CN114090240A - Scheduling method, device, equipment and storage medium for public cloud computing resources - Google Patents
Scheduling method, device, equipment and storage medium for public cloud computing resources Download PDFInfo
- Publication number
- CN114090240A CN114090240A CN202111304483.0A CN202111304483A CN114090240A CN 114090240 A CN114090240 A CN 114090240A CN 202111304483 A CN202111304483 A CN 202111304483A CN 114090240 A CN114090240 A CN 114090240A
- Authority
- CN
- China
- Prior art keywords
- public cloud
- resource
- computing
- computing resource
- cloud server
- 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
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
- G06F9/5027—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
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
-
- 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]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
Abstract
The embodiment of the invention discloses a scheduling method, a device, equipment and a storage medium for public cloud computing resources, which are applied to a plug-in a public cloud server, wherein the plug-in is used for managing virtual equipment applied by a user to the public cloud server, and the method comprises the following steps: monitoring a first computing resource occupied by a work task running on a virtual device; when monitoring that the public cloud server recovers at least part of the first computing resources, applying for second computing resources for the virtual equipment from the public cloud server according to a preset resource scheduling strategy; and replacing the first computing resource to be recycled with the second computing resource to run the work task. By the technical scheme, the work tasks of the users can be ensured to continuously and stably run in the public cloud computing resources, and the data loss caused by the interruption of the work tasks is avoided.
Description
Technical Field
The embodiment of the invention relates to the field of cloud computing, in particular to a scheduling method, device, equipment and storage medium for public cloud computing resources.
Background
With the development of over a decade, cloud computing technology has become mature and widely accepted and applied. With the development of the cloud computing concept, it is also gradually differentiated into public clouds, private clouds, and hybrid clouds. The public cloud has the advantages of high availability, low cost and expansion as required compared with independent maintenance of hardware resources of enterprises.
Cloud manufacturers with massive cloud computing resources can provide a large amount of computing resources of public clouds for enterprises, and different charging modes are designed for the computing resources of the public clouds, and generally divided into a reserved pricing mode, an on-demand pricing mode and a bidding mode (note that different public clouds are slightly different in calling law). The method comprises the steps that a pricing mode is purchased and used as needed, capacity is calculated and paid according to running examples (examples are cloud virtual hosts) in an hourly or second mode, and the price is the most expensive; the reservation pricing model has certain usage commitments (e.g., 1 year, 3 years usage commitments), and provides a substantial discount (typically 60% of the on-demand computing resources) compared to the on-demand pricing model; the bidding mode is an inexpensive mode for obtaining public cloud computing resources, which is provided by cloud manufacturers, and the offered public cloud computing resources have great flexibility, the price of the offered public cloud computing resources changes according to the market supply and demand relationship, and the offered public cloud computing resources have very obvious price advantage (the price of the offered public cloud computing resources is usually 10% -20% of the price of the on-demand pricing mode) compared with the on-demand pricing mode. Therefore, the bidding mode for acquiring the computing resources of the public cloud is undoubtedly the most cost-effective choice for most enterprises. However, an instance interruption mechanism exists in the bidding mode, and a cloud system of a cloud manufacturer comprehensively considers the price of computing resources and the stock of a resource pool, and performs interruption recovery on the public cloud computing resources acquired by the enterprise on an indefinite basis, which may cause interruption of a business process running in the public cloud computing resources, cause partial data loss, and cause a certain loss to the enterprise.
Disclosure of Invention
The invention provides a scheduling method, a scheduling device, scheduling equipment and a storage medium for public cloud computing resources, and aims to solve the problem that when the public cloud computing resources are recycled, a service process running in the public cloud computing resources is interrupted.
In a first aspect, an embodiment of the present invention provides a public cloud computing resource-oriented scheduling method, which is applied to a plug-in a public cloud server, where the plug-in is used to manage a virtual device that a user applies for a public cloud server, and the method includes:
monitoring a first computing resource occupied by a work task running on the virtual device;
when monitoring that the public cloud server recovers at least part of the first computing resources, applying a second computing resource for the virtual equipment from the public cloud server according to a preset resource scheduling strategy;
and replacing the first computing resource to be recycled with the second computing resource to run the work task.
In a second aspect, an embodiment of the present invention further provides a scheduling apparatus for public cloud computing resources, which is applied to a plug-in a public cloud server, where the plug-in is used to manage a virtual device that a user applies for a public cloud server, and the apparatus includes:
the environment sensor is used for monitoring a first computing resource occupied by a work task running on the virtual equipment;
the resource combination optimizer is used for applying a second computing resource for the virtual equipment to the public cloud server according to a preset resource scheduling strategy when monitoring that the public cloud server recovers at least part of the first computing resource;
and the scheduling controller is used for replacing the first computing resource to be recovered with the second computing resource to run the work task.
In a third aspect, an embodiment of the present invention further provides a computer device, where the computer device includes:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the public cloud computing resource oriented scheduling method of the first aspect.
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 public cloud computing resource oriented scheduling method according to the first aspect.
The invention provides a scheduling method, a device, equipment and a storage medium for public cloud computing resources, which are applied to a plug-in a public cloud server, the plug-in is used for managing virtual equipment applied by a user to the public cloud server, monitoring first computing resources occupied by work tasks running on the virtual equipment, feeding back the load state of the first computing resources in real time, applying for second computing resources for the virtual equipment by the public cloud server according to a preset resource scheduling strategy when monitoring that at least part of the first computing resources are recycled by the public cloud server, replacing the first computing resources to be recycled with the second computing resources to run the work tasks, ensuring the continuous and stable running of the work tasks of the user in the public cloud computing resources, avoiding the interruption of the work tasks and causing data loss, and further knowing the requirements of the current user on the public cloud computing resources by monitoring the work state of the first computing resources, the scheduling strategy of applying computing resources to the public cloud server is dynamically perfected according to actual requirements, and different user requirements are met.
Drawings
Fig. 1 is a flowchart of a public cloud computing resource-oriented scheduling method according to an embodiment of the present invention;
fig. 2 is a flowchart of a public cloud computing resource-oriented scheduling method according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of a public cloud computing resource-oriented scheduling apparatus according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device according to a fourth 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.
Cloud Computing (Cloud Computing) is an internet-based mode of addition, usage, and delivery of related services that provide dynamically scalable and virtualized resources over the internet. The cloud is the other end of internet connection, can visit various application and service from the high in the clouds, also can be at high in the clouds safe storage data, and the high in the clouds can constantly expand the capacity in order to satisfy different users 'demand, and more specifically, cloud computing is one kind and is supplied and managed with scalable, elasticity, shared physics and virtual resource pool in the mode of serving as required certainly to provide network access's mode.
Cloud computing is an internet-based computing manner by which shared software and hardware resources and information can be provided to various terminals and other devices of a computer as required. Cloud computing is a pay-per-use model that provides available, convenient, on-demand network access into a configurable shared pool of computing resources (resources including networks, servers, storage, applications, services) that can be provisioned quickly, with little administrative effort, or interaction with service providers.
Cloud Computing is a product of development and fusion of traditional computer and Network Technologies, such as Distributed Computing (Distributed Computing), Parallel Computing (Parallel Computing), Utility Computing (Utility Computing), Network Storage (Network Storage Technologies), Virtualization (Virtualization), Load balancing (Load Balance), hot backup redundancy (High Available), and the like.
The characteristic of 'acquiring according to requirements' of cloud computing enables a user to acquire required computing resources from a cloud at any time, and dynamic matching of the computing resources and the business scale is achieved. In the cloud computing era, computing resources are no longer the bottleneck in building large-scale systems, and in most cases, cost becomes the biggest challenge for enterprises in building large-scale applications.
In order to reduce the cost of the cloud system, most importantly, the characteristics of cloud computing resources, especially 'acquisition on demand' and 'pay on use', are fully utilized. For example, scaling computing resources in time according to traffic changes is a common way to reduce costs. The micro service architecture replaces the single architecture in architecture, so that the system obtains finer and more scalable granularity, and more appropriate and cheaper computing resources are utilized.
Example one
Fig. 1 is a flowchart of a scheduling method for public cloud computing resources according to an embodiment of the present invention, where this embodiment is applicable to a case where a user applies for a public cloud server provided by a cloud vendor to lease cloud computing resources, where the method is applied to a plug-in the public cloud server, where the plug-in is used to manage virtual equipment applied by the user to the public cloud server, the virtual equipment is used to run a work task allocated by the user, the scheduling method for public cloud computing resources may be executed by a scheduling device for public cloud computing resources, the scheduling device may be implemented by software and/or hardware, and may be configured in a computer device, such as a server, a workstation, a personal computer, and the like, and the scheduling method specifically includes the following steps:
s101, monitoring a first computing resource occupied by a work task running on the virtual equipment.
In this embodiment, the virtual device refers to a virtual resource provided by a public cloud service of a cloud manufacturer and allowed to be rented by a user, and may be a virtual host (an executable instance) that is prepared in advance by the cloud manufacturer at a cloud end for different service scenarios, or a virtual host (an executable instance) that is formed by combining various computing resources selected by the user from various computing resources provided by the cloud end according to the service requirements of the user. It can be understood that the virtual device in this embodiment refers to a sum of various computing resources provided by the public cloud service for the user, and the number of the virtual devices is not limited in this embodiment of the present invention. The public cloud service refers to that a cloud computing operator (cloud manufacturer for short) provides cloud services to the outside by using ultra-large-scale infrastructure.
Among the computing resources available in the public cloud, the computing resources generally refer to CPU resources and memory resources required for running a computer program, and the virtual device is configured with at least CPU and memory resources, for example, a universal balanced virtual device provided by a cloud manufacturer a for a high network packet transceiving scene (video barrage, telecommunication service forwarding, and the like) includes a CPU memory ratio of 1:4 (dual-core CPU, 8G memory, quad-core CPU, 16G memory, and the like), the CPU type is intel pentium series, the maximum available basic bandwidth capability is 32.0Gbit/s, and the maximum available network packet transceiving capability is 2400 ten million PPS.
In this embodiment, a user applies for the number of virtual devices from a public cloud server to be set according to actual service requirements, when the service requirements are large, the number of the obtained virtual devices is large, when the service requirements are small, the number of the obtained virtual devices is small, and the computing resources in each virtual device are also configured and selected for a specific work task.
In this embodiment, a plug-in of a public cloud server is installed with a software/hardware device for monitoring and managing virtual devices, and the device includes an environment sensor, a resource combination optimizer, and a scheduling controller, where the environment sensor is responsible for collecting and counting load and traffic change conditions of computing resources for running a user job task, and providing an optimization basis for the computing resources for the resource combination optimizer; the resource combination optimizer is responsible for monitoring all computing resources applied by a user to the public cloud server and determining an optimal resource scheduling strategy; and the scheduling controller is responsible for flexibly scheduling the applied public cloud computing resources, and when receiving a computing resource capacity expansion, capacity reduction or replacement request, the scheduling controller applies for the optimal computing resources to the resource combination optimizer.
In the specific implementation, the environment sensor is initialized, the resource combination optimizer is initialized, the environment sensor monitors computing resources occupied by work tasks running on the virtual device, the computing resources are used as first computing resources, the first computing resources comprise a CPU and a memory, the environment sensor collects information of load, the memory and flow change conditions of the first computing resources, monitoring information related to the first computing resources is obtained, the environment sensor regularly sends the collected monitoring information in a period of time to the resource combination optimizer, the resource combination optimizer adjusts a resource scheduling strategy according to the monitoring information, and sends a resource expansion request, a resource contraction request, a resource replacement request and the like to the scheduling controller according to the monitoring information.
It is understood that, in the present embodiment, the user does not refer to a natural person, but refers to an identity authentication account (ID number) authenticated by the public cloud server, and the account is allowed to apply for the cloud computing resource from the public cloud server.
S102, when it is monitored that the public cloud server recovers at least part of the first computing resources, applying for second computing resources for the virtual equipment from the public cloud server according to a preset resource scheduling strategy.
In this embodiment, the public cloud server refers to a virtual server that operates on a network side and is responsible for organizing and coordinating cloud computing resources and providing various computing services based on user requirements, and may be understood as a public cloud service in a cloud computing platform.
Because the computing resources provided by the bidding mode are large in scale and extremely cheap in price, although an instance interruption mechanism exists, in order to save cost, a user also can select to put services except the core services into public cloud computing resources obtained through the bidding mode, such as big data computing, machine learning model training, batch processing of non-real-time response data, and the like. Therefore, the safe and stable operation of the core service can be ensured, the load of the user side is reduced, and the requirements of other non-core services on high flow consumption and large data calculation amount can be met.
In an actual situation, the cloud end comprehensively considers conditions such as resource price and resource pool stock, and partially or completely recycles the computing resources acquired by the user through the bidding mode, for example, the supply and demand relationship of a certain resource pool is in short supply and demand relationship, the resource is in short supply, and for example, the highest bidding price set for the certain resource pool is lower than the current market price of the resource pool.
In this embodiment, when monitoring that the public cloud server recovers at least part of the first computing resources, the resource combination optimizer applies for a second computing resource for the virtual device from the public cloud server according to a preset resource scheduling policy.
Specifically, the resource combination optimizer may obtain configuration parameters, an interruption rate, and a lease price of an idle computing resource from a public cloud server; and applying for idle computing resources for the virtual equipment from the public cloud server as second computing resources according to a preset resource scheduling strategy by referring to the configuration parameters, the interruption rate and the lease price. The configuration parameters of the computing resources comprise parameters such as a CPU memory ratio, a maximum operation bandwidth, a maximum hard disk capacity and the like.
In this embodiment, the preset resource scheduling policy includes a cost priority policy, an availability priority policy, and the like.
When a user applies for an idle computing resource for a public cloud server, a computing resource application list is generally configured in a plug-in the public cloud server according to a service requirement of the user, and the list includes virtual device configuration information required by the user, specifically, detailed information such as a CPU memory ratio, network resources, hard disk capacity, bandwidth requirements and the like. Because the cloud possibly cannot provide virtual resources completely consistent with the computing resource application list, in order to meet the requirements of work tasks, a user can also set a recommended machine type list which is allowed to be used in the plug-in, and the recommended machine type list is a compatible list which is returned by the resource combination optimizer according to the type of the virtual machine provided by the current public cloud server, is compatible with the computing resource application list and is configured with machines not lower than the application list.
The cost priority strategy takes the computing resource with the lowest lease price applied to the public cloud server as a first priority, and if the current public cloud server cannot provide the computing resource specified in the computing resource application list, other models in the public cloud server are sequentially applied according to the sequence of the lease price of the computing resource from low to high. Since the price of the computing resource in the public cloud bidding mode is lower than the price of the computing resource in the on-demand pricing mode, the cost priority policy follows the order of the bidding mode priority and the on-demand pricing mode.
As a specific example of this embodiment, when the preset resource scheduling policy includes a cost priority policy, the resource combination optimizer may obtain a lease price of a computing resource that is idle in the public cloud server when monitoring that the public cloud server recovers at least a part of the first computing resource; and according to the sequence of the lease price from low to high, applying idle computing resources for the virtual equipment from the public cloud server in sequence as second computing resources.
The availability priority strategy takes the application of the most stable (low interruption rate) computing resources to the public cloud server as the first priority, and simultaneously has the principle of low cost, and the computing resources are applied to the public cloud server according to the sequence of a bidding mode priority and an on-demand pricing mode. Based on the computing resources applied by the bidding mode, a certain random probability exists in the computing resources and is recycled by the public cloud server, so that a user can conveniently consider the stability of the computing resources in the bidding mode, a cloud manufacturer generally provides interruption rate data of the computing resources, and the interruption rate represents the recycling probability of the computing resources in the bidding mode of the past year.
As a specific example of this embodiment, when the preset resource scheduling policy includes an availability priority policy, and when it is monitored that the public cloud server recovers at least part of the first computing resources, the resource combination optimizer applies to the public cloud server to access idle computing resources provided in the bidding mode, calculates a current recovery probability of the idle computing resources with reference to an interruption rate of the computing resources in the bidding mode, and sequentially applies to the public cloud server for the virtual device for the idle computing resources as the second computing resources in an order from a low recovery probability to a high recovery probability.
It should be noted that, in the above example, if a computing resource matching the computing resource application list cannot be applied from the public cloud server, and the plug-in is not configured with the list of allowable recommended machine types, the resource combination optimizer returns that the computing resource is empty. If the recommended machine type list is allowed to be used by configuration in the plug-in, when the resource combination optimizer cannot apply for the computing resource matched with the computing resource application list from the public cloud server, the resource combination optimizer applies for the computing resource in the recommended machine type list to the public cloud server as a second computing resource replacing the first computing resource to be recycled according to the priority of the bidding mode and the rear order of the pricing-on-demand mode based on the recommended machine type list and following the preset resource scheduling strategy.
S103, replacing the first computing resource to be recycled with the second computing resource to run the work task.
In this embodiment, when the resource combination optimizer successfully applies for the second computing resource to the public cloud server, the resource combination optimizer sends a replacement instruction for the first computing resource to the scheduling controller; and when the scheduling controller receives the replacement instruction, replacing the first computing resource to be recycled with the second computing resource to run the work task.
In this embodiment, a plug-in of the public cloud server is installed with a software/hardware device for monitoring and managing virtual devices, and the device further includes a change response component manager for managing components adapted to changes of computing resources. The scheduling controller is responsible for communicating with the change response component manager, and the scheduling controller can invoke component actions in the change response component manager.
When the computing environment where the plug-in is located is initialized, a change response component manager is initialized, and all components for responding to changes of computing resources are loaded in the change response component manager, where the components include, but are not limited to, a data storage component, a service cluster component (e.g., a Consul component), a container cluster component (e.g., a Kubernates component), and the like.
In a specific implementation manner of this embodiment, when the scheduling controller receives a replacement instruction for the first computing resource issued by the resource combination optimizer, the scheduling controller sends a component action instruction to the change response component manager, and the change response component manager calls the data storage component to mount the storage mounted by the first computing resource to be recovered on the second computing resource; invoking the container cluster component to remove the container associated with the first computing resource to be reclaimed to replace the first computing resource to be reclaimed with the second computing resource to run the work task.
Considering that the computing resource applied by the user to the public cloud server may not be the computing resource specified in the computing resource application list, or there is a case that the resource combination optimizer does not successfully apply for replacing the second computing resource of the first computing resource to be recovered, or there is a case that the performance of the applied second computing resource is not as good as the performance of the first computing resource to be recovered before.
Based on the above situation, the scheduling method further includes: the resource combination optimizer polls and monitors idle computing resources in the public cloud server as third computing resources; comparing the priority of the third computing resource to the priority of the first computing resource; if the priority of the third computing resource is higher than the priority of at least part of the first computing resource, applying for the third computing resource from the public cloud server; the third computing resource is substituted for at least a portion of the first computing resource to run the work task.
The third computing resource replaces at least part of the first computing resource to run the working task, so that the computing resource can be effectively operated in the plug-in unit on the basis of not reducing the continuity and the service stability of the user service, the optimized updating of the applied computing resource is realized, the user can be reasonably helped to plan the use cost of the cloud computing resource, the optimal computing resource in the bidding mode is applied as much as possible, the high efficiency and stability of the service operation are ensured, and the cost is reduced.
It should be noted that, in this embodiment, the second computing resource replaces the first computing resource to be recovered to run the work task allocated by the user, at this time, the second computing resource is changed to the first computing resource occupied by the work task running in the virtual device, and the plug-in for managing the user to apply for the computing resource from the public cloud server continues to execute the scheduling method. Thus, the priority of the third computing resource is compared to the priority of the first computing resource, which refers to the computing resource currently responsible for running the work task, including the second computing resource that is continuously replacing the first computing resource during execution of the scheduling method.
In the embodiment, by monitoring the first computing resource occupied by the work task running on the virtual device, the load state of the first computing resource can be fed back in real time, when the public cloud server is monitored to recover at least part of the first computing resources, the public cloud server applies for second computing resources for the virtual equipment according to a preset resource scheduling strategy, the second computing resources replace the first computing resources to be recovered to run work tasks, the continuous and stable running of user work tasks in the public cloud computing resources is ensured, the interruption of work tasks is avoided, and data loss is caused, the current user's demand for the public cloud computing resources can be further understood by monitoring the working state of the first computing resource, according to actual requirements, resource scheduling strategies for applying computing resources from users to the public cloud server are dynamically perfected, and different user requirements are met.
Example two
Fig. 2 is a flowchart of a public cloud computing resource-oriented scheduling method according to a second embodiment of the present invention, where the second embodiment is based on the foregoing embodiment, and the present embodiment adds technical steps to the public cloud computing resource-oriented scheduling method, so as to add processing operations for a resource capacity expansion request and a resource capacity reduction request. The method specifically comprises the following steps:
s201, monitoring a first computing resource occupied by a work task running on the virtual equipment.
In this embodiment, a computing environment in a plug-in is initialized, a change response component manager is initialized, and all components for responding to changes of computing resources are loaded; initializing an environment sensor; and initializing a resource combination optimizer.
The resource combination optimizer is in charge of communicating with the public cloud server, receiving resource scheduling instructions issued by the public cloud server to users, including resource recovery instructions, and meanwhile sending resource scheduling requests to the public cloud server according to service requirements of the users.
The method comprises the steps that an environment sensor continuously monitors the working state of computing resources occupied by working tasks running in virtual equipment, acquires monitoring information such as memory occupation and flow change of first computing resources, and sends the monitoring information to a resource combination optimizer, and the resource combination optimizer sends a computing resource change request to a scheduling controller according to the monitoring information and instructions issued by a public cloud server, wherein the computing resource change request comprises a resource expansion request, a resource replacement request, a resource recovery request and the like.
S202, when it is monitored that the public cloud server recovers at least part of the first computing resources, applying for second computing resources for the virtual equipment from the public cloud server according to a preset resource scheduling strategy.
In this embodiment, when monitoring that the public cloud server recovers at least part of the first computing resources, the resource combination optimizer applies a second computing resource to the public cloud server as the virtual device according to a preset resource scheduling policy; when successfully applying for the second computing resource, the resource combination optimizer sends a resource reclamation request to the scheduling controller.
S203, replacing the first computing resource to be recovered with the second computing resource to run the work task.
In this embodiment, the scheduling controller continuously monitors a computing resource change request from the resource combination optimizer, and when a resource recovery request sent by the resource combination optimizer is monitored, the scheduling controller calls a component action in the change response component manager to replace the first computing resource to be recovered with the second computing resource to run the work task.
And S204, when the first computing resource does not meet the load work task, applying for a second computing resource for the virtual device from the public cloud server according to a preset resource scheduling strategy.
In this embodiment, the resource combination optimizer receives monitoring information sent by the environment sensor, analyzes the monitoring information, and detects that a first computing resource occupied by a work task running in the virtual device does not satisfy a load work task, at this time, the resource combination optimizer applies a second computing resource for the virtual device to the public cloud server according to a preset resource scheduling policy.
The preset resource scheduling policy includes a cost priority policy, an availability priority policy, and the like.
In a specific implementation of the present implementation, the resource combination optimizer may obtain configuration parameters, an interruption rate, and a lease price of an idle computing resource from a public cloud server; and applying for idle computing resources for the virtual equipment from the public cloud server as second computing resources according to a preset resource scheduling strategy by referring to the configuration parameters, the interruption rate and the lease price.
S205, the second computing resource and the first computing resource are operated together to work the task.
In this embodiment, when the second computing resource is successfully applied, the resource combination optimizer sends a resource capacity expansion request to the scheduling controller, and when the scheduling controller monitors the resource capacity expansion request, the scheduling controller invokes a component action in the change response component manager, adds the second computing resource to the resource pool of the first computing resource, and runs a work task together with the first computing resource.
In this embodiment, when the first computing resource does not satisfy the load work task, a new computing resource is applied to the public cloud server, and resource expansion is performed on the computing resource group currently applied by the user, so that continuous and stable operation of the work task in the cloud computing resource can be ensured, and the situations that the first computing resource is overloaded, the task is crashed, and data loss is caused are avoided.
EXAMPLE III
Fig. 3 is a schematic structural diagram of a scheduling apparatus for public cloud computing resources according to a third embodiment of the present invention, where the apparatus is applied to a plug-in a public cloud server, where the plug-in is used to manage a virtual device that a user applies to the public cloud server, and the apparatus may specifically include an environment sensor 301, a resource combination optimizer 302, and a scheduling controller 303, where the environment sensor 301 is responsible for collecting and counting load, memory, and flow change conditions of computing resources that run a user job task, and providing an optimization basis for the computing resources for the resource combination optimizer; the resource combination optimizer 302 is responsible for monitoring all computing resources applied by a user to a public cloud server, determining an optimal resource scheduling strategy and optimizing the computing resources applied by the user to a cloud end; the scheduling controller 303 is responsible for flexibly scheduling the applied public cloud computing resources, and when the scheduling controller 303 receives a computing resource expansion or replacement request, the scheduling controller 303 applies for the optimal computing resource to the resource combination optimizer 302.
The environment sensor 301 is configured to monitor a first computing resource occupied by a work task running on the virtual device;
a resource combination optimizer 302, configured to apply for a second computing resource for the virtual device to the public cloud server according to a preset resource scheduling policy when it is monitored that the public cloud server recovers at least part of the first computing resource;
and the scheduling controller 303 is configured to replace the first computing resource to be recovered with the second computing resource to run the work task.
In an embodiment of the present invention, the resource combination optimizer 302 is further configured to poll and monitor that there is an idle computing resource in the public cloud server as a third computing resource; comparing the priority of the third computing resource to the priority of the first computing resource; if the priority of the third computing resource is higher than the priority of at least part of the first computing resource, applying for the third computing resource from the public cloud server;
the scheduling controller 303 is further configured to replace at least a portion of the first computing resource with the third computing resource to execute the work task.
In an embodiment of the present invention, the resource combination optimizer 302 is further configured to apply for a second computing resource for the virtual device from the public cloud server according to a preset resource scheduling policy when the first computing resource does not satisfy the load of the work task;
the scheduling controller 303 is further configured to run the work task with the first computing resource using the second computing resource.
In an embodiment of the present invention, the resource combination optimizer 302 is specifically configured to:
acquiring configuration parameters, interruption rate and lease price of idle computing resources from the public cloud server;
and applying the idle computing resources for the virtual equipment to the public cloud server as second computing resources according to a preset resource scheduling strategy by referring to the configuration parameters, the interruption rate and the lease price.
In an embodiment of the present invention, the resource scheduling policy includes a cost-first policy, and the resource combination optimizer 302 is specifically configured to:
acquiring the lease price of the idle computing resources in the public cloud server;
and according to the sequence of the lease price from low to high, sequentially applying for the idle computing resources as second computing resources for the virtual equipment from the public cloud server.
In an embodiment of the present invention, the resource scheduling policy includes an availability priority policy, and the resource combination optimizer 302 is specifically configured to:
applying for access to idle computing resources provided in a bidding mode to the public cloud server;
obtaining the interruption rate of idle computing resources provided by the bidding mode;
calculating a recovery probability for the idle computing resource based on the outage rate;
and sequentially applying for the idle computing resources as second computing resources to the public cloud server for the virtual equipment according to the sequence from low to high of the recovery probability.
In an embodiment of the present invention, the scheduling controller 303 is specifically configured to:
mounting storage mounted by the first computing resource to be recovered on the second computing resource;
removing a container associated with the first computing resource to be reclaimed to replace the first computing resource to be reclaimed with the second computing resource to run the work task.
The scheduling device provided by the embodiment of the invention can execute the scheduling method facing the public cloud computing resource provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 4 is a schematic structural diagram of a computer apparatus according to a fourth embodiment of the present invention, as shown in fig. 4, the computer apparatus includes a processor 400, a memory 401, a communication module 402, an input device 404, and an output device 404; the number of processors 400 in the computer device may be one or more, and one processor 400 is taken as an example in fig. 4; the processor 400, the memory 401, the communication module 402, the input device 404 and the output device 404 in the computer apparatus may be connected by a bus or other means, and fig. 4 illustrates an example of connection by a bus.
The memory 401 is used as a computer-readable storage medium and can be used for storing software programs, computer-executable programs, and modules, such as the modules corresponding to the public cloud computing resource-oriented scheduling method in the embodiment of the present invention (for example, the context sensor 301, the resource combination optimizer 302, and the scheduling controller 303 in the public cloud computing resource-oriented scheduling apparatus shown in fig. 3). The processor 400 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 401, that is, the public cloud computing resource oriented scheduling method is implemented.
The memory 401 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 memory 401 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, memory 401 may further include memory located remotely from processor 400, which may be connected to a computer device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
And the communication module 402 is used for establishing connection with the display screen and realizing data interaction with the display screen.
The input device 404 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the computer apparatus.
The output device 404 may include a display device such as a display screen.
It should be noted that the specific composition of the input device 404 and the output device 404 can be set according to actual situations.
The computer device provided by this embodiment of the present invention is capable of executing the scheduling method for public cloud computing resources provided by any embodiment of the present invention, and has corresponding functions and beneficial effects.
EXAMPLE five
The fifth embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the scheduling method for public cloud computing resources according to any of the above embodiments is implemented.
The scheduling method for the public cloud computing resources is applied to a plug-in a public cloud server, the plug-in is used for managing virtual equipment applied by a user to the public cloud server, and the method comprises the following steps:
monitoring a first computing resource occupied by a work task running on the virtual device;
when monitoring that the public cloud server recovers at least part of the first computing resources, applying a second computing resource for the virtual equipment from the public cloud server according to a preset resource scheduling strategy;
and replacing the first computing resource to be recycled with the second computing resource to run the work task.
Of course, the computer program of the computer-readable storage medium provided in the embodiment of the present invention is not limited to the method operations described above, and may also perform related operations in the public cloud computing resource-oriented scheduling method provided in 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, which can 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 for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the scheduling apparatus for public cloud computing resources, each unit and each module included in the scheduling apparatus 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.
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 scheduling method for public cloud computing resources is applied to a plug-in a public cloud server, and the plug-in is used for managing virtual equipment applied by a user to the public cloud server, and the method comprises the following steps:
monitoring a first computing resource occupied by a work task running on the virtual device;
when monitoring that the public cloud server recovers at least part of the first computing resources, applying a second computing resource for the virtual equipment from the public cloud server according to a preset resource scheduling strategy;
and replacing the first computing resource to be recycled with the second computing resource to run the work task.
2. The method of claim 1, further comprising:
polling and monitoring idle computing resources in the public cloud server as third computing resources;
comparing the priority of the third computing resource to the priority of the first computing resource;
if the priority of the third computing resource is higher than the priority of at least part of the first computing resource, applying for the third computing resource from the public cloud server;
executing the work task with the third computing resource replacing at least a portion of the first computing resource.
3. The method of claim 1 or 2, further comprising:
when the first computing resource does not meet the load of the work task, applying a second computing resource for the virtual equipment from the public cloud server according to a preset resource scheduling strategy;
running the second computing resource with the first computing resource for the work task.
4. The method according to claim 1 or 2, wherein the applying for the second computing resource for the virtual device from the public cloud server according to a preset resource scheduling policy comprises:
acquiring configuration parameters, interruption rate and lease price of idle computing resources from the public cloud server;
and applying the idle computing resources for the virtual equipment to the public cloud server as second computing resources according to a preset resource scheduling strategy by referring to the configuration parameters, the interruption rate and the lease price.
5. The method according to any of claims 1 or 2, wherein the resource scheduling policy comprises a cost-first policy;
applying for a second computing resource for the virtual device from the public cloud server according to a preset resource scheduling policy, including:
acquiring the lease price of the idle computing resources in the public cloud server;
and according to the sequence of the lease price from low to high, sequentially applying for the idle computing resources as second computing resources for the virtual equipment from the public cloud server.
6. The method according to any of claims 1 or 2, wherein the resource scheduling policy comprises an availability precedence policy;
applying for a second computing resource for the virtual device from the public cloud server according to a preset resource scheduling policy, including:
applying for access to idle computing resources provided in a bidding mode to the public cloud server;
obtaining the interruption rate of idle computing resources provided by the bidding mode;
calculating a recovery probability for the idle computing resource based on the outage rate;
and sequentially applying for the idle computing resources as second computing resources to the public cloud server for the virtual equipment according to the sequence from low to high of the recovery probability.
7. The method of claim 1 or 2, wherein the replacing the first computing resource to be reclaimed by the second computing resource to run the work task comprises:
mounting storage mounted by the first computing resource to be recovered on the second computing resource;
removing a container associated with the first computing resource to be reclaimed to replace the first computing resource to be reclaimed with the second computing resource to run the work task.
8. A scheduling device for public cloud computing resources is applied to a plug-in a public cloud server, and is used for managing virtual equipment applied by a user to the public cloud server, and the scheduling device comprises:
the environment sensor is used for monitoring a first computing resource occupied by a work task running on the virtual equipment;
the resource combination optimizer is used for applying a second computing resource for the virtual equipment to the public cloud server according to a preset resource scheduling strategy when monitoring that the public cloud server recovers at least part of the first computing resource;
and the scheduling controller is used for replacing the first computing resource to be recovered with the second computing resource to run the work task.
9. A computer device, characterized in that the computer device comprises:
one or more processors;
a memory for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the public cloud computing resource oriented scheduling method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the public cloud computing resource oriented scheduling method of any one of claims 1-7.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2020112491599 | 2020-11-10 | ||
| CN202011249159.9A CN112269662A (en) | 2020-11-10 | 2020-11-10 | Scheduling method, device, equipment and storage medium for public cloud computing resources |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114090240A true CN114090240A (en) | 2022-02-25 |
Family
ID=74339728
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011249159.9A Pending CN112269662A (en) | 2020-11-10 | 2020-11-10 | Scheduling method, device, equipment and storage medium for public cloud computing resources |
| CN202111304483.0A Pending CN114090240A (en) | 2020-11-10 | 2021-11-05 | Scheduling method, device, equipment and storage medium for public cloud computing resources |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011249159.9A Pending CN112269662A (en) | 2020-11-10 | 2020-11-10 | Scheduling method, device, equipment and storage medium for public cloud computing resources |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112269662A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115499388A (en) * | 2022-08-30 | 2022-12-20 | 阿里巴巴（中国）有限公司 | Virtual host resource allocation method, device, equipment and storage medium |
| CN115525425A (en) * | 2022-09-16 | 2022-12-27 | 中国电信股份有限公司 | Federal learning calculation engine arrangement method and device based on cloud native technology |
| CN115525425B (en) * | 2022-09-16 | 2024-05-14 | 中国电信股份有限公司 | Federal learning calculation engine arrangement method and equipment based on cloud primordial technology |
-
2020
- 2020-11-10 CN CN202011249159.9A patent/CN112269662A/en active Pending
-
2021
- 2021-11-05 CN CN202111304483.0A patent/CN114090240A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115499388A (en) * | 2022-08-30 | 2022-12-20 | 阿里巴巴（中国）有限公司 | Virtual host resource allocation method, device, equipment and storage medium |
| CN115499388B (en) * | 2022-08-30 | 2024-04-12 | 阿里巴巴（中国）有限公司 | Virtual host resource allocation method, device, equipment and storage medium |
| CN115525425A (en) * | 2022-09-16 | 2022-12-27 | 中国电信股份有限公司 | Federal learning calculation engine arrangement method and device based on cloud native technology |
| CN115525425B (en) * | 2022-09-16 | 2024-05-14 | 中国电信股份有限公司 | Federal learning calculation engine arrangement method and equipment based on cloud primordial technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112269662A (en) | 2021-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108881495B (en) | Resource allocation method, device, computer equipment and storage medium | |
| CN102844724B (en) | Power supply in managing distributed computing system | |
| US9154549B2 (en) | Dynamic server farms | |
| US10630559B2 (en) | Virtual machine (VM) realm integration and management | |
| CN108337109B (en) | Resource allocation method and device and resource allocation system | |
| US11372689B1 (en) | Cloud bursting technologies | |
| CN109451051B (en) | Service request processing method and device, electronic equipment and storage medium | |
| CN105357296A (en) | Elastic caching system based on Docker cloud platform | |
| CN109873868A (en) | A kind of computing capability sharing method, system and relevant device | |
| CN105159775A (en) | Load balancer based management system and management method for cloud computing data center | |
| EP2484096A1 (en) | Dynamic load balancing and scaling of allocated cloud resources in an enterprise network | |
| CN105183554A (en) | Hybrid computing system of high-performance computing and cloud computing, and resource management method therefor | |
| WO2013158139A1 (en) | Virtual computing resource orchestration | |
| CN107111520A (en) | Method and system for the real time resources consumption control in DCE | |
| CN106095581B (en) | Network storage virtualization scheduling method under private cloud condition | |
| CN114090240A (en) | Scheduling method, device, equipment and storage medium for public cloud computing resources | |
| CN113821340A (en) | Dynamic balancing method, system, terminal and storage medium of distributed system | |
| Karthikeyan et al. | Preserving resource handiness and exigency-based migration algorithm (PRH-EM) for energy efficient federated cloud management systems | |
| CN114860425A (en) | Service cloud platform based on cloud resources | |
| JP6700552B2 (en) | Process control program, process control device, and process control method | |
| CN111401752A (en) | Flywheel energy storage frequency modulation sharing processing method, device and system and electronic equipment | |
| Costache et al. | Themis: Economy-based automatic resource scaling for cloud systems | |
| Usman et al. | A conceptual framework for realizing energy efficient resource allocation in cloud data centre | |
| CN114675972A (en) | Method and system for flexibly scheduling cloud network resources based on integral algorithm | |
| CN112073223B (en) | System and method for managing and controlling operation of cloud computing terminal and cloud server |
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 |