CN111736981A - Container resource allocation method, device, equipment and storage medium - Google Patents
Container resource allocation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111736981A CN111736981A CN201910708644.9A CN201910708644A CN111736981A CN 111736981 A CN111736981 A CN 111736981A CN 201910708644 A CN201910708644 A CN 201910708644A CN 111736981 A CN111736981 A CN 111736981A
- Authority
- CN
- China
- Prior art keywords
- resource
- target
- container
- standby
- resources
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000013468 resource allocation Methods 0.000 title claims abstract description 61
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 63
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
-
- 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
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
-
- 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
-
- 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
- G06F9/505—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 considering the load
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The embodiment of the invention discloses a container resource allocation method, a device, equipment and a storage medium. The method comprises the following steps: acquiring a maximum resource allocation value of a target container, and determining a target private resource and a target standby resource of the target container according to the maximum resource allocation value and a target percentile; determining a shared standby resource corresponding to the target container according to the target standby resource and the rest standby resources of at least one rest container except the target container in the physical machine corresponding to the target container; and configuring the resources of the target container according to the target private resources and the shared standby resources. Through the technical scheme, the container resources are more reasonably configured, and the resources are saved while performance guarantee is provided to a greater extent.
Description
Technical Field
The embodiment of the invention relates to computer technology, in particular to a container resource configuration method, a device, equipment and a storage medium.
Background
Data centers of platforms such as internet service, e-commerce service, public cloud, private cloud, and internet of things are usually operated on a server or a server cluster (called a physical machine), and at present, the standard architecture of the physical machine mostly uses container and micro-service technology to realize server resource sharing. On the container platform, one physical machine can operate a plurality of containers which are isolated from each other and independently operate; a service or application is divided into a plurality of logically independent microservices, instances of which run in containers.
In order to meet the service performance and optimize the resource utilization, resources with appropriate sizes, such as Central Processing Unit (CPU) resources, internal Memory (Memory) space, Disk (Disk) space, and the like, need to be allocated to the container. The performance of the container or the service can be guaranteed by configuring excessive resources, but the resources are wasted; conversely, if the resource allocation is insufficient, the performance of the container or service cannot be guaranteed. Therefore, configuring appropriate container resources to ensure performance without wasting resources is a crucial issue for efficient operation of physical machines, especially for online services with large load variations and strict performance requirements, such as e-commerce services.
At present, the methods for implementing container resource allocation mainly include: first, the maximum based container resource configuration. According to the scheme, the maximum required resource required by the container bearing service can be estimated according to the experience of an administrator, or historical resource use data of a container is analyzed to determine the maximum used resource of the container, so that the resource is configured for the container conservatively according to the maximum required resource or the maximum used resource, and the performance can be guaranteed when the maximum load is achieved. Second, percentile-based container resource configuration. The solution is that under the condition of determining the maximum value, the resource allocation of a container is determined according to the maximum value and a certain percentile (such as a 90% percentile) so as to ensure that 90% of the resource requirements can be met. And thirdly, dynamic container resource allocation. This scheme requires real-time monitoring of the current load of a container and periodically adjusting the maximum value to dynamically adjust the resource configuration of the container.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: first, maximum-based container resource allocation, while ensuring performance, often does not require maximum resources for containers and services with very dynamic loads and demands, such as internet services and online services, thereby wasting many resources and increasing server costs. Second, the percentile-based container resource allocation, although saving resources to some extent, cannot guarantee to meet all the load and request resource requirements, resulting in certain performance conflicts, and may cause significant loss for important online services, such as e-commerce services, even 1% of the performance problems. Thirdly, the dynamic container resource allocation has the problems of stability and accuracy due to the fact that the technology is not mature enough, and large-scale application cannot be carried out at present.
Disclosure of Invention
Embodiments of the present invention provide a method, an apparatus, a device, and a storage medium for configuring container resources, so as to implement more reasonable configuration of container resources, provide performance guarantee, and save resources.
In a first aspect, an embodiment of the present invention provides a method for configuring container resources, including:
acquiring a maximum resource allocation value of a target container, and determining a target private resource and a target standby resource of the target container according to the maximum resource allocation value and a target percentile;
determining a shared standby resource corresponding to the target container according to the target standby resource and the rest standby resources of at least one rest container except the target container in the physical machine corresponding to the target container;
and configuring the resources of the target container according to the target private resources and the shared standby resources.
In a second aspect, an embodiment of the present invention further provides a device for configuring container resources, where the device includes:
the target private resource determining module is used for acquiring the maximum resource configuration value of the target container and determining the target private resource and the target standby resource of the target container according to the maximum resource configuration value and the target percentile;
a shared standby resource determining module, configured to determine, according to the target standby resource and the remaining standby resources of at least one remaining container except the target container in the physical machine corresponding to the target container, a shared standby resource corresponding to the target container;
and the container resource configuration module is used for configuring the resources of the target container according to the target private resources and the shared standby resources.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors implement the container resource allocation method provided by any embodiment of the invention.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the container resource configuration method provided in any embodiment of the present invention.
According to the embodiment of the invention, the resource configuration maximum value of the target container is obtained, and the target private resource and the target standby resource of the target container are determined according to the resource configuration maximum value and the target percentile, so that the resource of one container is divided into two parts, the target private resource is a resource independently used by the container to meet most of performance requirements of the container, the target standby resource is used for determining the shared standby resource to meet the additional requirements exceeding the target private resource, the problems of resource waste and performance conflict caused by resource reduction are solved, the resource utilization rate is optimized to a certain extent, and the performance guarantee is considered. The shared standby resource corresponding to the target container is determined according to the target standby resource and the other standby resources of at least one other container except the target container in the physical machine corresponding to the target container, so that the resource allocation of a plurality of containers on the physical machine is comprehensively considered when the resource of one container is allocated, the shared standby resource can be shared by the plurality of containers, the problem that the resource and the performance cannot be considered simultaneously due to the fact that a single container resource is allocated in an isolated mode is solved, the container resource is allocated more reasonably, and the allocated container resource can guarantee the performance and save the resource to a greater extent.
Drawings
Fig. 1 is a flowchart of a container resource allocation method according to a first embodiment of the present invention;
fig. 2 is a flowchart of a container resource allocation method according to a second embodiment of the present invention;
fig. 3a is a flowchart of a container resource allocation method in a third embodiment of the present invention;
FIG. 3b is a diagram illustrating the configuration result of two container resources according to the third embodiment of the present invention;
FIG. 3c is a flowchart of another method for allocating container resources according to a third embodiment of the present invention;
FIG. 3d is a diagram illustrating a resource allocation result of a sub-container type according to a third embodiment of the present invention;
fig. 3e is a diagram illustrating comparison of results of configuring container resources based on different container resource configuration methods according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a container resource allocation apparatus according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of an apparatus in the fifth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
The container resource allocation method provided by the embodiment can be applied to container resource allocation in a server or server cluster architecture of a data center. The method may be performed by a container resource configuration apparatus, which may be implemented by software and/or hardware, and may be integrated in a device based on a container technology architecture, such as a personal computer, a server cluster or an intelligent device cluster. Referring to fig. 1, the method of the present embodiment specifically includes the following steps:
s110, obtaining the maximum resource configuration value of the target container, and determining the target private resource and the target standby resource of the target container according to the maximum resource configuration value and the target percentile.
The target container is a container to be configured with the resource, and may be one container or a plurality of containers. The target container may be a newly created container or may be a container that has already been created and run. The resource may be at least one of a CPU resource, an internal memory (memory) resource, a video card memory (video memory) resource, and a disk resource. The maximum resource allocation value refers to a maximum resource value that needs to be allocated for the target container, and may be a maximum resource demand value or a maximum historical resource usage value. The resource requirement maximum refers to the maximum resource required by the service carried by the container, for example, it is determined empirically how much resource the container needs according to the service carried by the container. The historical resource usage maximum refers to the maximum of the resources that the container actually uses during operation. The target percentile is a predetermined percentile value that is used to determine the target private resource of the target container. The target percentile may be a predetermined fixed value, such as an empirical 90%, or a dynamic value that dynamically changes based on the operating conditions of the vessel. The target private resource refers to a resource private to the target container, which is directly allocated to the target container and can only be used by the target container. The target standby resource refers to a standby resource calculated for the target container, which is not directly allocated to the target container, but is a resource at a conceptual level.
In the related art, when resources are determined to be configured for a container, either a maximum value (which may be a fixed value or a dynamic value) or a percentile resource value is configured, which has a certain defect and cannot give consideration to both container performance guarantee and high resource utilization rate, so that in the embodiment of the invention, when resources are configured for a target container, the configured resources are divided into two parts, namely a target private resource and a target standby resource, and the target private resource is used for meeting the resource requirements of most of the performance of the container so as to guarantee most of the performance of the target container; the target standby resource serves as an optional resource for providing resource requirements beyond the target private resource to guarantee additional performance of the target container. By the arrangement, excessive resources cannot be distributed to the target container, sufficient guarantee can be provided for the performance of the target container, and the resource utilization rate and the performance guarantee are considered to a great extent.
Illustratively, obtaining the maximum resource configuration value of the target container comprises: and acquiring historical resource use data corresponding to the target container, and determining the maximum historical resource use value of the target container according to the historical resource use data to serve as the maximum resource configuration value.
The historical resource usage data refers to data of resources used by the container in actual operation in a historical time period before the current time, and may be time-continuous resource usage data, that is, historical resource usage time sequence data, or time-discontinuous resource usage data.
When the maximum resource configuration value is the maximum historical resource usage value, it is necessary to obtain resource usage data of the target container in a period of time before the current time, that is, historical resource usage data. If the target container is a newly created container, the historical resource usage data of the same kind of container with the same load characteristics can be acquired as the historical resource usage data of the target container. The load characteristics here refer to CPU, memory, video memory, disk storage, and the like. The resource allocation process of the target container is the resource allocation process of creating the container. If the target container is a container which is created and operated, the historical resource usage data of the target container can be directly acquired. The resource allocation process of the target container is the process of updating the container resource.
After the historical resource usage data of the target container is acquired, data statistics analysis can be performed on the historical resource usage data to obtain a maximum resource value in the historical resource usage data, which is used as the maximum historical resource usage value of the target container, that is, to obtain a maximum resource configuration value max of the target container. The method has the advantages of reducing the dependence of the resource allocation process on human experience and improving the objectivity and rationality of resource allocation.
Illustratively, determining the target private resource and the target standby resource of the target container according to the resource configuration maximum and the target percentile comprises: determining target private resources of the target container according to the maximum resource configuration value and the target percentile; and determining the target standby resource according to the maximum resource configuration value and the target private resource.
The determination process of the target private resource and the target standby resource comprises the following steps: and calculating the product of the maximum value max of the resource configuration and the target percentile, namely determining the resource percentile value Tail-bound of the target container, and taking the value Tail-bound as the target private resource of the target container. Then, the target private resource is subtracted from the maximum resource configuration value to obtain the target standby resource of the target container, i.e. the target standby resource is max-Tail-bound.
S120, determining the shared standby resource corresponding to the target container according to the target standby resource and the other standby resources of at least one other container except the target container in the physical machine corresponding to the target container.
The physical machine is a name of a physical computer relative to a virtual machine, and may be a single computer or a server, or may be a server cluster or an intelligent device cluster. The remaining containers are concepts corresponding to the target container, which are containers in the physical machine other than the target container. In the embodiment of the invention, the resources of each container on the physical machine are set as the private resources and the standby resources, and the rest of the standby resources are the standby resources of the rest of the containers. Shared standby resources refer to standby resources shared by multiple containers, which are actually allocated resource space, rather than conceptual resources that can interface multiple containers simultaneously, rather than a single container. A shared spare resource is a resource that actually carries a container beyond the performance of a private resource. The shared standby resource on one physical machine can be one, so that all containers on the physical machine share the shared standby resource; or a plurality of containers, wherein the first part of containers share one shared standby resource on the physical machine, the second part of containers share the second shared standby resource, and the like. The number of the shared standby resources is determined according to the service requirement, and in this embodiment, it is preferably set that one shared standby resource is set on one physical machine, so as to maximally compatible with the resource utilization rate and the performance guarantee.
In the embodiment of the present invention, the determination of the shared standby resource needs to depend on the standby resource of the container to which the shared standby resource can be interfaced, that is, the resource configuration of multiple containers needs to be considered comprehensively. When the shared standby resource on the physical machine is only docked with a part of containers, determining that the shared standby resource corresponding to the target container needs to depend on the target standby resource of the target container and the remaining standby resources of the respective remaining containers corresponding to the shared standby resource. When the shared standby resources on the physical machine are docked with all containers, determining that the shared standby resources corresponding to the target container need to depend on the target standby resources of the target container and the remaining standby resources of all remaining containers. The integrated manner of the target standby resource and the at least one remaining standby resource, i.e. the determination manner of the shared standby resource, may be to take the maximum value of each standby resource, or to group (or classify) each standby resource, to take the average value or the maximum value of the sum of the grouped (or each class) standby resources, etc.
S130, configuring resources of the target container according to the target private resources and the shared standby resources.
After the target private resource of the target container and the corresponding shared standby resource are determined, resource allocation can be performed according to the target private resource and the shared standby resource, and then resource allocation of the target container is completed.
According to the technical scheme of the embodiment, the resource configuration maximum value of the target container is obtained, the target private resource and the target standby resource of the target container are determined according to the resource configuration maximum value and the target percentile, the resource of one container is divided into two parts, the target private resource is a resource independently used by the container to meet most of performance requirements of the container, the target standby resource is used for determining the shared standby resource to meet the additional requirements exceeding the target private resource, the problems of resource waste and performance conflict caused by resource reduction are solved, the resource utilization rate is optimized to a certain extent, and the performance guarantee is considered. The shared standby resource corresponding to the target container is determined according to the target standby resource and the other standby resources of at least one other container except the target container in the physical machine corresponding to the target container, so that the resource allocation of a plurality of containers on the physical machine is comprehensively considered when the resource of one container is allocated, the shared standby resource can be shared by the plurality of containers, the problem that the resource and the performance cannot be considered simultaneously due to the fact that a single container resource is allocated in an isolated mode is solved, the container resource is allocated more reasonably, and the allocated container resource can guarantee the performance and save the resource to a greater extent.
Example two
In this embodiment, a step of "generating the target percentile" is added on the basis of the first embodiment. Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted. Referring to fig. 2, the method for configuring container resources provided in this embodiment includes:
s210, acquiring the maximum resource configuration value of the target container.
And S220, determining target resource use time sequence data and initial percentile corresponding to the target container.
The target resource usage time series data is historical resource usage data obtained when the target container or a container of the same kind of the target container (when the target container is newly created) is configured with resources based on the percentile container resource configuration method. The initial percentile refers to a percentile according to which the target container or the like performs resource configuration based on the percentile before the current operation.
The percentile in the related art percentile-based container resource allocation method is usually set by human experience, and has human subjectivity. Therefore, before the target private resource is determined according to the percentile, the initial percentile is dynamically adjusted according to the resource use condition of the target container, so that the percentile which is more objective and more adaptive to the target container, namely the target percentile, is obtained.
To generate the target percentile, an initial percentile corresponding to the target container and target resource usage timing sequence data need to be obtained first. When the target container or the similar container thereof has configured the resources according to the percentile container resource configuration method, the historical resource use time sequence data of the target container or the similar container thereof and the percentile used by the configured resources can be directly obtained and respectively used as the target resource use time sequence data and the initial percentile corresponding to the target container.
When the target container or the similar container thereof does not configure the resources according to the percentile container resource configuration method, initial resources need to be configured for the target container according to the percentile container resource configuration method, and the initial resources can be called as initial private resources because the resources are determined according to the percentile. The percentile that is relied upon at this time is the initial percentile. Illustratively, determining the target resource usage timing data and the initial percentile corresponding to the target container comprises: determining initial private resources of the target container according to the maximum resource configuration value and the initial percentile; and when the target container runs with the initial private resource, obtaining target resource use time sequence data of the target container in a preset time period. When the method is specifically implemented, the initial private resource of the target container is determined according to the product of the maximum resource configuration value of the target container and the initial percentile. And then triggering the target container to run for at least the duration of a preset time period under the condition of the initial private resources, and taking the resource use time sequence data in the preset time period as the target resource use time sequence data of the target container.
And S230, determining the duration ratio of the continuous resource usage value greater than the initial private resource in the target resource usage time sequence data.
The continuous resource usage value refers to a set number of resource usage values, where a time difference between every two resource usage values is smaller than a preset time threshold, and the preset time threshold may be set manually in advance, for example, 3 minutes. The initial private resource is the private resource of the target container corresponding to the initial percentile.
The basis for judging whether the initial percentile is matched with the performance requirement of the target container is mainly based on whether the resources allocated by the initial percentile can guarantee most of the container performance, that is, whether the conflict between the resources and the performance (the allocated resources cannot guarantee the service performance, referred to as performance conflict for short) caused by the allocated resources during the container operation period meets the tolerance of the user to the performance conflict degree, and the performance conflict degree can be represented by the time length of the performance conflict in the whole operation time length. When the performance conflict degree is characterized, the persistent performance conflict has a large influence on the performance of the container, so in order to improve the efficiency of determining the performance conflict degree, the continuous resource usage value is adopted in the embodiment, instead of the single resource usage value.
One way to characterize performance conflicts more intuitively is for the resource usage value to exceed the initial private resource allocated by the container. Then, when the performance conflict degree of the target container is determined, the ratio of the duration that the continuous resource usage value in the target resource usage time sequence data is greater than the initial private resource to the total duration of the target resource usage time sequence data is counted, that is, the duration ratio is determined.
Exemplarily, S230 includes: determining that the peak value of each resource use in the target resource use time sequence data is larger than the peak value length of the initial private resource; generating first peak value time sequence data according to each peak value length, a preset peak value length threshold value and target resource use time sequence data; and determining the duration ratio of continuous peaks in the first peak time sequence data.
The resource usage peak value is a maximum value of the resource usage value in a certain time interval, and in a curve corresponding to the resource usage time series data, the resource usage peak value is a peak value of each curve. The peak length refers to the magnitude of the resource usage peak greater than the initial private resource. The preset peak length threshold is a preset peak length, and is used for screening all peak lengths to filter data with smaller peak lengths. The first peak timing data is timing data in which only resource usage data having a peak length exceeding a preset peak length threshold is retained. The continuous peak value refers to a set number of resource usage peak values, wherein the time difference between every two resource usage peak values is smaller than a preset time threshold value.
The above process of determining the time-length ratio may be further improved as follows: determining each resource use peak value in the target resource use time sequence data, calculating the difference value of each resource use peak value and the initial private resource, and determining the peak value length corresponding to each resource use peak value. Then, in the target resource usage time series data, reserving each resource usage value with the peak length larger than a preset peak length threshold value, and generating first peak value time series data. And finally, counting the ratio of the duration occupied by the continuous peaks in the first peak time sequence data to the total duration of the first peak time sequence data, namely determining the duration occupied ratio. The method has the advantages that on one hand, through the filtering processing of the resource use value, the data volume is reduced, the determination efficiency of the time length ratio is improved, and therefore the adjustment efficiency of the percentile is further improved; on the other hand, the resource use value with smaller peak value length is filtered, the sensitivity of percentile adjustment can be reduced, the percentile adjustment operation with too small percentile adjustment amplitude is reduced to a certain extent, and the actual value of the percentile adjustment operation is enhanced.
S240, adjusting the initial percentile according to the duration percentage, the preset percentage threshold and the demand percentage, and generating a target percentile.
The preset duty ratio threshold refers to a preset duration ratio, for example, 10%, which is used to characterize the tolerance of the user to the performance conflict degree. The demand duty is a preset duration duty that is used to characterize the degree of performance conflict of the container desired by the user. The demand fraction may be equal to a preset fraction threshold, such as 10%, or may not be equal.
Whether the initial percentage needs to be adjusted and how to adjust are determined by comparing the duration ratio with a preset ratio threshold. And if the duration ratio is greater than the preset ratio threshold, increasing the initial percentile according to the demand ratio to generate a target percentile. This situation illustrates that the private resources allocated to the target container at present cannot guarantee most of the performance of the target container, and the configuration of the target private resources needs to be increased according to the magnitude relationship between the demand duty ratio and the duration duty ratio, so as to guarantee the operational performance of the convergence and the service. And if the duration ratio is smaller than the preset ratio threshold, reducing the initial percentile according to the demand ratio, and generating a target percentile. This situation indicates that the private resources allocated to the target container are more, which is wasted on the basis of ensuring most of the performance of the target container, and the configuration of the target private resources needs to be reduced according to the relationship between the demand duty ratio and the duration duty ratio, so as to further save resources.
And S250, determining the target private resource and the target standby resource of the target container according to the maximum resource configuration value and the target percentile.
S260, determining the shared standby resource corresponding to the target container according to the target standby resource and the other standby resources of at least one other container except the target container in the physical machine corresponding to the target container.
S270, configuring resources of the target container according to the target private resources and the shared standby resources.
According to the technical scheme of the embodiment, the duration ratio that the continuous resource usage value is greater than the initial private resource in the target resource usage time sequence data is determined through the target resource usage time sequence data and the initial percentile corresponding to the target container, and the initial percentile is adjusted according to the duration ratio, the preset ratio threshold and the demand ratio to generate the target percentile. The dynamic determination of the target percentile which is more objective and more adaptive to the target container is realized, so that the determination of the target private resource and the target standby resource can be more objective and more suitable for the target container, the flexibility of resource allocation of the target container is further improved, and the resource utilization rate of the target container and the compatibility of performance guarantee are further improved.
EXAMPLE III
In this embodiment, based on the first embodiment, further optimization is performed on "determining the shared standby resource corresponding to the target container according to the target standby resource and the remaining standby resources of at least one remaining container except the target container in the physical machine corresponding to the target container". Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted. Referring to fig. 3a, the method for configuring container resources provided in this embodiment includes:
s310, obtaining the maximum resource allocation value of the target container, and determining the target private resource and the target standby resource of the target container according to the maximum resource allocation value and the target percentile.
S320, determining the resource demand correlation between every two containers according to the historical resource use time sequence data corresponding to the target container and each other container.
The resource demand correlation refers to the correlation of the resource demand variation trend between two containers. If the resource demand correlation of the two containers is high, it means that the resource demand variation of the two containers is consistent. Conversely, if the resource demand of two containers is low in correlation, it means that their resource demands are not correlated or inversely correlated, e.g., when one container has a high resource demand, the other container has a low resource demand.
The determination of the shared standby resources needs to comprehensively consider the standby resources of the corresponding containers, and in this embodiment, the comprehensive manner of the multiple standby resources is determined to be based on the resource demand correlation between the containers. Then the resource demand dependencies between the containers need to be determined before the shared standby resources are determined. Since the historical resource usage timing data may reflect the resource demand condition of the container in a period of time, the resource demand correlation in this embodiment may be directly calculated according to the historical resource usage timing data of any two containers of the target container and the remaining containers, for example, the covariance of formula (1) may be used for calculation:
wherein COV (x, y) represents covariance of two containers, x, y represent historical resource usage timing data of two containers respectively,respectively, a resource usage average value of the historical resource usage time series data of the corresponding container, and n represents the number of resource usage values in the historical resource usage time series data.
The resource demand correlation can also be calculated by utilizing partial characteristic data in the historical resource usage time sequence data. Since the peak usage rate (probability of occurrence of a peak) in the historical resource usage data can more significantly reflect the resource demand of the container, the resource demand correlation between two containers can be calculated by using the peak data in the historical resource usage time series data corresponding to the two containers.
Exemplarily, S320 includes: generating corresponding second peak value time sequence data according to the historical resource usage time sequence data corresponding to the target container and each of the other containers; and respectively determining the similarity between every two second peak value time sequence data as the resource demand correlation between every two corresponding containers.
Wherein the second peak time series data is time series data of resource usage peaks in the reserved historical resource usage time series data.
Determining the resource demand correlation by using the peak data requires first obtaining second peak time sequence data corresponding to each container, specifically filtering historical resource use time sequence data of the containers, and removing non-peak resource use values. In order to further improve the determination efficiency of the subsequent resource demand correlation and improve the reference value of the resource demand correlation, the resource usage peak value which does not exceed the target private resource in all the resource usage peak values can be further removed, because the resource usage peak value which exceeds the target private resource can better represent the usage of the standby resource by the container. Then, the similarity between every two second peak time series data is calculated by using the Jacard coefficient such as the formula (2), and the similarity can represent the resource demand correlation between two containers.
Where J (a, B) represents the resource demand correlation between the two containers, and A, B represents the second peak timing data of the two containers, respectively.
The advantage of this arrangement is that the determination efficiency of the resource demand dependency can be improved and the reference value of the resource demand dependency can be improved.
S330, classifying the target container and other containers based on a clustering algorithm according to the resource demand correlation to obtain at least one container category.
According to the characteristics of the resource demand correlation, it can be known that the probability that each container with consistent resource demand correlation reaches a performance peak value in the same time period is high, the probability that each container with inconsistent resource demand correlation reaches the performance peak value in the same time period is low, so that a target container and each other container can be classified according to the resource demand correlation, the containers with consistent correlation are in the same container class, the containers with inconsistent correlation are in different container classes, and then shared standby resources are determined according to each standby resource corresponding to each container class, so that each container corresponding to the shared standby resources can be ensured not to reach the performance peak value at the same time, the sum of the standby resources which are less than all the containers and have high probability of the shared standby resources is ensured, peak resource demand is greatly reduced, and resource allocation is reduced as much as possible, but also can ensure the performance of the related container and reduce the performance conflict of the container to the maximum extent.
In particular, a clustering algorithm such as K-Means may be used to classify the target container and each of the remaining containers based on the resource demand dependency of each container to obtain at least one container class. The process of classifying the container based on the K-Means algorithm may be:
1) randomly selecting N containers from the containers as the initial "center point" of each category.
2) Euclidean distances are calculated from each container to the N central points, and the distances are defined as covariance COV of the container and historical resource usage time sequence data of the container at the central point.
3) And classifying the containers into the nearest classes according to Euclidean distance, wherein each class is a container which is associated with each other.
4) The center of each category is updated, the center being defined as the average of the historical resource usage timing data for all containers within the same category.
5) And continuing to 2) iterating until a certain number of times or a specified error requirement is met.
S340, determining shared standby resources according to the sum of the standby resources corresponding to the container types.
Wherein the sum of the standby resources is the sum of the standby resources of each container in the corresponding container category.
First, the sum of the standby resources of all containers in each container category is determined, and then the sum of the standby resources corresponding to the number of container categories can be obtained. If there is a target container and remaining containers within the container category, then the reserve resource sum is the sum of the target reserve resource and each remaining reserve resource, and if there is no target container within the container category, then the reserve resource sum is the sum of each remaining reserve resource. Then, the maximum value of the sum of the standby resources is determined, and the maximum value of the sum of the standby resources is determined as the shared standby resource of the target container and the related rest containers.
If the number of container categories is at least two, then the shared standby resource is less than the sum of the target standby resource and each of the remaining standby resources; if the number of container classes is one, then the shared standby resource is equal to the sum of the target standby resource and each of the remaining standby resources.
Taking two containers as an example, if the resource requirement correlation of the two containers is low (uncorrelated or negative correlation), their shared spare resource is equal to the maximum value of the spare resources of the two containers, i.e. the shared spare resource is MAX (MAX)1–Tail_bound1,max2–Tail_bound2) Wherein max1And max2The maximum resource allocation values, Tail _ bound, for container 1 and container 2, respectively1And Tail _ bound2The resource percentile values for container 1 and container 2, respectively. If the resource demand dependency of two containers is high, their shared standby resource is equal to the SUM of the respective standby resources, i.e. the shared standby resource is SUM (max)1–Tail_bound1,max2–Tail_bound2)。
According to the method of this embodiment, the two containers 1 and 2 with low correlation of resource demand are configured with resources, so as to obtain the resource configuration result and the resource usage time sequence data curve of the two containers shown in fig. 3 b.
And S350, configuring the resources of the target container according to the target private resources and the shared standby resources.
It should be noted that, in this embodiment, the corresponding operations in the second embodiment may also be adopted for determining the target percentile.
Referring to fig. 3c, if resource allocation is performed on all containers on one physical machine, the flow of resource allocation based on the maximum value of the historical resource usage and the target percentile is roughly:
A. the method comprises the steps of statistically analyzing historical resource use data of a single container on a physical machine, and determining the maximum value of historical resource use of the single container;
B. configuring a private resource of a single container, the determination of the private resource being calculable from a maximum value of historical resource usage and a target percentile;
C. calculating the standby resources of the single container, and calculating the standby resources of the single container by using the maximum historical resource usage value and the private resources of the single container;
D. carrying out statistical analysis on the resource demand correlation among all containers on the physical machine;
E. performing container clustering analysis according to the resource demand correlation between every two containers to obtain each container type;
F. and calculating and configuring the shared standby resources of all containers on the physical machine according to the standby resources of all containers and the container types.Here, the sum of the standby resources for a container class is the sum of the standby resources for each container in that container class, i.e.The shared spare resource is the maximum of the sum of the spare resources of the container classes, i.e.
Based on the above description and fig. 3d, when resource allocation is performed on a container in which 9 containers are divided into 3 container types, as a result, the total amount of private resources allocated to the 9 containers is C1+ C2+ … + C9, and the shared standby resources are max (H1+ H2+ H3, H4+ H5+ H6+ H7, and H8+ H9).
Referring to fig. 3e, resource allocation of 18 containers is performed by using the conventional percentile (90%) based container resource allocation method (method 1), the maximum value based container resource allocation method (method 2), the average value based container resource allocation method (method 3) and the resource allocation method (method 4) of the present invention, and the total amount of allocated resources (the total amount of 90% percentile resource allocation is used as a standardized reference) and the performance conflict (the percentage of duration of the performance conflict) are compared, and the smaller the two comparison indexes are, the better the values are. Although the total resource allocation amount of the method 4 is not the minimum, the total resource allocation amount is far less than that of the method 2, and the performance conflict of the method 4 is also small. In summary, other methods have difficulty optimizing resource usage while satisfying performance, while method 4 works best in terms of ensuring container performance and conserving resources at the same time.
According to the technical scheme of the embodiment, the resource demand correlation between every two containers is determined, the containers are classified according to the resource demand correlation, the container types with consistent resource demand correlation of the containers in the container types and inconsistent resource demand correlation of the containers in different container types are obtained, the shared standby resources are determined according to the standby resource sum corresponding to the container types, the arrival time of the performance peak value of each container is staggered, the shared standby resources can be determined according to the maximum value of the standby resource sum, the configuration of the shared standby resources is further optimized by using the resource demand correlation of each container, the shared standby resources are smaller than the standby resource sum of each container, and the performance of each container is further guaranteed while the resources are saved.
Example four
The present embodiment provides a container resource allocation apparatus, referring to fig. 4, the apparatus specifically includes:
a target private resource determining module 410, configured to obtain a maximum resource configuration value of the target container, and determine a target private resource and a target standby resource of the target container according to the maximum resource configuration value and the target percentile;
a shared standby resource determining module 420, configured to determine, according to the target standby resource and the remaining standby resources of at least one remaining container except the target container in the physical machine corresponding to the target container, a shared standby resource corresponding to the target container;
and a container resource configuration module 430, configured to configure resources of the target container according to the target private resource and the shared standby resource.
Optionally, the target private resource determining module 410 includes:
the target private resource determining submodule is used for determining the target private resources of the target container according to the maximum resource configuration value and the target percentile;
and the target standby resource determining submodule is used for determining the target standby resource according to the maximum resource configuration value and the target private resource.
Optionally, on the basis of the foregoing apparatus, the apparatus further includes a target percentile generation module, configured to:
determining target resource usage time sequence data and an initial percentile corresponding to a target container before determining a target private resource and a target standby resource of the target container according to the maximum resource configuration value and the target percentile;
determining the duration ratio that the continuous resource usage value in the target resource usage time sequence data is greater than the initial private resource, wherein the initial private resource is the private resource of the target container corresponding to the initial percentile;
and adjusting the initial percentile according to the duration percentage, the preset percentage threshold and the demand percentage to generate the target percentile.
Further, the target percentile generation module is specifically configured to:
determining that the peak value of each resource use in the target resource use time sequence data is larger than the peak value length of the initial private resource;
generating first peak value time sequence data according to each peak value length, a preset peak value length threshold value and target resource use time sequence data;
and determining the duration ratio of continuous peaks in the first peak time sequence data.
Optionally, the shared standby resource determining module 420 is specifically configured to:
determining the resource demand correlation between every two containers according to the historical resource use time sequence data corresponding to the target container and each of the other containers;
classifying the target container and other containers based on a clustering algorithm according to the correlation of the resource demands to obtain at least one container category;
and determining shared standby resources according to the standby resource sum corresponding to each container type, wherein the standby resource sum is the sum of the standby resources of each container in the corresponding container type.
Further, the shared standby resource determining module 420 is further specifically configured to:
generating corresponding second peak value time sequence data according to the historical resource usage time sequence data corresponding to the target container and each of the other containers;
and respectively determining the similarity between every two second peak value time sequence data as the resource demand correlation between every two corresponding containers.
Optionally, the target private resource determining module 410 is specifically configured to:
and acquiring historical resource use data corresponding to the target container, and determining the maximum historical resource use value of the target container according to the historical resource use data to serve as the maximum resource configuration value.
By the container resource allocation device in the fourth embodiment of the invention, more reasonable allocation of container resources is realized, and the resources are saved while performance guarantee is provided to a greater extent.
The container resource allocation device provided by the embodiment of the invention can execute the container resource allocation method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, in the embodiment of the container resource allocation apparatus, each unit and each module included in the embodiment are only divided according to functional logic, but are not limited to the above division as long as the corresponding function can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
EXAMPLE five
Referring to fig. 5, the present embodiment provides an apparatus, which includes: one or more processors 520; the storage 510 is configured to store one or more programs, and when the one or more programs are executed by the one or more processors 520, the one or more processors 520 implement the method for configuring container resources according to the embodiment of the present invention, including:
acquiring a maximum resource allocation value of the target container, and determining a target private resource and a target standby resource of the target container according to the maximum resource allocation value and the target percentile;
determining shared standby resources corresponding to the target container according to the target standby resources and the rest standby resources of at least one rest container except the target container in the physical machine corresponding to the target container;
and configuring the resources of the target container according to the target private resources and the shared standby resources.
Of course, those skilled in the art can understand that the processor 520 may also implement the technical solution of the container resource configuration method provided in any embodiment of the present invention.
The device shown in fig. 5 is only an example and should not bring any limitation to the function and the scope of use of the embodiments of the present invention. As shown in fig. 5, the apparatus includes a processor 520 and a storage device 510; the number of the processors 520 in the device may be one or more, and one processor 520 is taken as an example in fig. 5; the processor 520 and the storage means in the device may be connected by a bus or other means, in fig. 5 by way of example only, a bus 530.
The storage device 510 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the container resource configuration method in the embodiment of the present invention (for example, a target private resource determining module, a shared standby resource determining module, and a container resource configuration module in the container resource configuration device).
The storage device 510 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the storage 510 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, storage 510 may further include memory located remotely from processor 520, which may be connected to devices over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
EXAMPLE six
The present embodiments provide a storage medium containing computer-executable instructions which, when executed by a computer processor, are operable to perform a method of container resource configuration, the method comprising:
acquiring a maximum resource allocation value of the target container, and determining a target private resource and a target standby resource of the target container according to the maximum resource allocation value and the target percentile;
determining shared standby resources corresponding to the target container according to the target standby resources and the rest standby resources of at least one rest container except the target container in the physical machine corresponding to the target container;
and configuring the resources of the target container according to the target private resources and the shared standby resources.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the above method operations, and may also perform related operations in the container resource configuration method provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, where the computer software product may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk, or an optical disk of a computer, and includes several instructions to enable a device (which may be a server or a network device) to execute the container resource configuration method provided in the embodiments of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method for configuring container resources, comprising:
acquiring a maximum resource allocation value of a target container, and determining a target private resource and a target standby resource of the target container according to the maximum resource allocation value and a target percentile;
determining a shared standby resource corresponding to the target container according to the target standby resource and the rest standby resources of at least one rest container except the target container in the physical machine corresponding to the target container;
and configuring the resources of the target container according to the target private resources and the shared standby resources.
2. The method of claim 1, wherein determining the target private resource and the target standby resource of the target container according to the resource configuration maximum and the target percentile comprises:
determining target private resources of the target container according to the maximum resource configuration value and the target percentile;
and determining the target standby resource according to the maximum resource configuration value and the target private resource.
3. The method of claim 1, further comprising, prior to the determining the target private resource and the target standby resource of the target container according to the resource configuration maximum and the target percentile:
determining target resource usage time sequence data and an initial percentile corresponding to the target container;
determining a duration ratio that a continuous resource usage value in the target resource usage timing sequence data is greater than an initial private resource, wherein the initial private resource is a private resource of the target container corresponding to the initial percentile;
and adjusting the initial percentile according to the duration percentage, a preset percentage threshold and a demand percentage to generate the target percentile.
4. The method of claim 3, wherein the determining that the duration of time that the continuous resource usage value is greater than the initial private resource in the target resource usage timing data comprises:
determining that each resource usage peak in the target resource usage timing data is greater than a peak length of the initial private resource;
generating first peak value time sequence data according to each peak value length, a preset peak value length threshold value and the target resource use time sequence data;
and determining the duration ratio of continuous peaks in the first peak time sequence data.
5. The method of claim 1, wherein the determining the shared standby resource corresponding to the target container based on the target standby resource and the remaining standby resources of at least one remaining container in the physical machine corresponding to the target container comprises:
determining the resource demand correlation between every two containers according to the historical resource use time sequence data corresponding to the target container and each of the other containers;
classifying the target container and each of the other containers based on a clustering algorithm according to each of the resource demand correlations to obtain at least one container category;
and determining the shared standby resources according to the standby resource sum corresponding to each container type, wherein the standby resource sum is the sum of the standby resources of each container in the corresponding container type.
6. The method of claim 5, wherein determining the resource demand correlation between two containers according to the historical resource usage timing data corresponding to the target container and each of the remaining containers comprises:
generating corresponding second peak value time sequence data according to the historical resource usage time sequence data corresponding to the target container and each of the other containers;
and respectively determining the similarity between every two second peak value time sequence data as the resource demand correlation between every two corresponding containers.
7. The method of claim 1, wherein obtaining the maximum resource configuration value of the target container comprises:
obtaining historical resource usage data corresponding to the target container, and determining a maximum historical resource usage value of the target container according to the historical resource usage data to serve as the maximum resource configuration value.
8. A container resource allocation apparatus, comprising:
the target private resource determining module is used for acquiring the maximum resource configuration value of the target container and determining the target private resource and the target standby resource of the target container according to the maximum resource configuration value and the target percentile;
a shared standby resource determining module, configured to determine, according to the target standby resource and the remaining standby resources of at least one remaining container except the target container in the physical machine corresponding to the target container, a shared standby resource corresponding to the target container;
and the container resource configuration module is used for configuring the resources of the target container according to the target private resources and the shared standby resources.
9. An apparatus, characterized in that the apparatus comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the container resource configuration method of any one of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for configuring container resources according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910708644.9A CN111736981A (en) | 2019-08-01 | 2019-08-01 | Container resource allocation method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910708644.9A CN111736981A (en) | 2019-08-01 | 2019-08-01 | Container resource allocation method, device, equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111736981A true CN111736981A (en) | 2020-10-02 |
Family
ID=72646296
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910708644.9A Pending CN111736981A (en) | 2019-08-01 | 2019-08-01 | Container resource allocation method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111736981A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113641453A (en) * | 2021-08-17 | 2021-11-12 | 中国联合网络通信集团有限公司 | Node selection method and node selection equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102822798A (en) * | 2010-03-31 | 2012-12-12 | 国际商业机器公司 | Methods and apparatus for resource capacity evaluation in a system of virtual containers |
CN108090225A (en) * | 2018-01-05 | 2018-05-29 | 腾讯科技(深圳)有限公司 | Operation method, device, system and the computer readable storage medium of database instance |
CN108415772A (en) * | 2018-02-12 | 2018-08-17 | 腾讯科技(深圳)有限公司 | A kind of resource adjusting method, device and medium based on container |
WO2019019807A1 (en) * | 2017-07-24 | 2019-01-31 | ***股份有限公司 | Resource allocation method and device |
-
2019
- 2019-08-01 CN CN201910708644.9A patent/CN111736981A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102822798A (en) * | 2010-03-31 | 2012-12-12 | 国际商业机器公司 | Methods and apparatus for resource capacity evaluation in a system of virtual containers |
WO2019019807A1 (en) * | 2017-07-24 | 2019-01-31 | ***股份有限公司 | Resource allocation method and device |
CN108090225A (en) * | 2018-01-05 | 2018-05-29 | 腾讯科技(深圳)有限公司 | Operation method, device, system and the computer readable storage medium of database instance |
CN108415772A (en) * | 2018-02-12 | 2018-08-17 | 腾讯科技(深圳)有限公司 | A kind of resource adjusting method, device and medium based on container |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113641453A (en) * | 2021-08-17 | 2021-11-12 | 中国联合网络通信集团有限公司 | Node selection method and node selection equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107045456B (en) | Resource allocation method and resource manager | |
US8869160B2 (en) | Goal oriented performance management of workload utilizing accelerators | |
CN111176792B (en) | Resource scheduling method and device and related equipment | |
CN110221920B (en) | Deployment method, device, storage medium and system | |
CN111176840B (en) | Distribution optimization method and device for distributed tasks, storage medium and electronic device | |
CN108681481B (en) | Service request processing method and device | |
CN113038609B (en) | Method, device and equipment for allocating bandwidth resources based on communication demands | |
CN111614754B (en) | Fog-calculation-oriented cost-efficiency optimized dynamic self-adaptive task scheduling method | |
WO2023087658A1 (en) | Task scheduling method, apparatus and device, and readable storage medium | |
CN104881322A (en) | Method and device for dispatching cluster resource based on packing model | |
CN110888732A (en) | Resource allocation method, equipment, device and computer readable storage medium | |
CN113296929A (en) | Resource matching method, device and system based on cloud computing | |
CN111611076B (en) | Fair distribution method for mobile edge computing shared resources under task deployment constraint | |
CN114968601B (en) | Scheduling method and scheduling system for AI training jobs with resources reserved in proportion | |
CN117608840A (en) | Task processing method and system for comprehensive management of resources of intelligent monitoring system | |
JPWO2022093319A5 (en) | ||
CN113535398B (en) | Resource allocation adjustment method, device, electronic equipment and readable storage medium | |
CN111736981A (en) | Container resource allocation method, device, equipment and storage medium | |
CN109039694B (en) | Global network resource allocation method and device for service | |
CN109495595B (en) | IP address allocation method, device, communication system and storage medium | |
CN113918301A (en) | Request processing method and device, electronic equipment and storage medium | |
CN111614735B (en) | Novel fog computing architecture based on weighted round robin algorithm and task scheduling method thereof | |
CN116204293A (en) | Resource scheduling method, device, computer equipment and storage medium | |
CN116954905A (en) | Task scheduling and migration method for large Flink data | |
CN114860449B (en) | Data processing method, device, equipment and storage medium |
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 |