CN111796934B - Task issuing method and device, storage medium and electronic equipment - Google Patents
Task issuing method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111796934B CN111796934B CN202010601477.0A CN202010601477A CN111796934B CN 111796934 B CN111796934 B CN 111796934B CN 202010601477 A CN202010601477 A CN 202010601477A CN 111796934 B CN111796934 B CN 111796934B
- Authority
- CN
- China
- Prior art keywords
- resource
- target
- quota
- task
- platform
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000004590 computer program Methods 0.000 claims description 2
- 238000012384 transportation and delivery Methods 0.000 claims description 2
- 238000013468 resource allocation Methods 0.000 abstract description 7
- 238000007726 management method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000002716 delivery method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure relates to a task issuing method, a device, a storage medium and electronic equipment, wherein the method comprises the following steps: after receiving a resource request of a target task, determining the resource occupation amount, the name space, the priority and the resource model corresponding to the target task, wherein the target resource type is the equipment model of the computing resource of a resource platform required for executing the target task; determining a first quota limit for the namespace for the target task and a second quota limit for the resource platform for the target task; and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result. The quota limit can be carried out on the occupation of the computing tasks with different priorities and different resource models on the total amount of the resources allocated to the namespaces and the platform resources, the pertinence and the accuracy of the resource allocation are improved, and the utilization rate of the platform resources is improved.
Description
Technical Field
The disclosure relates to the field of computing resource management, and in particular relates to a task issuing method, a device, a storage medium and electronic equipment.
Background
With the development of artificial intelligence technology, the data volume and the calculation volume involved in the development and reference processes related to artificial intelligence are also increasing, and a single resource device cannot meet the large calculation volume required by artificial intelligence calculation. Therefore, a resource platform based on a cloud computing platform needs to be built for artificial intelligent computing. When the user scale and the resource scale of the resource platform are large, reasonable resource quota management is needed to meet the resource scheduling requirement in a multi-user multi-task scene under the condition that the whole resource platform is ensured to be stably available.
Disclosure of Invention
In order to overcome the problems in the related art, the present disclosure provides a task issuing method, a task issuing device, a storage medium, and an electronic device.
According to a first aspect of an embodiment of the present disclosure, there is provided a task issuing method applied to a cloud computing platform, the method including:
under the condition that a resource request of a target task is received, determining a resource occupation amount, a target name space, a target priority and a target resource model corresponding to the target task, wherein the target name space is a name space (namespace) where a service end of the target task is sent out, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task;
determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; wherein the first quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and the target resource has the target resource model;
and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result.
Optionally, before determining the target namespace, the target priority and the target resource model corresponding to the target task in the case of receiving the resource request of the target task, the method further includes:
identifying device information of each computing resource in the resource platform, so as to determine the device model of each computing resource according to the device information;
and outputting the equipment model of each computing resource so that the service end sets the target resource model of the target task according to the equipment model of each computing resource.
Optionally, the determining a first quota limit for the target task for the target namespace and a second quota limit for the target task for the resource platform includes:
acquiring a first resource quota table corresponding to the preset target name space and a second resource quota table corresponding to the resource platform; the first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task;
determining the first resource quota limit according to the target priority, the target resource model and the first resource quota table; the method comprises the steps of,
and determining the second resource quota limit according to the target priority, the target resource model and the second resource quota table.
Optionally, the comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result, includes:
determining whether the resource occupancy exceeds the first quota limit;
determining whether the resource occupancy exceeds the second quota limit if it is determined that the resource occupancy does not exceed the first quota limit;
under the condition that the occupied amount of the resources does not exceed the second quota limit, the target task is issued to the resource platform for execution; or,
and outputting information for indicating that the target task cannot be issued to the resource platform under the condition that the occupied amount of the resources exceeds the first quota limit or the occupied amount of the resources exceeds the second quota limit.
According to a second aspect of embodiments of the present disclosure, there is provided a task issuing device applied to a cloud computing platform, the device including:
the information determining module is configured to determine a resource occupation amount, a target name space, a target priority and a target resource model corresponding to a target task under the condition that a resource request of the target task is received, wherein the target name space is a name space where a service end of the target task is sent out, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task;
a quota determination module configured to determine a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; wherein the first quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and the target resource has the target resource model;
and the task issuing module is configured to compare the resource occupation amount with the first quota limit and the second quota limit respectively so as to determine whether to issue the target task to the resource platform according to a comparison result.
Optionally, the apparatus further includes:
a device identification module configured to identify device information for each computing resource in the resource platform to determine a device model for the each computing resource based on the device information;
the resource model output module is configured to output the equipment model of each computing resource so that the service end can set the target resource model of the target task according to the equipment model of each computing resource.
Optionally, the quota determination module is configured to:
acquiring a first resource quota table corresponding to the preset target name space and a second resource quota table corresponding to the resource platform; the first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task;
determining the first resource quota limit according to the target priority, the target resource model and the first resource quota table; the method comprises the steps of,
and determining the second resource quota limit according to the target priority, the target resource model and the second resource quota table.
Optionally, the task issuing module is configured to:
determining whether the resource occupancy exceeds the first quota limit;
determining whether the resource occupancy exceeds the second quota limit if it is determined that the resource occupancy does not exceed the first quota limit;
under the condition that the occupied amount of the resources does not exceed the second quota limit, the target task is issued to the resource platform for execution; or,
and outputting information for indicating that the target task cannot be issued to the resource platform under the condition that the occupied amount of the resources exceeds the first quota limit or the occupied amount of the resources exceeds the second quota limit.
According to a third aspect of embodiments of the present disclosure, there is provided a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the steps of the task issuing method provided by the first aspect of the present disclosure.
According to a fourth aspect of embodiments of the present disclosure, there is provided an electronic device, comprising: the task issuing device provided by the second aspect of the present disclosure.
According to the technical scheme provided by the embodiment of the disclosure, under the condition that a resource request of a target task is received, the resource occupation amount, the target name space, the target priority and the target resource model corresponding to the target task can be determined, wherein the target name space is the name space where a service end sending the target task is located, and the target resource type is the equipment model of the computing resource of a resource platform required for executing the target task; determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; the first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and is provided with the target resource model; and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result. The quota limit can be carried out on the occupation of the computing tasks with different priorities and different resource models on the resource quantity corresponding to the name space and the total resource quantity of the platform resources, the pertinence and the accuracy of the resource allocation are improved, and the utilization rate of the platform resources is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a flowchart illustrating a method of task delivery according to an exemplary embodiment;
FIG. 2 is a flow chart of another task delivery method according to the one shown in FIG. 1;
FIG. 3 is a flow chart of a method of determining a resource quota limit in accordance with the method shown in FIG. 1;
FIG. 4 is a flow chart of a task delivery method according to the one shown in FIG. 1;
FIG. 5 is a block diagram of a task issuing device according to an example embodiment;
FIG. 6 is a block diagram of another task issuing device according to the illustration of FIG. 5;
fig. 7 is a block diagram of an electronic device, according to an example embodiment.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
In the related art of resource scheduling, resource quota management for multiple users (or service ends) or groups of users is typically implemented through namespaces. Namespaces provide a way to logically partition and isolate resources of an entire computing platform. In particular, before a user issues a task, it is necessary to create separate namespaces for different groups of users, where the names of the same computing task instance objects are unique in the namespaces, and task instances in the namespaces will have the same control policies. In a multi-user scene, in order to avoid that individual users occupy excessive resources and influence the use of other users of the platform, the limitation of the allowed resource amount of each namespace is defined according to the actual service requirement of each user and the total resource amount of the resource platform through a resource quota concept.
However, in the above-mentioned resource quota management method, only the GPU (Graphics Processing Unit, graphics processor) computing resources, CPU (Central Processing Unit ) computing resources, memory resources and other resource types are distinguished, that is, all the resource devices having the same resource type are regarded as the same resource to perform quota management, and the difference of parameters and performances of graphics cards (GPU devices), memory banks or processors (CPU devices) of different models is not considered, so that the pertinence of performing quota limitation on the resource is poor, and the utilization rate of platform resources is further reduced. And only the quota limitation on the task is performed on the resource quantity of each name space, and the quota limitation on the actual resource quantity of the whole resource platform is not considered, so that the accuracy of resource allocation is lower, and the utilization rate of the platform resources is further reduced.
In this regard, the present disclosure proposes a task issuing method, which specifically includes:
before introducing the task issuing method provided by the present disclosure, first, a target application scenario related to each embodiment in the present disclosure is described, where the target application scenario includes a resource platform, the resource platform is a cloud computing platform, the cloud computing platform includes at least one scheduler provided with a resource scheduling system, and a plurality of resource servers, where the resource servers are cloud servers that have GPU computing resources, CPU computing resources, and memory resources and are capable of providing cloud data acquisition and cloud computing services for a received computing task.
Fig. 1 is a flowchart of a task issuing method according to an exemplary embodiment, as shown in fig. 1, applied to the cloud computing platform described in the application scenario, where the method includes the following steps:
in step 101, when a resource request of a target task is received, a resource occupation amount, a target namespace, a target priority, and a target resource model corresponding to the target task are determined.
The target name space is a name space where a service end sending the target task is located, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task.
Illustratively, the resource platform has a plurality of namespaces, each namespace corresponding to a property end group (or user group), wherein each business end can create tasks according to business requirements. After the resource request of the target task is acquired, four resource information including the resource occupation amount, the target name space, the target priority and the target resource model corresponding to the target task can be acquired from the resource request information. The target name space acquisition method comprises the following steps: and determining the service end group of the service end sending the target task, and determining the corresponding naming space of the service end group as the target naming space. The four pieces of information are resource information corresponding to the computing resources of the same resource type. The resource categories may include: GPU computing resources, memory resources, CPU computing resources, and the like. For example, the resource occupation amount may be a resource occupation amount of the target task with respect to the GPU computing resource, the target priority is used to represent a priority order of occupation of the GPU computing resource corresponding to the target task, the target resource model is a device model of the graphics card, or the resource occupation amount may be a resource occupation amount of the target task with respect to the CPU computing resource, the target priority is used to represent a priority order of occupation of the CPU computing resource corresponding to the target task, and the target resource model is a device model of the processor chip. In the embodiment of the disclosure, the task issuing method is described by taking the resource type of the computing resource as quota management of GPU computing resources.
In step 102, a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task are determined.
The first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and the target resource has the target resource model.
In step 103, the resource occupation amount is compared with the first quota limit and the second quota limit, so as to determine whether to issue the target task to the resource platform according to the comparison result.
Illustratively, for a computing resource, e.g., a GPU computing resource, the resource platform corresponds to a set of quota limiting policies, and the resource platform corresponds to a set of quota limiting policies associated with each of the namespaces. And under the condition that the resource requirement of the target task simultaneously meets the quota limiting policy of the resource platform and the quota limiting policy corresponding to the naming space.
In this embodiment, the expression "resource amount" and the "total amount of resources" and "occupation amount of resources" related thereto may represent the number of entity devices that calculate resources, or the capacity or calculation capability of the resource devices. For example, a task may take up 6 GPU computing resources, which may actually indicate that 6 graphics cards (with the same computing power) are required to support the task. Alternatively, the resource occupation amount of a task is 2 memory resources, which actually means that 2 capacity units of storage capacity are needed to support the operation of the task, and the capacity units may be MB (Mbyte), GB (Gigabyte) or larger.
In summary, according to the technical solution provided by the embodiments of the present disclosure, under a condition that a resource request of a target task is received, a resource occupation amount, a target namespace, a target priority, and a target resource model corresponding to the target task can be determined, where the target namespace is a namespace where a service end sending the target task is located, and the target resource type is a device model of a computing resource of a resource platform required for executing the target task; determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; the first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and is provided with the target resource model; and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result. The quota limit can be carried out on the occupation of the computing tasks with different priorities and different resource models on the resource quantity corresponding to the name space and the total resource quantity of the platform resources, the pertinence and the accuracy of the resource allocation are improved, and the utilization rate of the platform resources is improved.
FIG. 2 is a flow chart of another task delivery method according to the one shown in FIG. 1, as shown in FIG. 2, the method may further include, prior to the step 101:
in step 104, device information for each computing resource in the resource platform is identified to determine a device model for each computing resource based on the device information.
In step 105, the device model of each computing resource is output, so that the service end sets the target resource model of the target task according to the device model of each computing resource.
For example, the scheduling system of the resource platform may perform scheduling on all computing resources in the resource platform at intervals of a preset duration, for example, all the performance of the graphics cards, the size of the video memory, the production merchant, the production number, the sales number, and other device information. The production number of each display card can be directly obtained, and the display cards with the same production number are used as a resource model, or the equipment information can be clustered through a preset equipment identification model, so that the computing resources corresponding to the equipment information clustered into one type have the same equipment model. After each equipment model is output or reported to the service end, the service end can determine the target resource model corresponding to the target task according to the number of each equipment model in the resource platform, the computing capacity of each equipment and the computing capacity of the target task. It should be noted that, the setting process of the target resource model (including the target priority as described above) may include: the method is realized through manual setting of operators at the service end, or the automatic division of the resource model can be performed through the service end by a preset classification algorithm based on the number of each equipment model in the resource platform, the computing capacity of each equipment and the computing capacity of the target task.
FIG. 3 is a flow chart of a method of determining a resource quota limit according to the one shown in FIG. 1, as shown in FIG. 3, the step 102 may include:
in step 1021, a first resource quota table corresponding to the target namespace and a second resource quota table corresponding to the resource platform, which are preset, are obtained.
The first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task.
In step 1022, the first resource quota limit is determined according to the target priority, the target resource model, and the first resource quota table.
In step 1023, the second resource quota limit is determined based on the target priority, the target resource model, and the second resource quota table.
Illustratively, the first resource quota table and the second resource quota table may each be stored in the scheduling system of the resource platform in the form of the following table 1, where the resource platform corresponds to one resource quota table and each namespace corresponds to one namespace resource quota table. In the process of initializing the resource platform or in the process of restarting the resource platform after updating, the resource scheduling system of the resource platform can count and output the number of each computing resource in the resource platform, and an operator can set the second resource quota table according to the number of each computing resource. And as each namespace is created, the operator can set the resource quota table corresponding to each namespace based on the business requirements and the number of each computing resource.
TABLE 1
| Resource quota restriction | Resource model a | Resource model b | Resource model c |
| Priority A | 3 | 4 | 2 |
| Priority B | 4 | 5 | 4 |
| Priority C | 3 | 6 | 6 |
| Priority D | 2 | 5 | 7 |
For example, after determining the target namespace, a plurality of target namespace corresponding resource quota tables may be determined from a plurality of namespace resource quota tables. Taking the table 1 as the first resource quota table (or the second resource quota table) as an example, if the target priority corresponding to the target task is the priority B and the target resource model is the resource model c, the resource quota of the target namespace (or the resource platform) for the target task is limited to 4.
It should be noted that, the total amount of the resource quota corresponding to each namespace may be adjusted according to the task amount or the resource occupation amount of the task actually received in a period of time, and specific values of the resource quota limits corresponding to different priorities in the resource quota table of each namespace.
For example, if it is detected that the task amount or the resource occupation amount corresponding to the namespace a is smaller and the task amount or the resource occupation amount corresponding to the namespace B is larger in one week, the total amount of the resource quota corresponding to the namespace a may be reduced at the beginning of the next week, and the total amount of the resource quota corresponding to the namespace B is increased, but the sum of the total amounts of the resource quota corresponding to each namespace is kept larger than the total amount of the resource platform. Or, for any namespace C, if it is detected that the task amount or the resource occupation amount of the task with the priority a is smaller and the task amount or the resource occupation amount of the task with the priority B is larger within one week, the total amount of the resource quota corresponding to the priority a may be reduced and the total amount of the resource quota corresponding to the priority B may be increased at the beginning of the next week, but the total amount of the resource quota of the namespace C is kept unchanged.
FIG. 4 is a flowchart of a task delivery method according to the one shown in FIG. 1, and as shown in FIG. 4, the step 103 may include: steps 1031-1032, or steps 1031 and 1034, or steps 1031, 1032 and 1034.
In step 1031, a determination is made as to whether the resource occupancy exceeds the first quota limit.
In step 1032, in the event that it is determined that the resource occupancy does not exceed the first quota limit, it is determined whether the resource occupancy exceeds the second quota limit.
In step 1033, the target task is issued to the resource platform for execution if it is determined that the resource occupancy does not exceed the second quota limit.
In step 1034, in the event that the resource occupancy exceeds the first quota limit or the resource occupancy exceeds the second quota limit, information is output indicating that the target task cannot be issued to the resource platform.
For example, considering that in the actual application process, the situation that all namespaces issue tasks to the resource platform at the same time is rarely occurred, a certain redundancy is generally considered when setting the resource quota corresponding to each namespace, that is, the sum of the total resource quota corresponding to each namespace may be set to be greater than the total resource of the resource platform. Therefore, the situation may occur that the resource occupation amount does not exceed the first quota limit but exceeds the second quota limit, and thus, in steps 1031 and 1032, the resource occupation amount needs to be compared with the first quota limit and the second quota limit sequentially. If both can meet the resource occupation amount (not exceeded), allowing the target task to be issued to the resource platform and executing through the computing resource improved by the resource platform; if any one of the two can not meet the occupation amount of the resource (exceeded), the issuing of the target task is refused, and information for indicating that the target task can not be issued to the resource platform is further output.
In summary, according to the technical solution provided by the embodiments of the present disclosure, under a condition that a resource request of a target task is received, a resource occupation amount, a target namespace, a target priority, and a target resource model corresponding to the target task can be determined, where the target namespace is a namespace where a service end sending the target task is located, and the target resource type is a device model of a computing resource of a resource platform required for executing the target task; determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; the first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and is provided with the target resource model; and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result. The quota limit can be carried out on the occupation of the computing tasks with different priorities and different resource models on the resource quantity corresponding to the name space and the total resource quantity of the platform resources, the pertinence and the accuracy of the resource allocation are improved, and the utilization rate of the platform resources is improved.
Fig. 5 is a block diagram of a task issuing device according to an exemplary embodiment, as shown in fig. 5, applied to the cloud computing platform described in the application scenario, where the device 500 includes:
the information determining module 510 is configured to determine, when a resource request of a target task is received, a resource occupation amount, a target namespace, a target priority and a target resource model corresponding to the target task, where the target namespace is a namespace where a service end sending the target task is located, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task;
a quota determination module 520 configured to determine a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; the first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and is provided with the target resource model;
the task issuing module 530 is configured to compare the resource occupation amount with the first quota limit and the second quota limit, so as to determine whether to issue the target task to the resource platform according to the comparison result.
Fig. 6 is a block diagram of another task issuing device according to fig. 5, and as shown in fig. 6, the device 500 may further include:
a device identification module 540 configured to identify device information of each computing resource in the resource platform, so as to determine a device model of each computing resource according to the device information;
the resource model output module 550 is configured to output the device model of each computing resource, so that the service end sets the target resource model of the target task according to the device model of each computing resource.
Optionally, the quota determination module 520 is configured to:
acquiring a first resource quota table corresponding to the preset target name space and a second resource quota table corresponding to the resource platform; the first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task;
determining the first resource quota limit according to the target priority, the target resource model and the first resource quota table; the method comprises the steps of,
and determining the second resource quota limit according to the target priority, the target resource model and the second resource quota table.
Optionally, the task issuing module 530 is configured to:
determining whether the resource occupancy exceeds the first quota limit;
determining whether the resource occupancy exceeds the second quota limit if it is determined that the resource occupancy does not exceed the first quota limit;
under the condition that the occupation amount of the resources does not exceed the second quota limit, the target task is issued to the resource platform for execution; or,
and outputting information for indicating that the target task cannot be issued to the resource platform under the condition that the resource occupation amount exceeds the first quota limit or the resource occupation amount exceeds the second quota limit.
In summary, according to the technical solution provided by the embodiments of the present disclosure, under a condition that a resource request of a target task is received, a resource occupation amount, a target namespace, a target priority, and a target resource model corresponding to the target task can be determined, where the target namespace is a namespace where a service end sending the target task is located, and the target resource type is a device model of a computing resource of a resource platform required for executing the target task; determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; the first quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for representing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and is provided with the target resource model; and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result. The quota limit can be carried out on the occupation of the computing tasks with different priorities and different resource models on the resource quantity corresponding to the name space and the total resource quantity of the platform resources, the pertinence and the accuracy of the resource allocation are improved, and the utilization rate of the platform resources is improved.
Fig. 7 is a block diagram of an electronic device, according to an example embodiment. For example, the electronic device 700 may be a scheduler or server in a cloud computing platform. Referring to fig. 7, electronic device 700 includes a processing component 722 that further includes one or more processors and memory resources represented by memory 732 for storing instructions, such as application programs, executable by processing component 722. The application programs stored in memory 732 may include one or more modules that each correspond to a set of instructions. Further, the processing component 722 is configured to execute instructions to perform the task issuing methods illustrated in fig. 1-4.
The electronic device 700 may also include a power supply component 726 configured to perform power management of the electronic device 700, a wired or wireless network interface 770 configured to connect the electronic device 700 to a network, and an input/output (I/O) interface 778. The electronic device 700 may operate based on memoryOperating system for memory 732, e.g., windows Server TM ，Mac OS X TM ，Unix TM ，Linux TM ，FreeBSD TM Or the like.
The electronic equipment provided by the embodiment of the disclosure can limit the occupation of the computing tasks with different priorities and different resource models to the resource quantity corresponding to the name space and the total resource quantity of the platform resources, thereby improving the pertinence and the accuracy of the resource allocation and the utilization rate of the platform resources.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure. This application is intended to cover any adaptations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. A task issuing method applied to a cloud computing platform, the method comprising:
under the condition that a resource request of a target task is received, determining a resource occupation amount, a target name space, a target priority and a target resource model corresponding to the target task, wherein the target name space is a name space where a service end of the target task is positioned, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task;
determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; wherein the first quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and the target resource has the target resource model;
and comparing the resource occupation amount with the first quota limit and the second quota limit respectively, so as to determine whether to issue the target task to the resource platform according to a comparison result.
2. The method according to claim 1, wherein before determining the target namespace, the target priority, and the target resource model corresponding to the target task in the case of receiving the resource request of the target task, the method further comprises:
identifying device information of each computing resource in the resource platform, so as to determine the device model of each computing resource according to the device information;
and outputting the equipment model of each computing resource so that the service end sets the target resource model of the target task according to the equipment model of each computing resource.
3. The method of claim 1, wherein the determining a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task comprises:
acquiring a first resource quota table corresponding to the preset target name space and a second resource quota table corresponding to the resource platform; the first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task;
determining the first resource quota limit according to the target priority, the target resource model and the first resource quota table; the method comprises the steps of,
and determining the second resource quota limit according to the target priority, the target resource model and the second resource quota table.
4. The method of claim 1, wherein comparing the resource occupancy with the first quota limit and the second quota limit, respectively, to determine whether to issue the target task to the resource platform based on the comparison results comprises:
determining whether the resource occupancy exceeds the first quota limit;
determining whether the resource occupancy exceeds the second quota limit if it is determined that the resource occupancy does not exceed the first quota limit;
under the condition that the occupied amount of the resources does not exceed the second quota limit, the target task is issued to the resource platform for execution; or,
and outputting information for indicating that the target task cannot be issued to the resource platform under the condition that the occupied amount of the resources exceeds the first quota limit or the occupied amount of the resources exceeds the second quota limit.
5. A task issuing device applied to a cloud computing platform, the device comprising:
the information determining module is configured to determine a resource occupation amount, a target name space, a target priority and a target resource model corresponding to a target task under the condition that a resource request of the target task is received, wherein the target name space is a name space where a service end of the target task is sent out, and the target resource model is a device model of a computing resource of a resource platform required for executing the target task;
a quota determination module configured to determine a first quota limit for the target namespace for the target task and a second quota limit for the resource platform for the target task; wherein the first quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in all target resources allocated for the target namespace, the second quota limit is used for characterizing a limit value of a resource amount of the target resource allowed to be occupied by the target task in the resource platform, the target resource corresponds to the target priority and the target resource has the target resource model;
and the task issuing module is configured to compare the resource occupation amount with the first quota limit and the second quota limit respectively so as to determine whether to issue the target task to the resource platform according to a comparison result.
6. The apparatus of claim 5, wherein the apparatus further comprises:
a device identification module configured to identify device information for each computing resource in the resource platform to determine a device model for the each computing resource based on the device information;
the resource model output module is configured to output the equipment model of each computing resource so that the service end can set the target resource model of the target task according to the equipment model of each computing resource.
7. The apparatus of claim 5, wherein the quota determination module is configured to:
acquiring a first resource quota table corresponding to the preset target name space and a second resource quota table corresponding to the resource platform; the first resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the target namespace for each task, which are sent by all service ends in the target namespace, and the second resource quota table is used for representing the corresponding relation among the priority, the resource model and the resource quota limit of the resource platform for each task;
determining the first resource quota limit according to the target priority, the target resource model and the first resource quota table; the method comprises the steps of,
and determining the second resource quota limit according to the target priority, the target resource model and the second resource quota table.
8. The apparatus of claim 5, wherein the task delivery module is configured to:
determining whether the resource occupancy exceeds the first quota limit;
determining whether the resource occupancy exceeds the second quota limit if it is determined that the resource occupancy does not exceed the first quota limit;
under the condition that the occupied amount of the resources does not exceed the second quota limit, the target task is issued to the resource platform for execution; or,
and outputting information for indicating that the target task cannot be issued to the resource platform under the condition that the occupied amount of the resources exceeds the first quota limit or the occupied amount of the resources exceeds the second quota limit.
9. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the steps of the method of any of claims 1-4.
10. An electronic device, comprising: task issuing device according to any of claims 5-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010601477.0A CN111796934B (en) | 2020-06-28 | 2020-06-28 | Task issuing method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010601477.0A CN111796934B (en) | 2020-06-28 | 2020-06-28 | Task issuing method and device, storage medium and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111796934A CN111796934A (en) | 2020-10-20 |
| CN111796934B true CN111796934B (en) | 2023-11-21 |
Family
ID=72804648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010601477.0A Active CN111796934B (en) | 2020-06-28 | 2020-06-28 | Task issuing method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111796934B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801430B (en) * | 2021-04-13 | 2021-11-12 | 贝壳找房(北京)科技有限公司 | Task issuing method and device, electronic equipment and readable storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107040479A (en) * | 2016-02-04 | 2017-08-11 | 华为软件技术有限公司 | A kind of method and apparatus of cloud computing resources regulation |
| CN109542603A (en) * | 2018-11-22 | 2019-03-29 | 北京航空航天大学 | It is a kind of improve different priorities task between isolation multi dimensional resource shielding system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109726005B (en) * | 2017-10-27 | 2023-02-28 | 伊姆西Ip控股有限责任公司 | Method, server system and computer readable medium for managing resources |
-
2020
- 2020-06-28 CN CN202010601477.0A patent/CN111796934B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107040479A (en) * | 2016-02-04 | 2017-08-11 | 华为软件技术有限公司 | A kind of method and apparatus of cloud computing resources regulation |
| CN109542603A (en) * | 2018-11-22 | 2019-03-29 | 北京航空航天大学 | It is a kind of improve different priorities task between isolation multi dimensional resource shielding system |
Non-Patent Citations (1)
| Title |
|---|
| 基于改进Stackelberg模型的卫星网络虚拟资源分配算法;刘治国;金朦朦;刘庆利;;计算机工程(第10期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111796934A (en) | 2020-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109684065B (en) | Resource scheduling method, device and system | |
| US8639792B2 (en) | Job processing system, method and program | |
| US7716193B2 (en) | Ensuring timely servicing of desired transactions in a database server | |
| CN110413412B (en) | GPU (graphics processing Unit) cluster resource allocation method and device | |
| CN112463375A (en) | Data processing method and device | |
| CN111309440B (en) | Method and equipment for managing and scheduling multiple types of GPUs | |
| WO2019205370A1 (en) | Electronic device, task distribution method and storage medium | |
| CN111798113A (en) | Resource allocation method, device, storage medium and electronic equipment | |
| CN111796933B (en) | Resource scheduling method, device, storage medium and electronic equipment | |
| CN115292016A (en) | Task scheduling method based on artificial intelligence and related equipment | |
| CN113360269A (en) | Task allocation method, device, server and storage medium | |
| CN111796934B (en) | Task issuing method and device, storage medium and electronic equipment | |
| CN109582439B (en) | DCN deployment method, device, equipment and computer readable storage medium | |
| CN110751376A (en) | Work order distribution scheduling method and device, computer equipment and storage medium | |
| CN107193749B (en) | Test method, device and equipment | |
| CN114116173A (en) | Method, device and system for dynamically adjusting task allocation | |
| CN110413393B (en) | Cluster resource management method and device, computer cluster and readable storage medium | |
| CN111475251A (en) | Cluster container scheduling method, system, terminal and storage medium | |
| US20140047454A1 (en) | Load balancing in an sap system | |
| CN112988383A (en) | Resource allocation method, device, equipment and storage medium | |
| CN114237902A (en) | Service deployment method and device, electronic equipment and computer readable medium | |
| CN114090201A (en) | Resource scheduling method, device, equipment and storage medium | |
| CN113504988A (en) | Control method, system, medium and device for system computing power resource scheduling | |
| CN113703945A (en) | Scheduling method, device, equipment and storage medium of micro-service cluster | |
| CN110673931A (en) | Distributed calculation method for document synthesis, document synthesis system and control device thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |