CN110928649A - Resource scheduling method and device - Google Patents

Abstract

The application discloses a method and a device for scheduling resources. The method comprises the following steps: classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources. By the method and the device, the problems that in the related technology, the quantity of the resources distributed to each resource pool by the distributed system is fixed, and mutual utilization of the resources among the resource pools is difficult to realize according to the resource demand in the system operation process are solved.

Description

Resource scheduling method and device

Technical Field

The present application relates to the technical field of resource allocation of distributed systems, and in particular, to a method and an apparatus for resource scheduling.

Background

When the distributed system is queried, the client submits a query demand, the system allocates a resource pool for the query demand in the database, and the resource in the resource pool is utilized to execute query operation. Specifically, when a query is performed in Apache Hadoop, the client submits an SQL query task to a coordinator node of the Apache impala, the Apache impala analyzes the SQL query task and predicts a memory value required for executing the SQL query task, and then the SQL query task is put into a corresponding resource pool to be executed. If the resources in the resource pool are sufficient, allowing the SQL query task and the current SQL query task to be executed concurrently, wherein the resources required by the concurrent execution of all the SQL query tasks cannot exceed the total quantity of the resources in the resource pool; and if the resources in the resource pool are insufficient, the SQL query task is executed after the current SQL query task is executed to release the resources.

Currently, the resource scheduling of Apache impala belongs to a static scheduling mode, that is, each resource pool allocates a resource coefficient, usually a percentage, before the Apache impala cluster runs, so that some resource pools in a plurality of resource pools are in a busy state and some resource pools are in a idle state in a certain time period. Because the resource quantity in each resource pool is preset and is a fixed value, the resource pool in the busy state cannot acquire more resources to process the SQL query task waiting for queuing, and the resource pool in the idle state cannot transfer the idle resources to the resource pool in the busy state for utilization. The utilization rate of the whole cluster resources is reduced, the working efficiency of the Apache impala is influenced, and meanwhile the interactive experience of the system and the user is influenced.

Aiming at the problems that the quantity of resources distributed to each resource pool by a distributed system in the related technology is fixed and the mutual utilization of the resources among the resource pools is difficult to realize according to the resource demand in the system operation process, an effective solution is not provided at present.

Disclosure of Invention

The main purpose of the present application is to provide a method and an apparatus for resource scheduling, so as to solve the problems that the number of resources allocated to each resource pool by a distributed system in the related art is fixed, and it is difficult to implement mutual utilization of resources between resource pools according to the resource demand in the system operation process.

In order to achieve the above object, according to an aspect of the present application, a resource scheduling method is provided. The method comprises the following steps: classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources.

Further, before classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: acquiring historical data in a preset time interval, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than the preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool; and determining the quantity of the remaining resources in each resource pool according to the quantity of the resources requested by the query request received by each resource pool and the quantity of the available resources of each resource pool.

Further, after classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: under the condition that the quantity of resources requested by a query request received by a resource pool is greater than the quantity of available resources of the resource pool, calculating the quantity of the resources lacked in the resource pool; sequencing the resource pools in a resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first-class resource pool sequence; and sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

Further, determining a scheduling policy for the resource based on the type of each resource pool includes: according to the number of the deficient resources of the resource pool in the resource busy state, sequentially scheduling the resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the sequence order of the second type resource pool sequence, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequence is preferentially scheduled; in the process of scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state in sequence according to the sequence of the second type of resource pool sequences, detecting whether the number of the remaining resources in the resource pool in the resource busy state is greater than a preset threshold value; and under the condition that the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value, stopping scheduling the resources from the second type resource pool sequence.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the quantity of the remaining resources in the preset resource pool is greater than a preset value or not; under the condition that the number of the residual resources in the preset resource pool is greater than or equal to a preset value, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the quantity of the residual resources in the preset resource pool is less than a preset value, determining the preset resource pool as a resource pool in a resource busy state.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the proportion of the number of the remaining resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than a preset proportion or not; under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to a preset proportion, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion, determining the preset resource pool as a resource pool in a resource busy state.

Further, after determining the scheduling policy of the resource based on the type of each resource pool, the method further comprises: scheduling resources among all resource pools according to a scheduling strategy of the resources; and executing query operation according to the received query request in each resource pool after resource scheduling.

To achieve the above object, according to another aspect of the present application, there is provided a resource scheduling apparatus. The device includes: the classification unit is used for classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; a first determining unit, configured to determine a scheduling policy of the resource based on the type of each resource pool, where the scheduling policy is used to determine a scheduling flow direction of the resource and a scheduling number of the resource.

In order to achieve the above object, according to another aspect of the present application, there is provided a storage medium including a stored program, wherein the program performs any one of the above resource scheduling methods.

In order to achieve the above object, according to another aspect of the present application, there is provided a processor configured to execute a program, where the program executes any one of the above resource scheduling methods.

Through the application, the following steps are adopted: classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling quantity of the resources, and the problems that in the related technology, the quantity of the resources distributed to each resource pool by a distributed system is fixed, and the mutual utilization of the resources among the resource pools is difficult to realize according to the resource demand in the system operation process are solved. The resource scheduling strategy is determined by judging whether the resource pool is in a resource idle state or a resource pool in a resource busy state, so that the mutual utilization of resources among the resource pools is realized according to the resource demand in the system operation process, and the system operation efficiency is improved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the application and, together with the description, serve to explain the application and are not intended to limit the application. In the drawings:

fig. 1 is a flowchart of a resource scheduling method provided according to an embodiment of the present application; and

fig. 2 is a schematic diagram of a resource scheduling apparatus according to an embodiment of the present application.

Detailed Description

It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the application described herein may be used. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

For convenience of description, some terms or expressions referred to in the embodiments of the present application are explained below:

apache Hadoop: the method is a framework for running the application program on a large cluster constructed by hardware, and in the running process, a computing task is divided into small blocks (multiple times) to run on different nodes.

Apache Impala: is an open source, distributed, interactive database. The method and the Apache Hive share metadata, and near real-time SQL query can be provided for massive data stored based on Apache Hadoop.

Cloudera Manager: is a tool for simplifying Apache Hadoop management.

coordinator node: is the node in Apache Impala that is used to distribute the query request.

SQL query task: a query task implemented using a structured query language.

According to an embodiment of the present application, a resource scheduling method is provided.

Fig. 1 is a flowchart of a resource scheduling method according to an embodiment of the present application. As shown in fig. 1, the method comprises the steps of:

step S101, classifying a plurality of resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in the resource idle state and the resource pool in the resource busy state.

For example, when Apache Impala provides a real-time SQL query, the client submits an SQL query task to the Apache Impala, the Apache Impala corresponds to a plurality of resource pools, and the SQL query task executes the query by using resources in the corresponding resource pools. In the process of providing real-time SQL query by the Apache Impala, the query tasks received by each resource pool are different, the quantity of resources required by different query tasks is different, the resource occupation conditions in each resource pool in the same time period are different, in order to reasonably utilize the resources in each resource pool in the operation process of the Apache Impala, a plurality of resource pools of the Apache Impala are classified according to the resource occupation conditions in each resource pool, and the resource occupation conditions are measured by the quantity of the residual resources in the resource pools.

For example, Apache Impala corresponds to 5 resource pools, at the current moment, there are 3 SQL query tasks executed in parallel in the first resource pool, and there are 3 SQL query tasks waiting to be executed in queue, and there is no remaining resource in the first resource pool; the number of SQL query tasks executed in the second type resource pool is 2, the number of the resources left in the second type resource pool is 10M, and no query tasks waiting to be executed in a queue exist; 1 SQL query task is executed in the third resource pool, and the number of the resources left in the third resource pool is 500M; there are no SQL query tasks being executed in the fourth resource pool and the fifth resource pool, and no new query requests are currently received. The 5 resource pools corresponding to Apache Impala are classified, the first resource pool and the second resource pool can be divided into resource pools in a resource busy state, and the third resource pool, the fourth resource pool and the fifth resource pool can be divided into resource pools in a resource idle state.

Optionally, in the resource scheduling method provided in this embodiment of the present application, the classifying the plurality of resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types includes: judging whether the quantity of the remaining resources in the preset resource pool is greater than a preset value or not; under the condition that the number of the residual resources in the preset resource pool is greater than or equal to a preset value, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the quantity of the residual resources in the preset resource pool is less than a preset value, determining the preset resource pool as a resource pool in a resource busy state.

It should be noted that, because the total resource amount of different resource pools is different, the number of query tasks received by different resource pools is different, and the number of resources required by each query task is also different, in order to facilitate classification of resource pools and thus scheduling of resources, uniform preset values can be set for a plurality of resource pools of an Apache Impala, and the preset values are used as criteria for resource pool classification.

For example, different Apache Impala face different client groups, the types of the received query tasks are different, if the number of resources required by the minimum query task that can be received by the target Apache Impala is 50M, the number of resources required by 10 minimum query tasks can be set as a preset value, the preset value is 500M, it is detected whether the number of remaining resources in each resource pool of the target Apache Impala at the current time is less than 500M, the resource pool with the number of remaining resources less than 500M in the resource pool is determined as a resource pool in a resource busy state, and the resource pool with the number of remaining resources greater than or equal to 500M in the resource pool is determined as a resource pool in a resource idle state.

For another example, because the total resource amounts of different resource pools of the same Apache Impala are different, half of the total resource amount of the minimum resource pool may be used as a preset value, if the total resource amount allocated to the minimum resource pool by the target Apache Impala is 800M, the preset value is 400M, it is detected whether the amount of the remaining resources in each resource pool of the target Apache Impala at the current time is less than 400M, the resource pool whose amount of the remaining resources in the resource pool is less than 400M is determined as the resource pool in the resource busy state, and the resource pool whose amount of the remaining resources in the resource pool is greater than or equal to 400M is determined as the resource pool in the resource idle state.

Through the embodiment, the unified preset value is set to serve as the standard for classifying the plurality of resource pools, the states of whether the resources of the resource pools are busy or not are simply and conveniently distinguished, specifically, the preset value can be set according to different query tasks received by a database of a distributed system, and the preset value can also be set according to different total amounts of the resources in the resource pools, and the set standard of the preset value is not limited in the application.

Optionally, in the resource scheduling method provided in this embodiment of the present application, the classifying the plurality of resource pools based on the number of remaining resources in the resource pools to obtain different types of resource pools includes: judging whether the proportion of the number of the remaining resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than a preset proportion or not; under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to a preset proportion, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion, determining the preset resource pool as a resource pool in a resource busy state.

It should be noted that, if the total number of resources in the first resource pool in the distributed system is 10G, the number of resources required for the received query task is large, the total number of resources in the second type of resource pool is 500M, the number of resources required for the received query task is large, and if a uniform value is set to distinguish whether the resources in the resource pool are busy, for example, 300M, and in the case that the number of remaining resources in the first resource pool is less than 300M, the first resource pool is divided into resource pools in a busy state, so that the status of whether the resources in the first resource pool are busy can be properly represented; however, for the second type of resource pool, in the case that the amount of the remaining resources in the resource pool is less than 300M, some query requests with smaller resource demand can still be processed, and it is not appropriate to divide the query requests into the resource pool in a busy state. Under the condition that the difference of the total quantity of the resources in each resource pool is large, the proportion of the quantity of the remaining resources in the resource pool to the total quantity of the resources in the resource pool can be used as the standard for classifying the resource pools.

For example, the ratio of the number of the remaining resources in the resource pool to the total number of the resources in the resource pool is 30% as a standard for classifying the resource pool, and for a first resource pool with the total number of the resources of 10G, the resource pool is determined to be in a resource busy state when the number of the remaining resources in the resource pool is less than 3G, and the resource pool is determined to be in a resource idle state when the number of the remaining resources in the resource pool is greater than or equal to 3G; for a first resource pool with the total resource amount of 500M, determining the resource pool as a resource pool in a resource busy state when the number of the remaining resources in the resource pool is less than 150M, and determining the resource pool as a resource pool in a resource idle state when the number of the remaining resources in the resource pool is greater than or equal to 150M.

According to the embodiment, under the condition that the difference of the total quantity of the resources in each resource pool is large, the proportion of the quantity of the residual resources in the resource pool to the total quantity of the resources in the resource pool is used as the standard for classifying the resource pools, so that the accuracy of distinguishing whether the resources in each resource pool are busy or not is improved.

Step S102, determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources.

For example, according to the number of idle resources in the resource pool in the resource idle state and the number of resource pool shortage resources in the resource pool in the resource busy state, the idle resources in the resource pool in the resource idle state are scheduled to the resource pool in the resource busy state, so as to improve the resource utilization rate of the resource pool in the resource idle state, and improve the query efficiency of the query tasks in the resource pool in the resource busy state, for example, the number of remaining resources in the first resource pool is 900M, the number of remaining resources in the second type resource pool is 500M, no remaining resources in the third resource pool, 2 query tasks waiting to be queried in line exist, the first query task waiting in line needs 300M resources, the second query task waiting in line needs 200M resources, no remaining resources in the fourth resource pool, there is one query task waiting to be queried, and 400M resources are needed, at least 500M resources can be scheduled for the third resource pool from the first resource pool, and at least 400M resources can be scheduled for the fourth resource pool from the second resource pool, so that the resource requirement of the query task in the resource pool in the resource busy state is met.

The resource scheduling method provided by the embodiment of the application classifies the plurality of resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling quantity of the resources, and the problems that in the related technology, the quantity of the resources distributed to each resource pool by a distributed system is fixed, and the mutual utilization of the resources among the resource pools is difficult to realize according to the resource demand in the system operation process are solved. The resource scheduling strategy is determined by judging whether the resource pool is in a resource idle state or a resource pool in a resource busy state, so that the mutual utilization of resources among the resource pools is realized according to the resource demand in the system operation process, and the system operation efficiency is improved.

In order to determine the number of remaining resources in each current resource pool, optionally, in the resource scheduling method provided in the embodiment of the present application, before classifying the plurality of resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: acquiring historical data in a preset time interval, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than the preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool; and determining the quantity of the remaining resources in each resource pool according to the quantity of the resources requested by the query request received by each resource pool and the quantity of the available resources of each resource pool.

For example, the preset length may be 5 minutes, the preset time interval may be the previous 5 minutes of the current time, the preset length may be 10 minutes, the preset time interval may be the previous 10 minutes of the current time, because a certain time is needed for processing the query request, the historical data in the previous period of time at the current moment can represent whether the resources in each resource pool at the current moment are busy or not, the historical data may be data generated in the process that the Apache Impala receives and executes the query task within a preset time interval, may include the amount of resources requested by the SQL query request, and in particular, after receiving the SQL query task from Apache Impala, the predicted required memory value may further include the amount of available resources in each resource pool, that is, the amount of resources occupied by the currently executed SQL query task is excluded, and the resources that can be provided in each resource pool are also included. If a new query task to be executed is not received except the SQL query task being executed in one resource pool, the quantity of the available resources of the resource pool is the quantity of the remaining resources in the resource pool; if a new query task to be executed is received in one resource pool except the SQL query task which is being executed, the quantity of the remaining resources in the resource pool is the difference value between the quantity of the available resources in the resource pool and the quantity of the resources required by the new query task to be executed; if the executed SQL query task in one resource pool is about to be executed, the quantity of the remaining resources in the resource pool is the sum of the quantity of the available resources in the resource pool and the quantity of the resources occupied by the executed query task.

For example, data generated in the process that an Apache Impala receives and executes a query task within 5 minutes before the current time is collected, 2 SQL query tasks executed in parallel exist in the first resource pool, no new query task to be executed is received within 5 minutes, 500M idle resources exist in the first resource pool after resource support is provided for the query task being executed, which indicates that the available resources of the first resource pool are 500M, and the number of the remaining resources in the first resource pool is also 500M; the method comprises the steps that 2 SQL query tasks executed in parallel exist in a second resource pool, available resources of the second resource pool are 600M, the second resource pool receives a new query task to be executed, the number of resources required by the new query task is 400M, and the number of the remaining resources in a first resource pool is 200M; in addition to 3 SQL query tasks executed in parallel and 2 query tasks queued to be executed exist in the third resource pool, so that no residual resource exists in the third resource pool; an executed SQL query task exists in the fourth resource pool, the available resource of the fourth resource pool is 800M, the query task occupies 200M of resources, and the quantity of the remaining resources in the fourth resource pool is 1000M after the execution is about to be completed.

By the embodiment, the number of resources requested by the query request received by each resource pool in the historical time period before the current time and the number of available resources in each resource pool are analyzed, so that the resource occupation condition in each current resource pool can be accurately known, and the number of the remaining resources in each resource pool can be known.

Optionally, in the resource scheduling method provided in this embodiment of the present application, after classifying the plurality of resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: under the condition that the quantity of resources requested by a query request received by a resource pool is greater than the quantity of available resources of the resource pool, calculating the quantity of the resources lacked in the resource pool; sequencing the resource pools in a resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first-class resource pool sequence; and sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

For example, the first resource pool and the second resource pool are resource pools in a resource busy state, in addition to the query task being executed, there are 2 query tasks queued for execution in the first resource pool, the available resources in the first resource pool are 200M, and the number of resources requested by the 2 query tasks is 300M and 400M, respectively, which indicates that there are no remaining resources in the first resource pool, and the number of shortage resources in the first resource pool is 500M; besides the query tasks being executed, there are 1 query task waiting to be executed in the second type resource pool, the available resources in the second type resource pool are 100M, and the number of resources requested by the query tasks waiting to be executed is 500M respectively, which indicates that there are no remaining resources in the second type resource pool, and the number of shortage resources in the second type resource pool is 400M. Sequencing according to the number of the deficient resources in the resource pool to obtain a first resource pool sequence as follows: a first resource pool and a second type resource pool.

The third resource pool, the fourth resource pool and the fifth resource pool are resource pools in a resource idle state, and the number of the remaining resources in the third resource pool is 800M; the number of the remaining resources in the fourth resource pool is 900M; the number of remaining resources in the fifth resource pool is 500M. Sequencing the number of the remaining resources in the resource pool to obtain a second type resource pool sequence as follows: a fourth resource pool, a third resource pool, and a fifth resource pool.

Optionally, in the resource scheduling method provided in the embodiment of the present application, determining the scheduling policy of the resource based on the type of each resource pool includes: according to the number of the deficient resources of the resource pool in the resource busy state, sequentially scheduling the resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the sequence order of the second type resource pool sequence, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequence is preferentially scheduled; in the process of scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state in sequence according to the sequence of the second type of resource pool sequences, detecting whether the number of the remaining resources in the resource pool in the resource busy state is greater than a preset threshold value; and under the condition that the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value, stopping scheduling the resources from the second type resource pool sequence.

For example, within a first type of resource pool sequence: the available resources in the first resource pool are 200M, the quantity of the resources requested by 2 tasks waiting for query is 300M and 400M respectively, and the quantity of the shortage resources is 500M; the available resources in the second type of resource pool are 100M, the number of resources requested by the task waiting for query is 500M, which indicates that no residual resources exist in the second type of resource pool, and the number of the shortage resources is 400M.

Within the second type of resource pool sequence: the number of the remaining resources in the fourth resource pool is 900M, the number of the remaining resources in the third resource pool is 800M, and the number of the remaining resources in the fifth resource pool is 500M.

The scheduling strategy specifically comprises: the resource scheduling is performed from the fourth resource pool to the first resource pool, the scheduling 500M can minimally satisfy the number of resources required for executing the query task in the first resource pool, and in order to prevent an error from occurring when the system analyzes the number of resources required for the query task, resources exceeding 500M can be scheduled for the first resource pool to ensure that the query task in the first resource pool is normally executed after the resource scheduling, for example, the preset threshold value can be set to 100M, and 600M resources are scheduled for the first resource pool, that is, the requirement that the remaining number in the first resource pool is greater than the preset threshold value can be satisfied.

And scheduling resources for the second resource pool under the condition of ensuring that the query task in the first resource pool can be normally executed. In the second type of resource pool sequence, the number of the remaining resources in the fourth resource pool is 300M, the number of the remaining resources in the third resource pool is 800M, and the number of the remaining resources in the fifth resource pool is 500M. The number of the shortage resources of the second type resource pool is 400M, 500M resources can be scheduled for the second type resource pool, the number of the remaining resources in the fourth type resource pool is 300M and is less than 400M, and the fourth type resource pool needs to reserve part of resources to ensure that the executing query task does not have errors, 100M resources can be reserved for the fourth type resource pool, 200M resources are scheduled from the fourth type resource pool to the second type resource pool, and 300M resources are scheduled from the third type resource pool to the second type resource pool.

According to the method, the scheduling strategy is formulated, the resources are scheduled from the resource pools with more residual resources, the reasonable utilization of the resources among the resource pools is achieved, the resource requirements of the resource pools with more serious resource shortage conditions are met preferentially, the resource requirements of the resource pools with the second time resource shortage conditions are met, the execution efficiency of the query task is improved, in addition, partial resources are reserved for each resource pool in the resource scheduling process, and the normal operation of the task under the condition that the resource requirements of the query task have small-range errors is guaranteed.

Optionally, in the resource scheduling method provided in the embodiment of the present application, after determining the scheduling policy of the resource based on the type of each resource pool, the method further includes: scheduling resources among all resource pools according to a scheduling strategy of the resources; and executing query operation according to the received query request in each resource pool after resource scheduling.

For example, in Apache Impala, resource scheduling may be completed by using a Cloudera Manager as a management tool, and specifically, according to a set scheduling policy, resource scheduling between resource pools may be implemented by transferring resources in a resource pool in a resource idle state from an API interface of the Cloudera Manager into a resource pool in a resource busy state. After resources are scheduled for the resource pool in the resource busy state, the resource pool in the resource idle state continues to execute the existing query task or waits for the query task, the resources are not in shortage in the resource pool in the resource busy state originally, the query task to be executed can complete the query by using the scheduled resources, the resources are not required to be released by the current query task, the query requirement of a client side is quickly responded, the execution efficiency of the system on the query task is improved, and the user experience is improved.

It should be noted that, in the resource scheduling method provided in the embodiment of the present application, after the query operation is performed in each resource pool according to the received query request, the classification of the plurality of resource pools is updated according to the number of remaining resources in the resource pool according to the preset time period, and the scheduling policy of the resource is adjusted based on the type of each resource pool after the update.

For example, as each resource pool receives the query task, occupies the resource to execute the query task, and releases the resource after the query task is executed, the state of whether the resource in each resource pool is busy or not is dynamically changed, a preset time period can be set to 5 minutes, the number of the remaining resources in each resource pool is detected every 5 minutes, each resource pool is reclassified according to the number of the remaining resources, the scheduling policy of the resource is adjusted based on a new classification result, and the adjusted resource scheduling policy is executed.

According to the embodiment, the resource scheduling strategy is adjusted according to the dynamic change of the residual resources in each resource pool, the utilization rate of the resources is continuously improved in the system operation process, and therefore the execution efficiency of the query task is continuously improved.

It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowcharts, in some cases, the steps illustrated or described may be performed in an order different than presented herein.

The embodiment of the present application further provides a resource scheduling apparatus, and it should be noted that the resource scheduling apparatus in the embodiment of the present application may be used to execute the method for resource scheduling provided in the embodiment of the present application. The following describes a resource scheduling apparatus provided in an embodiment of the present application.

Fig. 2 is a schematic diagram of a resource scheduling apparatus according to an embodiment of the present application. As shown in fig. 2, the apparatus includes: a classification unit 10 and a first determination unit 20.

Specifically, the classifying unit 10 is configured to classify the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, where the types of the resource pools include at least one of the following: the resource pool in a resource idle state and the resource pool in a resource busy state;

a first determining unit 20, configured to determine a scheduling policy of the resource based on the type of each resource pool, where the scheduling policy is used to determine a scheduling flow direction of the resource and a scheduling number of the resource.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the apparatus further includes: the acquisition unit is used for classifying the resource pools based on the number of the residual resources in the resource pools, and acquiring historical data in a preset time interval before the resource pools of different types are obtained, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than the preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool; and a second determining unit, configured to determine the number of remaining resources in each resource pool according to the number of resources requested by the query request received by each resource pool and the number of available resources of each resource pool.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the apparatus further includes: a third determining unit, configured to calculate, after classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the number of resources that are insufficient in the resource pools when the number of resources requested by the query request received by the resource pools is greater than the number of available resources in the resource pools; the first sequencing unit is used for sequencing the resource pools in the resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first type of resource pool sequence; and the second sequencing unit is used for sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the first determining unit 20 includes: the scheduling module is used for sequentially scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the number of the deficient resources of the resource pool in the resource busy state and the sequence order of the second type resource pool sequences, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequences is scheduled preferentially; the detection module is used for detecting whether the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value or not in the process of sequentially scheduling the resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the sequence of the second type of resource pool sequence; and the execution stopping module is used for stopping scheduling the resources from the second type resource pool sequence under the condition that the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the classifying unit 10 includes: the first judgment module is used for judging whether the quantity of the residual resources in the preset resource pool is greater than a preset value or not; the first determining module is used for determining the preset resource pool as a resource pool in a resource idle state under the condition that the number of the residual resources in the preset resource pool is greater than or equal to a preset value; and the second determining module is used for determining the preset resource pool as the resource pool in the resource busy state under the condition that the quantity of the residual resources in the preset resource pool is less than the preset value.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the classifying unit 10 includes: the second judgment module is used for judging whether the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than the preset proportion or not; the third determining module is used for determining the preset resource pool as a resource pool in a resource idle state under the condition that the ratio of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to a preset ratio; and the fourth determining module is used for determining the preset resource pool as the resource pool in the resource busy state under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion.

Optionally, in the resource scheduling apparatus provided in the embodiment of the present application, the apparatus further includes: and the scheduling unit is used for scheduling the resources among the resource pools according to the scheduling strategies of the resources after the scheduling strategies of the resources are determined based on the types of the resource pools.

And the execution unit is used for executing the query operation in each resource pool after the resource scheduling according to the received query request.

The resource scheduling apparatus provided in the embodiment of the present application classifies a plurality of resource pools by the classification unit 10 based on the number of remaining resources in the resource pools to obtain resource pools of different types, where the types of the resource pools include at least one of the following: the resource pool in a resource idle state and the resource pool in a resource busy state; the first determining unit 20 determines a scheduling policy of the resource based on the type of each resource pool, where the scheduling policy is used to determine a scheduling flow direction of the resource and a scheduling number of the resource, so as to solve the problem that the number of the resources allocated to each resource pool by the distributed system in the related art is fixed, and it is difficult to implement mutual utilization of the resources between the resource pools according to the resource demand during the system operation process.

The resource scheduling device comprises a processor and a memory, wherein the classification unit 10, the first determination unit 20 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.

The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and the problem that the quantity of resources distributed to each resource pool by a distributed system in the related technology is fixed and the mutual utilization of the resources among the resource pools is difficult to realize according to the resource demand in the system operation process is solved by adjusting the kernel parameters.

The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.

An embodiment of the present invention provides a storage medium, on which a program is stored, which, when executed by a processor, implements the resource scheduling method.

The embodiment of the invention provides a processor, which is used for running a program, wherein the resource scheduling method is executed when the program runs.

The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps: classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources.

Further, before classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: acquiring historical data in a preset time interval, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than the preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool; and determining the quantity of the remaining resources in each resource pool according to the quantity of the resources requested by the query request received by each resource pool and the quantity of the available resources of each resource pool.

Further, after classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: under the condition that the quantity of resources requested by a query request received by a resource pool is greater than the quantity of available resources of the resource pool, calculating the quantity of the resources lacked in the resource pool; sequencing the resource pools in a resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first-class resource pool sequence; and sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

Further, determining a scheduling policy for the resource based on the type of each resource pool includes: according to the number of the deficient resources of the resource pool in the resource busy state, sequentially scheduling the resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the sequence order of the second type resource pool sequence, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequence is preferentially scheduled; in the process of scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state in sequence according to the sequence of the second type of resource pool sequences, detecting whether the number of the remaining resources in the resource pool in the resource busy state is greater than a preset threshold value; and under the condition that the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value, stopping scheduling the resources from the second type resource pool sequence.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the quantity of the remaining resources in the preset resource pool is greater than a preset value or not; under the condition that the number of the residual resources in the preset resource pool is greater than or equal to a preset value, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the quantity of the residual resources in the preset resource pool is less than a preset value, determining the preset resource pool as a resource pool in a resource busy state.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the proportion of the number of the remaining resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than a preset proportion or not; under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to a preset proportion, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion, determining the preset resource pool as a resource pool in a resource busy state.

Further, after determining the scheduling policy of the resource based on the type of each resource pool, the method further comprises: scheduling resources among all resource pools according to a scheduling strategy of the resources; and executing query operation according to the received query request in each resource pool after resource scheduling. The device herein may be a server, a PC, a PAD, a mobile phone, etc.

The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device: classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state; and determining a scheduling strategy of the resources based on the type of each resource pool, wherein the scheduling strategy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources.

Further, before classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: acquiring historical data in a preset time interval, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than the preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool; and determining the quantity of the remaining resources in each resource pool according to the quantity of the resources requested by the query request received by each resource pool and the quantity of the available resources of each resource pool.

Further, after classifying the resource pools based on the number of remaining resources in the resource pools to obtain resource pools of different types, the method further includes: under the condition that the quantity of resources requested by a query request received by a resource pool is greater than the quantity of available resources of the resource pool, calculating the quantity of the resources lacked in the resource pool; sequencing the resource pools in a resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first-class resource pool sequence; and sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

Further, determining a scheduling policy for the resource based on the type of each resource pool includes: according to the number of the deficient resources of the resource pool in the resource busy state, sequentially scheduling the resources from the resource pool in the resource idle state to the resource pool in the resource busy state according to the sequence order of the second type resource pool sequence, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequence is preferentially scheduled; in the process of scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state in sequence according to the sequence of the second type of resource pool sequences, detecting whether the number of the remaining resources in the resource pool in the resource busy state is greater than a preset threshold value; and under the condition that the quantity of the residual resources in the resource pool in the resource busy state is greater than a preset threshold value, stopping scheduling the resources from the second type resource pool sequence.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the quantity of the remaining resources in the preset resource pool is greater than a preset value or not; under the condition that the number of the residual resources in the preset resource pool is greater than or equal to a preset value, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the quantity of the residual resources in the preset resource pool is less than a preset value, determining the preset resource pool as a resource pool in a resource busy state.

Further, classifying the plurality of resource pools based on the number of remaining resources in the resource pools, and obtaining the resource pools of different types includes: judging whether the proportion of the number of the remaining resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than a preset proportion or not; under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to a preset proportion, determining the preset resource pool as a resource pool in a resource idle state; and under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion, determining the preset resource pool as a resource pool in a resource busy state.

Further, after determining the scheduling policy of the resource based on the type of each resource pool, the method further comprises: scheduling resources among all resource pools according to a scheduling strategy of the resources; and executing query operation according to the received query request in each resource pool after resource scheduling.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims (10)

1. A method for scheduling resources, comprising:

classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state;

and determining a scheduling policy of the resources based on the type of each resource pool, wherein the scheduling policy is used for determining the scheduling flow direction of the resources and the scheduling number of the resources.

2. The method of claim 1, wherein before classifying the plurality of resource pools based on the number of resources remaining in the resource pools for different types of resource pools, the method further comprises:

acquiring historical data in a preset time interval, wherein the preset time interval takes the current moment as an interval end point, the interval length of the preset time interval is not more than a preset length, and the historical data at least comprises: the number of resources requested by the query request received by each resource pool, and the number of available resources of each resource pool;

and determining the quantity of the remaining resources in each resource pool according to the quantity of the resources requested by the query request received by each resource pool and the quantity of the available resources of each resource pool.

3. The method of claim 2, wherein after classifying the plurality of resource pools based on the number of resources remaining in the resource pools for different types of resource pools, the method further comprises:

under the condition that the quantity of resources requested by a query request received by a resource pool is greater than the quantity of available resources of the resource pool, calculating the quantity of the resources lacked in the resource pool;

sequencing the resource pools in the resource busy state according to the quantity of the deficient resources in the resource pools to obtain a first class of resource pool sequence;

and sequencing the resource pools in the resource idle state according to the number of the residual resources in the resource pools to obtain a second type of resource pool sequence.

4. The method of claim 3, wherein determining the scheduling policy for the resource based on the type of each resource pool comprises:

according to the number of the deficient resources of the resource pool in the resource busy state, sequentially scheduling resources from the resource pool in the resource free state to the resource pool in the resource busy state according to the sequence order of the second type resource pool sequence, wherein the resource pool with the largest number of the residual resources in the second type resource pool sequence is scheduled preferentially;

in the process of scheduling resources from the resource pool in the resource idle state to the resource pool in the resource busy state in sequence according to the sequence order of the second-class resource pool sequence, detecting whether the number of the remaining resources in the resource pool in the resource busy state is greater than a preset threshold value;

and stopping scheduling resources from the second type resource pool sequence under the condition that the number of the remaining resources in the resource pool in the resource busy state is greater than the preset threshold value.

5. The method of claim 1, wherein classifying the plurality of resource pools based on the number of resources remaining in the resource pools to obtain different types of resource pools comprises:

judging whether the quantity of the remaining resources in the preset resource pool is greater than a preset value or not;

determining the preset resource pool as a resource pool in a resource idle state under the condition that the number of the residual resources in the preset resource pool is greater than or equal to the preset value;

and under the condition that the quantity of the residual resources in the preset resource pool is less than the preset value, determining that the preset resource pool is in a resource busy state.

6. The method of claim 1, wherein classifying the plurality of resource pools based on the number of resources remaining in the resource pools to obtain different types of resource pools comprises:

judging whether the proportion of the number of the remaining resources in a preset resource pool to the total number of the resources in the preset resource pool is greater than a preset proportion or not;

under the condition that the proportion of the number of the remaining resources in the preset resource pool to the total number of the resources in the preset resource pool is greater than or equal to the preset proportion, determining that the preset resource pool is a resource pool in a resource idle state;

and under the condition that the proportion of the number of the residual resources in the preset resource pool to the total number of the resources in the preset resource pool is smaller than the preset proportion, determining that the preset resource pool is a resource pool in a resource busy state.

7. The method of claim 1, wherein after determining the scheduling policy for the resource based on the type of each resource pool, the method further comprises:

scheduling resources among all resource pools according to the scheduling strategy of the resources;

and executing query operation according to the received query request in each resource pool after resource scheduling.

8. A resource scheduling apparatus, comprising:

the classification unit is used for classifying the resource pools based on the number of the remaining resources in the resource pools to obtain resource pools of different types, wherein the types of the resource pools include at least one of the following types: the resource pool in a resource idle state and the resource pool in a resource busy state;

a first determining unit, configured to determine a scheduling policy of the resource based on a type of each resource pool, where the scheduling policy is used to determine a scheduling flow direction of the resource and a scheduling number of the resource.

9. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program performs the resource scheduling method of any one of claims 1 to 7.

10. A processor, configured to run a program, wherein the program when running performs the resource scheduling method of any one of claims 1 to 7.

Images