 1, characterized in that: The obtaining of a center weight set according to the required service type includes: Obtaining the original weight set and resource usage data set of the data center at the current moment, as well as the historical resource data set; The original weight set is constrained and updated according to the resource usage data set and the historical resource data set to obtain the central weight set.

5. The method according to claim 4, characterized in that The step of performing constraint updating on the original weight set according to the resource usage data set and the historical resource data set to obtain the central weight set includes: Performing data analysis and load forecasting on the historical resource data set to obtain a load forecast data set; According to the load prediction data set, resource classification is performed on the resource usage data set to obtain a resource classification data set, wherein the resource classification data set includes processing classification information, memory classification information, storage classification information, and network classification information, and the target classification information is used to indicate whether the corresponding computing power resource belongs to a bottleneck resource or an idle resource, and the target classification information is any one of the processing classification information, the memory classification information, the storage classification information, or the network classification information; According to the processing classification information, the memory classification information, the storage classification information and the network classification information, the original weight set is updated accordingly to obtain the central weight set.

6. The method according to claim 5, characterized in that The step of classifying the resource usage data set according to the load prediction data set to obtain a resource classification data set includes: Obtaining processing resource prediction information, memory resource prediction information, storage resource prediction information, and network resource prediction information of the load prediction data set, and processing resource usage information, memory resource usage information, storage resource usage information, and network resource usage information in the resource usage data set; performing a first information classification on the processing resource prediction information according to the processing resource usage information to obtain the processing classification information; According to the memory resource usage information, performing a second information classification on the memory resource prediction information to obtain the memory classification information; According to the storage resource usage information, performing a third information classification on the storage resource prediction information to obtain the storage classification information; According to the network resource usage information, the network resource prediction information is subjected to a fourth information classification to obtain the network classification information.

7. The method according to claim 5, characterized in that The updating of the original weight set according to the processing classification information, the memory classification information, the storage classification information and the network classification information to obtain the central weight set includes: Obtaining original processing weights, original network weights, original memory weights, and original storage weights of the original weight set; If the processing classification information indicates that the processing resource is a bottleneck resource, then according to the processing classification information, the original processing weight is updated with a weight negative correlation constraint to obtain the target processing weight; or, if the processing classification information indicates that the processing resource is an idle resource, then according to the processing classification information, the original processing weight is updated with a weight positive correlation constraint to obtain the target processing weight; If the network classification information indicates that the network resource is a bottleneck resource, then according to the network classification information, the original network weight is updated with a weight negative correlation constraint to obtain the target network weight; or, if the network classification information indicates that the network resource is an idle resource, then according to the network classification information, the original network weight is updated with a weight positive correlation constraint to obtain the target network weight; If the memory classification information indicates that the memory resource is a bottleneck resource, then according to the memory classification information, the original memory weight is updated with a weight negative correlation constraint to obtain the target memory weight; or, if the memory classification information indicates that the memory resource is an idle resource, then according to the memory classification information, the original memory weight is updated with a weight positive correlation constraint to obtain the target memory weight; If the storage classification information indicates that the storage resource is a bottleneck resource, then the original storage weight is updated with a negatively correlated constraint based on the storage classification information to obtain the target storage weight; or, if the storage classification information indicates that the storage resource is an idle resource, then the original storage weight is updated with a positively correlated constraint based on the storage classification information to obtain the target storage weight.

8. A computing resource evaluation and scheduling system, characterized in that: include: A first processing unit is used to obtain business computing power demand data and the demand business type of the business computing power demand data, as well as processing computing power information, network computing power information, memory computing power information and storage computing power information of the data center; A second processing unit is used to obtain a center weight set according to the demand business type, where the center weight set includes a target processing weight, a target network weight, a target memory weight, and a target storage weight, where the object weight is negatively correlated with the usage of the corresponding resources of the data center, and the object weight is any one of the target processing weight, the target network weight, the memory weight, or the target storage weight; A third processing unit is used to perform resource evaluation on the processing computing power information, the network computing power information, the memory computing power information, and the storage computing power information according to the target processing weight, the target network weight, the target memory weight, and the target storage weight to obtain resource evaluation information; The fourth processing unit is used to schedule and allocate computing power for the business computing power demand data according to the resource evaluation information.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the method according to any one of claims 1 to 7 when executed by the processor.

Images