CN113312161A - Application scheduling method, platform and storage medium - Google Patents
Application scheduling method, platform and storage medium Download PDFInfo
- Publication number
- CN113312161A CN113312161A CN202110492267.7A CN202110492267A CN113312161A CN 113312161 A CN113312161 A CN 113312161A CN 202110492267 A CN202110492267 A CN 202110492267A CN 113312161 A CN113312161 A CN 113312161A
- Authority
- CN
- China
- Prior art keywords
- application
- container
- virtual node
- virtual
- target
- 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
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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
- G06F9/5077—Logical partitioning of resources; Management or configuration of virtualized resources
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45562—Creating, deleting, cloning virtual machine instances
-
- 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
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45595—Network integration; Enabling network access in virtual machine instances
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
Abstract
The embodiment of the application discloses an application scheduling method, which comprises the following steps: determining at least one first virtual node having first identification information; the first virtual node and an edge cluster managed by an application scheduling platform have a corresponding relation; determining resource demand information of a first container to be scheduled with second identification information; the first container to be scheduled corresponds to a virtual application instance to be created; determining a target virtual node from the at least one first virtual node based on the resource demand information; scheduling the first container to be scheduled to the target virtual node; and creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node. The embodiment of the application also discloses an application scheduling platform and a storage medium.
Description
Technical Field
The present application relates to the field of communications technologies, and in particular, to an application scheduling method, a platform, and a storage medium.
Background
With the rapid development of communication network technology, Edge Computing (MEC) technology is becoming mature and widely used, such as unmanned and smart factories. Currently, there are many implementations and open source communities for edge clustering, such as kubidge edge (kubenedege) of platform (kubernets, k8s) community for managing container clusters, k3s, openyurt, edge infrastructure software platform (Starlingx) of cloud computing management platform (openstack) community, and so on.
However, because the storage schemes of the edge clusters are different in implementation mechanisms and support various application types and formats, when the above scheme implements edge cluster storage management, only an edge cluster of a single application type can be managed, that is, the use efficiency of the current storage scheme of the edge cluster is low.
Content of application
In order to solve the above technical problems, embodiments of the present application are expected to provide an application scheduling method, a platform, and a storage medium, so as to solve the problem that only a single application type of edge cluster can be managed when edge cluster storage management is currently implemented, implement an edge cluster method capable of managing multiple application types when edge cluster storage management is implemented, and effectively improve the scheduling efficiency of edge cluster storage management on heterogeneous applications.
The technical scheme of the application is realized as follows:
in a first aspect, a method for application scheduling, the method comprising:
determining at least one first virtual node having first identification information; the first virtual node and an edge cluster managed by an application scheduling platform have a corresponding relation;
determining resource demand information of a first container to be scheduled with second identification information; the first container to be scheduled corresponds to a virtual application instance to be created;
determining a target virtual node from the at least one first virtual node based on the resource demand information;
scheduling the first container to be scheduled to the target virtual node;
and creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node.
Optionally, the determining at least one first virtual node having first identification information includes:
determining at least one edge cluster managed by the application scheduling platform;
determining resource information of the at least one edge cluster to obtain at least one target resource information;
generating at least one second virtual node corresponding to the at least one edge cluster based on each target resource information;
and identifying the at least one second virtual node by adopting the first identification information to obtain the at least one first virtual node.
Optionally, the generating at least one second virtual node corresponding to the at least one edge cluster based on each piece of target resource information includes:
determining a resource category corresponding to each target resource information based on each target resource information;
and generating a corresponding virtual node based on each resource category corresponding to each target resource information to obtain the at least one second virtual node.
Optionally, before determining the resource requirement information of the first container to be scheduled having the second identification information, the method further includes:
determining the configuration content of the virtual application instance to be created;
creating a target virtual application based on the configuration content;
creating a first reference scheduling container based on the target virtual application;
and identifying the first reference scheduling container by adopting the second identification information to obtain the first container to be scheduled.
Optionally, the determining a target virtual node from the at least one first virtual node based on the resource demand information includes:
screening at least one third virtual node from the at least one first virtual node based on the resource demand information and the service directory information of the at least one first virtual node;
evaluating each third virtual node based on the service information of the at least one third virtual node to obtain at least one evaluation value;
and determining the target virtual node from the at least one third virtual node based on the at least one evaluation value.
Optionally, after the target virtual node creates the virtual application instance to be created based on the resource demand information of the first container to be scheduled, the method further includes:
monitoring the state of the first container to be dispatched;
if the state of the first container to be scheduled is a first state, updating the state of the target virtual application to be the first state; the first to-be-scheduled container indicates that the running state of the to-be-created virtual application instance on the edge cluster corresponding to the target virtual node is a target state for the first state;
if the state of the first container to be scheduled is a second state, creating a second reference scheduling container based on the target virtual application;
identifying the second reference scheduling container by using the second identification information to obtain a second container to be scheduled;
scheduling the second container to be scheduled to the target virtual node;
and creating the virtual application instance to be created by the target virtual node based on the resource demand information of the second container to be scheduled.
Optionally, after the updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the method further includes:
if the state of the target virtual application is detected to be the first state, deleting the target virtual application;
and if the target virtual application is detected to be deleted, deleting the first container to be scheduled.
Optionally, after the updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the method further includes:
and managing and controlling the target virtual node in a target communication mode so as to realize the operation and maintenance management of the virtual application instance to be created.
In a second aspect, an application scheduling platform, the application scheduling platform comprising at least: an edge system and a cloud end system; wherein:
the edge system is used for managing at least one edge cluster managed by the application scheduling platform;
the cloud system is configured to implement the steps of the application scheduling method according to any one of the above items based on at least one edge cluster managed by the edge system.
In a third aspect, a storage medium has an application scheduler stored thereon, which when executed by a processor implements the steps of the application scheduling method as defined in any one of the above.
The embodiment of the application scheduling method, the platform and the storage medium determines at least one first virtual node with first identification information, determines resource demand information of a first container to be scheduled with second identification information, then determines a target virtual node from the at least one first virtual node based on the resource demand information, schedules the first container to be scheduled to the target virtual node, and finally creates a virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on an edge cluster corresponding to the target virtual node. Therefore, after the target virtual node is determined from at least one first virtual node with the first identification information based on the resource demand information of the first container to be scheduled with the second identification information, the first container to be scheduled is scheduled to the target virtual node, so that the target virtual node creates a virtual application instance to be created based on the resource demand information of the first container to be scheduled.
Drawings
Fig. 1 is a schematic flowchart of an application scheduling method according to an embodiment of the present application;
fig. 2 is a schematic flowchart of another application scheduling method according to an embodiment of the present application;
fig. 3 is a schematic flowchart of another application scheduling method according to an embodiment of the present application;
fig. 4 is a flowchart illustrating an application scheduling method according to another embodiment of the present application;
fig. 5 is a schematic flowchart of another application scheduling method according to another embodiment of the present application;
fig. 6 is a schematic flowchart of another application scheduling method according to another embodiment of the present application;
fig. 7 is a schematic structural diagram of an application scheduling platform according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of a cloud system according to an embodiment of the present disclosure;
fig. 9 is a schematic structural diagram of an edge system and an edge cluster according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
An embodiment of the present application provides an application scheduling method, which is applied to an application scheduling platform and shown in fig. 1, and includes the following steps:
The first virtual node and the edge cluster managed by the application scheduling platform have a corresponding relation.
In this embodiment of the present application, the application scheduling platform may be an MEC platform for performing management control on an edge cluster, and the application scheduling platform may simultaneously manage and control the edge cluster of at least one application type and format. The first identification information is information for distinguishing at least one first virtual node from virtual nodes constituting the application scheduling platform, the first virtual node being different from physical nodes of the application scheduling platform. The at least one first virtual node is created based on the edge cluster managed by the application scheduling platform, that is, the application scheduling platform creates the corresponding at least one first virtual node based on each edge cluster managed by the application scheduling platform. The number of the at least one first virtual node corresponding to each edge cluster is determined by the type included in each edge cluster.
And 102, determining the resource requirement information of the first container to be scheduled with the second identification information.
And the first container to be scheduled corresponds to the virtual application instance to be created.
In this embodiment of the present application, the second identification information is used to distinguish the first container to be scheduled from the original container of the application scheduling platform. The container may be Pod. The resource demand information of the first container to be scheduled is obtained by the application scheduling platform in advance for the edge cluster, and the resource demand information of the first container to be scheduled at least comprises Service-Level Agreement (SLA) demands, third-party Service demands and the like.
In this embodiment of the present application, the target virtual node is an optimal virtual node in the at least one first virtual node, where the optimal virtual node is capable of satisfying the resource requirement information of the first container to be scheduled. That is, the target virtual node is the resource requirement information capable of satisfying the first container to be scheduled.
And step 104, scheduling the first container to be scheduled to the target virtual node.
In this embodiment of the present application, scheduling the first container to be scheduled to the target virtual node means that a corresponding relationship between the first container to be scheduled and the target virtual node is established, so that the target virtual node is bound with the first container to be scheduled, and the first container to be scheduled is operated on the target virtual node.
And 105, creating a virtual application instance to be created through the target virtual node based on the resource requirement information of the first container to be scheduled.
And creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node.
In the embodiment of the application, after the first container to be scheduled is scheduled to the target virtual node, the target virtual node creates the virtual application instance to be created corresponding to the first container to be scheduled according to the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs in the edge cluster corresponding to the target virtual node after the virtual application instance to be created is successfully created. Therefore, the virtual nodes are created to represent the edge clusters in the application scheduling platform, so that the virtual nodes are managed by the application scheduling platform to realize the management control of the corresponding edge clusters, and the virtual nodes are not limited by the types of the edge clusters, so that the application scheduling platform can manage various different types of edge clusters, and the management efficiency of the application scheduling platform on the edge clusters is effectively improved.
In the application scheduling method provided by the embodiment of the application scheduling method, at least one first virtual node with first identification information is determined, resource demand information of a first container to be scheduled with second identification information is determined, a target virtual node is determined from the at least one first virtual node based on the resource demand information, the first container to be scheduled is scheduled to the target virtual node, and finally a virtual application instance to be created is created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on an edge cluster corresponding to the target virtual node. Therefore, after the target virtual node is determined from at least one first virtual node with the first identification information based on the resource demand information of the first container to be scheduled with the second identification information, the first container to be scheduled is scheduled to the target virtual node, so that the target virtual node creates a virtual application instance to be created based on the resource demand information of the first container to be scheduled.
Based on the foregoing embodiments, an embodiment of the present application provides an application scheduling method, which is applied to an application scheduling platform and shown in fig. 2, and includes the following steps:
In the embodiment of the application, the application scheduling platform counts at least one edge cluster managed by the application scheduling platform.
In this embodiment of the present application, the application scheduling platform obtains resource information of each edge cluster in the at least one edge cluster, obtains at least one target resource information, that is, several edge clusters exist, and obtains several target resource information by statistics, where the target resource information includes multiple types of resource information of the edge clusters. For example, assuming that the application scheduling platform includes 3 edge clusters, namely, an edge cluster 1, an edge cluster 2, and an edge calculation 3, the target resource information 1 corresponding to the edge cluster 1, the target resource information 2 corresponding to the edge cluster 2, and the target resource information 3 corresponding to the edge cluster 3 are determined.
In this embodiment of the present application, based on each target resource information in the at least one target resource information, at least one second virtual node corresponding to each edge cluster is generated, that is, each target resource information corresponds to at least one second virtual node. However, in some specific application scenarios, when each edge cluster provides applications including two types, a second virtual node needs to be generated for each type of application.
Illustratively, if the edge cluster 1 only provides deployment and management (blueprint) applications of the orchestration system cloud of the cloud application, a second virtual node 1 corresponding to the edge cluster 1 is generated; if the edge cluster 2 only provides deployment and management (blueprint) application of the cloud application orchestration system Cloudify, a second virtual node 2 corresponding to the edge cluster 2 is generated; if the edge cluster 3 provides deployment and management (blueprint) application and Virtual Machine (VM) application of the cloud application orchestration system, two second Virtual nodes 3 and 4 corresponding to the edge cluster 3 need to be generated, where the blueprint application provided by the edge cluster 3 corresponding to the second Virtual node 3 and the VM application provided by the edge cluster 3 corresponding to the second Virtual node 4.
And step 204, identifying the at least one second virtual node by adopting the first identification information to obtain at least one first virtual node.
In this embodiment of the present application, the first identification information is used to distinguish at least one second virtual node from a virtual node that is originally present in the application scheduling platform, that is, the first identification information is information used to uniquely identify an edge cluster managed by the application scheduling platform. And identifying each second virtual node in the at least one second virtual node by adopting the first identification information so as to obtain at least one first virtual node, namely identifying the plurality of second virtual nodes by adopting the first identification information so as to obtain the plurality of first virtual nodes.
For example, the first identification information may be marked as label, so that as long as label is identified, it may be determined that the corresponding virtual node is a virtual node corresponding to an edge cluster managed by the application scheduling platform.
And the first container to be scheduled corresponds to the virtual application instance to be created.
In the embodiment of the present application, the first container to be scheduled is a Pod created according to a virtual application instance to be created, where the Pod is created according to needs, and the resource requirement information of the first container to be scheduled is the resource requirement information of the corresponding edge cluster. The second identification information is information for uniquely identifying a container corresponding to the virtual application instance to be created, for example, the second identification information may be a node selector.
In the embodiment of the application, based on the resource demand information, the resource filtering is performed on the edge cluster corresponding to the at least one first virtual node, and then the optimal matching analysis is performed on the at least one first virtual node which basically meets the requirement after filtering, so that the target virtual node is determined. The target virtual node is typically a virtual node. The resource requirement information includes local resources such as Graphics Processing Unit (GPU), Field Programmable Gate Array (FPGA), virtual network card, and the like
And step 207, scheduling the first container to be scheduled to the target virtual node.
And 208, creating a virtual application instance to be created through the target virtual node based on the resource requirement information of the first container to be scheduled.
And creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node.
In the embodiment of the application, the application scheduling platform creates the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the resources of the edge cluster are effectively utilized to provide the service corresponding to the virtual application instance to be created.
Based on the foregoing embodiments, in other embodiments of the present application, step 203 may be implemented by steps 203a to 203 b:
step 203a, determining a resource category corresponding to each target resource information based on each target resource information.
In this embodiment of the present application, the resource category corresponding to each target resource information is the category of the edge cluster corresponding to each target resource information. In some application scenarios, the resource category corresponding to each target resource information includes at least one category.
Step 203b, generating a corresponding virtual node based on each resource category corresponding to each target resource information, and obtaining at least one second virtual node.
In the embodiment of the present application, a corresponding virtual node is generated based on each resource category corresponding to each target resource information, so that when each target resource information corresponds to a plurality of resource categories, a plurality of virtual nodes are generated for each target resource information. In this way, the number of the second virtual nodes generated by the application scheduling platform is at least the same as the determined number of the target resource information. It should be noted that, assuming that the two counted target resource information are the same in type and the two target resource information are substantially the same, two second virtual nodes still need to be created based on the two target resource information, respectively.
Based on the foregoing embodiments, in other embodiments of the present application, referring to fig. 3, before the application scheduling platform performs step 205, the application scheduling platform is further configured to perform steps 209 to 212:
step 209 determines the configuration content of the virtual application instance to be created.
In the embodiment of the present application, the configuration content of the virtual application instance to be created may be obtained by setting by a user, and in some application scenarios, may also be default configuration content for the virtual application instance to be created when the user wishes to create the virtual application instance to be created. The configuration content may be configuration parameters for the virtual application instance to be created.
In the embodiment of the application, the application scheduling platform performs configuration adjustment according to the configuration content and the application template of the pre-packaged heterogeneous application, and creates the target virtual application.
In the embodiment of the present application, for example, based on the created target virtual application, a node Pod of k8s is created, and the first reference scheduling container is obtained.
And step 212, identifying the first reference scheduling container by using the second identification information to obtain a first container to be scheduled.
In this embodiment of the application, in order to distinguish the first reference scheduling container from the Pod node in the k8s platform, the Pod node created for the target virtual application needs to be used for distinguishing, and therefore, the unique identification information and the second identification information are used for identifying the first reference scheduling container, so that the first reference scheduling container is rapidly identified in the following process, and the first container to be scheduled is obtained.
Based on the foregoing embodiments, in other embodiments of the present application, step 206 may be implemented by steps 206a to 206 c:
step 206a, based on the resource demand information and the service directory information of the at least one first virtual node, at least one third virtual node is obtained by screening from the at least one first virtual node.
In this embodiment, the service directory information of at least one first virtual node is provided by each corresponding edge cluster. And determining at least one service directory information matched with the resource demand information from the service directory information of the at least one first virtual node, so that the first virtual node corresponding to the matched at least one service directory information can be determined to be at least one third virtual node. The process of screening at least one third virtual node can be realized by pre-programmed k8s pre-selection scheduling plug-in.
Illustratively, when the service directory information 1 of the first virtual node 1, the service directory information 2 of the second virtual node 2, and the service directory information 3 of the third virtual node 3 are currently included, it is assumed that the service directory information 2 and the service target information 3 match with the resource demand information, and therefore, it may be determined that at least one third virtual node is the second virtual node 2 corresponding to the service directory information 2 and the third virtual node 3 corresponding to the service directory information 3.
And step 206b, evaluating each third virtual node based on the service information of at least one third virtual node to obtain at least one evaluation value.
In this embodiment of the present application, the service information of at least one third virtual node is a computational resource that can be provided by the edge cluster corresponding to each third virtual node, so that each third virtual node is evaluated for the service information of each third virtual node to obtain an evaluation value of each third virtual node, and then evaluation values of all third virtual nodes are obtained.
And step 206c, determining a target virtual node from the at least one third virtual node based on the at least one evaluation value.
In the embodiment of the application, when the larger the evaluation value is, the larger the corresponding edge cluster can provide the optimal computing resource for the first container to be scheduled, the third virtual node corresponding to the largest evaluation value in the at least one evaluation value is determined as the target virtual node. Or, when the smaller evaluation value indicates that the corresponding edge cluster can provide the optimal computing resource for the first container to be scheduled, determining the third virtual node corresponding to the smallest evaluation value in the at least one evaluation value as the target virtual node. Wherein, the steps 206b and 206c can be realized by a pre-programmed k8s preferred scheduling plug-in.
If it is determined that the maximum evaluation value or the minimum evaluation value corresponds to the at least two third virtual nodes based on the at least one evaluation value, one third virtual node may be randomly determined from the at least two determined third virtual nodes as a target virtual node, or a certain resource in the service information of the third virtual node may be further preferentially selected, for example, according to the amount of remaining resources of a Central Processing Unit (CPU) corresponding to the at least two determined third virtual nodes, for example, determining the third virtual node with the largest amount of remaining resources of the CPU as the target virtual node.
Based on the foregoing embodiments, in other embodiments of the present application, referring to fig. 4, after the application scheduling platform performs step 208, the application scheduling platform is further configured to perform steps 213 to 218:
In the embodiment of the present application, the status of the first container to be scheduled includes a success status or a failure status; the successful state refers to a state that the scheduling between the first container to be scheduled and the target virtual node is successful or the connection is successful, and the failed state refers to a state that the scheduling between the first container to be scheduled and the target virtual node is failed or the connection is failed.
The first to-be-scheduled container is in a first state, which indicates that the running state of the to-be-created virtual application instance on the edge cluster corresponding to the target virtual node is in a target state.
In the embodiment of the present application, the first state refers to a success state of the first container to be scheduled. And under the condition that the first container to be scheduled is determined to be in the first state, updating the state of the target virtual application, and updating the state to be in the first state, namely a power state.
In an embodiment of the application, the second status is used to indicate a failure status of the first container to be scheduled. And when the state of the first container to be scheduled is determined to be a failure state, creating a second reference scheduling container based on the target virtual application again.
And step 216, identifying the second reference scheduling container by using the second identification information to obtain a second container to be scheduled.
And step 217, scheduling the second container to be scheduled to the target virtual node.
In the embodiment of the application, when the state of the first container to be scheduled is the second state, the first container to be scheduled is discarded, a new second container to be scheduled corresponding to the target virtual application is created again, and scheduling is performed again, so that effective management and application of the edge cluster of the target virtual application are realized.
Based on the foregoing embodiments, in other embodiments of the present application, referring to fig. 5, after the application scheduling platform performs step 214, the application scheduling platform is further configured to perform steps 219 to 220:
And step 220, if the target virtual application is detected to be deleted, deleting the first container to be scheduled.
In the embodiment of the application, when it is detected that the state of the target virtual application is the first state, it indicates that the application instance to be created has successfully run in the edge cluster corresponding to the target virtual application, and in order to reduce resource consumption of the application scheduling platform, the target virtual application may be deleted, and in the state where the target virtual application has been deleted, the first container to be scheduled is further deleted, so that the utilization rate of resources of the application scheduling platform is improved.
Based on the foregoing embodiments, in other embodiments of the present application, referring to fig. 6, after the application scheduling platform performs step 214, the application scheduling platform is further configured to perform step 221:
and 221, performing management control on the target virtual node by adopting a target communication mode to realize operation and maintenance management on the virtual application instance to be created.
In this embodiment of the present application, the target communication mode may be an asynchronous communication mode, and specifically may be an asynchronous communication mode in a queue form. The operation and maintenance management comprises management, scaling or deletion of the version of the created virtual application instance to be created and the like.
Wherein, different steps in fig. 4, fig. 5 and fig. 6 can be executed in parallel or in a progressive manner according to actual situations.
Based on the foregoing embodiments, an embodiment of the present application provides a schematic structural diagram of an application scheduling platform, and referring to fig. 7, an application scheduling platform 3 includes a cloud system 31 and an edge system 32. Referring to fig. 8, the cloud system 31 includes 5 components: cloud mirror management service 311, cloud mirror repository 312, cloud virtual application processor 313, virtual provider 314, and cloud application lifecycle management service 315; wherein:
the cloud image management service 311 is configured to provide an Application Programming Interface (API) for presentation State Transfer (RESTful) to the outside, and support functions of uploading, deleting, and viewing a virtual machine image, a container image, and an Application package. And the synchronous state of the virtual machine image, the container image and the application package is supported to be displayed.
And a cloud mirror repository (Harbor)312, which is used for uniformly storing the cloud virtual machine mirror image, the container mirror image and the application package, and directly packaging the virtual machine mirror image, the application package and related metadata into the container mirror image for storage and synchronization.
A cloud virtual application processor (Operator)313, configured to convert different third-party resource virtual applications into, for example, Pod schedulable pods according to kubernets. The third party virtual application may support multiple virtual application subtypes, more than Blueprint, a separate VM, and a separate Container (Container) by Cloudify. And the third-party virtual application records the corresponding resource requirements, such as SLA requirements and/or third-party service requirements in detail. The Operator monitors the third-party virtual applications, that is, when a third-party virtual application is newly created, the Operator creates a piece of Pod, Pod resource demand information of k8s to summarize the total demand of the third-party virtual application, wherein the SLA demand and/or the third-party service demand can be written in the Annotation (Annotation) of the Pod, and other information such as an application package can be put in a configuration map (Configmap) referred by the Pod.
A Virtual Provider (Virtual Kubelet Provider)314 is used to virtualize each edge cluster managed as a node of kubernets, thus facilitating application scheduling to heterogeneous edge clusters. The specific reasons are: some edge clusters provide different physical infrastructures for virtual machines and containers, so it is necessary to create a virtual kubernets node for running a virtual machine and a virtual kubernets node for running a container separately for each such edge cluster. The virtual provider is further configured to regard the GPU, the virtual network card, and the memory of the remote edge cluster as 3 types of devices, and respectively create corresponding device-plugin (device-plugin). Each device-plugin is used for acquiring corresponding resource information, that is, the device-plugin corresponding to the GPU is used for acquiring GPU information of the edge cluster, the device-plugin corresponding to the virtual network card is used for acquiring virtual network card information of the edge cluster, and the device-plugin corresponding to the memory is used for acquiring storage resource information of the edge cluster. The information collected by the three device-plugin programs is transmitted to the cloud virtual application processor through the virtual provider program, so that the cloud virtual application processor makes a scheduling decision. The storage resource information of the edge cluster can be divided into a plurality of types, such as file storage, object storage, block storage, fast storage, slow storage, and the like. Each type of storage corresponds to one device, and the capacity of the storage is converted into the number of devices.
And the cloud application lifecycle management service (Appmanager)315 is used for creating the virtual application based on the application template stored in the Harbor. Meanwhile, operation and maintenance management information such as deletion, version management, expansion and contraction capacity, SLA management and the like aiming at the application instance is forwarded to the edge application life cycle management service in the edge system through the message queue.
Referring to fig. 9, edge system 32 includes 4 components: mirror synchronization service 321, edge Harbor 322, edge application lifecycle management service 323, and scheduler adapter 324; wherein:
and the mirror image synchronization service 321 is configured to monitor states of the virtual machine mirror image, the container mirror image, and the application package in the edge Harbor, and synchronize the states to a manager corresponding to the edge cluster 4.
And the edge Harbor 322 is used for uniformly storing the virtual machine mirror image, the container mirror image and the application package of the edge cluster.
The edge application lifecycle management service (Appmanager)323 is used for directly interfacing with an application executor of an edge cluster, and is responsible for functions of creating and deleting application instances, version management, expansion and contraction capacity, SLA management and the like.
A Scheduler Adapter (Scheduler Adapter)324, configured to collect real-time resource information of the edge cluster 4, synchronize the real-time resource information to the Virtual kubel Provider in the cloud system, and forward a request for creating a Virtual application to the edge Appmanager after receiving the request for creating the Virtual application from the Virtual kubel Provider.
In this way, the implementation steps of the process for implementing the application scheduling method based on the application scheduling platform shown in fig. 7 to 9 may be as follows:
step a11, the front-end page calls the API of the mirror image management service in the edge system to upload the virtual machine mirror image, the container mirror image or the application package.
The front-end page is an interface operated by a user and is irrelevant to the application scheduling platform.
Step a12, the cloud Harbor receives the virtual machine image, the container image or the application package sent by the image management service, realizes synchronization of the virtual machine image, the container image or the application package, and stores the virtual machine image, the container image or the application package.
Step a13, calling an application deployment API of the cloud Appmanager by the front page so that the cloud Appmanager can obtain the virtual machine mirror image, the container mirror image or the application package.
The virtual machine image, the container image or the application package is the configuration content for creating the virtual application instance.
Step a14, the cloud Appmanager creates a third party resource (virtApp) and monitors the state of the virtApp.
Wherein, the third party resource is the target virtual application.
Step a15, the virtual application operator creates kubernets Pod according to VirtApp. Illustratively, Pod is also identified as nondeselector: virtualnode: vm (contractor).
Wherein, kubernetes Pod created according to virtupp is the first container to be scheduled. The nondeselector is the second identification information.
Step a16, the Virtual kubbelet provider constructs one or two kubberenets Virtual nodes as Virtual-kubbelets based on the edge cluster resource information returned by the Scheduler Adapter in the edge system.
The edge cluster resource information comprises GPU of the edge cluster, storage resource information, a virtual network card and the like, and the storage resource information, the virtual network card and the like can be used as equipment resource capacity of the created kubernetes virtual node to be output outwards. Meanwhile, first identification information is added to the kubernets virtual node, for example, the kubernets virtual node to which the first identification information is added may be recorded as label, virtual node, vm (container) NoScheduler.
Wherein, the kubernetes virtual node is the target virtual node.
Step a17, the cloud virtual application processor dispatches the created kubernets Pod to the target virtual node corresponding to the screened edge cluster.
The target virtual node filters the edge clusters according to service directory information provided by each edge cluster and the requirement of the kubernets Pod on third-party service, namely the resource requirement information, through a k8s pre-selection scheduling plug-in, scores the edge clusters through the SLA related information of the edge clusters obtained through filtering through a k8s preferred scheduling plug-in, automatically performs resource matching, and finds the target virtual node corresponding to the optimal edge cluster for the Kubernets Pod.
Step a18, the target virtual node creates an application instance according to Kubernets Pod information, an edge Scheduler Adapter and an edge applier; and updates the state of Kubernetes Pod.
Wherein, the created application instance is the created application instance.
Step a19, the virtual application operator monitors the status of kubernets Pod, if it is found that kubernets Pod is failure (failed), it re-creates kubernets Pod, reschedules, i.e. jumps to step a 16. If Kubernetes Pod is the successful (cached) state, the status of the VirtApp is updated to the cached state.
Wherein the failed state is the second state, and the failed state is the first state.
Step a20, the cloud Appmanager finds that the VirtApp state is succade, and deletes the VirtApp.
Step a21, when the virtual application Operator receives the information of deleting the VirtApp event, the corresponding Kubernetes Pod is deleted.
Step a22, after the application instance is successfully created, the cloud Appmanager asynchronously communicates with the edge Appmanager through the message queue, and executes operation and maintenance operations including version management, capacity expansion and/or instance deletion and the like, so as to manage the application instance running in the edge cluster through the edge Appmanager.
Therefore, the expandability is high, the preselection and optimization steps of the scheduling of the heterogeneous application are customized on the basis of the expansion framework of the kubernets scheduler, for example, the edge cluster is selected on the basis of the maximum network delay, the available storage/available gpu and other devices of the edge cluster are used as screening conditions of the edge cluster, and the expandability is high; when the selected edge cluster cannot create an application due to resource fragmentation, network and other problems, a new edge cluster can be found again in time to realize high availability; and based on a set of scheduling framework and interfaces, multiple types of applications are scheduled to various edge clusters, so that the usability is high.
It should be noted that, for the descriptions of the same steps and the same contents in this embodiment as those in other embodiments, reference may be made to the descriptions in other embodiments, which are not described herein again.
In the application scheduling method provided by the embodiment of the application scheduling method, at least one first virtual node with first identification information is determined, resource demand information of a first container to be scheduled with second identification information is determined, a target virtual node is determined from the at least one first virtual node based on the resource demand information, the first container to be scheduled is scheduled to the target virtual node, and finally a virtual application instance to be created is created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on an edge cluster corresponding to the target virtual node. Therefore, after the target virtual node is determined from at least one first virtual node with the first identification information based on the resource demand information of the first container to be scheduled with the second identification information, the first container to be scheduled is scheduled to the target virtual node, so that the target virtual node creates a virtual application instance to be created based on the resource demand information of the first container to be scheduled.
Based on the foregoing embodiments, an embodiment of the present application provides an application scheduling platform, where the application scheduling platform may be applied to the application scheduling methods provided in embodiments corresponding to fig. 1 to 6, and as shown in fig. 7, the application scheduling platform 3 may include: edge system 31 and cloud system 32, wherein:
an edge system 31, configured to manage at least one edge cluster managed by the application scheduling platform;
the cloud system 32 is configured to implement the following steps based on at least one edge cluster managed by the edge system:
determining at least one first virtual node having first identification information; the first virtual node and the edge cluster managed by the application scheduling platform have a corresponding relation;
determining resource demand information of a first container to be scheduled with second identification information; the first container to be scheduled corresponds to the virtual application instance to be created;
determining a target virtual node from the at least one first virtual node based on the resource demand information;
scheduling a first container to be scheduled to a target virtual node;
and creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node.
In other embodiments of the present application, when the cloud system performs the step of determining at least one first virtual node having the first identification information, the method may be implemented by:
determining at least one edge cluster managed by an application scheduling platform;
determining resource information of at least one edge cluster to obtain at least one target resource information;
generating at least one second virtual node corresponding to at least one edge cluster based on the at least one target resource information;
and identifying the at least one second virtual node by adopting the first identification information to obtain at least one first virtual node.
In other embodiments of the present application, when the cloud system executes the step of generating at least one second virtual node corresponding to at least one edge cluster based on at least one target resource information, the cloud system may be implemented by:
determining a resource category corresponding to each target resource information based on at least one target resource information;
and generating a corresponding second virtual node based on the resource category corresponding to each target resource information to obtain at least one second virtual node.
In other embodiments of the present application, before the cloud system performs the step of determining the resource requirement information of the first container to be scheduled having the second identification information, the cloud system is further configured to perform the following steps:
determining configuration content of a virtual application instance to be created;
creating a target virtual application based on the configuration content;
creating a first reference scheduling container based on the target virtual application;
and identifying the first reference scheduling container by adopting the second identification information to obtain a first container to be scheduled.
In other embodiments of the present application, when the cloud system executes the step of determining the target virtual node from the at least one first virtual node based on the resource demand information, the cloud system may be implemented by:
screening at least one third virtual node from at least one first virtual node based on the resource demand information and the service directory information of at least one first virtual node;
evaluating each third virtual node based on the service information of at least one third virtual node to obtain at least one evaluation value;
and determining a target virtual node from the at least one third virtual node based on the at least one evaluation value.
In other embodiments of the present application, after the cloud system executes the step of creating the virtual application instance to be created based on the resource demand information of the first container to be scheduled by the target virtual node, the cloud system is further configured to execute the following steps:
monitoring the state of a first container to be scheduled;
if the state of the first container to be scheduled is the first state, updating the state of the target virtual application to be the first state; the first container to be scheduled is in a first state, which indicates that the running state of the virtual application instance to be created on the edge cluster corresponding to the target virtual node is in a target state;
if the state of the first container to be scheduled is the second state, a second reference scheduling container is created based on the target virtual application;
and identifying the second reference scheduling container by adopting the second identification information to obtain a second container to be scheduled.
In other embodiments of the present application, after the cloud system executes the step of updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the cloud system is further configured to execute the following steps:
if the state of the target virtual application is detected to be the first state, deleting the target virtual application;
and if the target virtual application is detected to be deleted, deleting the first container to be scheduled.
In other embodiments of the present application, after the cloud system executes the step of updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the cloud system is further configured to execute the following steps:
and managing and controlling the target virtual node in a target communication mode to realize the operation and maintenance management of the virtual application instance to be created.
It should be noted that, for a specific implementation process of the steps executed by the processor in this embodiment, reference may be made to the implementation process in the application scheduling method provided in the embodiments corresponding to fig. 1 to 6, and details are not described here again.
The application scheduling platform provided by the embodiment of the application scheduling platform determines at least one first virtual node with first identification information, determines resource demand information of a first container to be scheduled with second identification information, then determines a target virtual node from the at least one first virtual node based on the resource demand information, schedules the first container to be scheduled to the target virtual node, and finally creates a virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on an edge cluster corresponding to the target virtual node. Therefore, after the target virtual node is determined from at least one first virtual node with the first identification information based on the resource demand information of the first container to be scheduled with the second identification information, the first container to be scheduled is scheduled to the target virtual node, so that the target virtual node creates a virtual application instance to be created based on the resource demand information of the first container to be scheduled.
Based on the foregoing embodiments, embodiments of the present application provide a computer-readable storage medium, referred to as a storage medium for short, where the computer-readable storage medium stores one or more programs to implement the implementation process in the application scheduling method provided in the embodiments corresponding to fig. 1 to 6, and details are not described here again.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present application, and is not intended to limit the scope of the present application.
Claims (10)
1. A method of application scheduling, the method comprising:
determining at least one first virtual node having first identification information; the first virtual node and an edge cluster managed by an application scheduling platform have a corresponding relation;
determining resource demand information of a first container to be scheduled with second identification information; the first container to be scheduled corresponds to a virtual application instance to be created;
determining a target virtual node from the at least one first virtual node based on the resource demand information;
scheduling the first container to be scheduled to the target virtual node;
and creating the virtual application instance to be created through the target virtual node based on the resource demand information of the first container to be scheduled, so that the virtual application instance to be created runs on the edge cluster corresponding to the target virtual node.
2. The method of claim 1, the determining at least one first virtual node having first identification information, comprising:
determining at least one edge cluster managed by the application scheduling platform;
determining resource information of the at least one edge cluster to obtain at least one target resource information;
generating at least one second virtual node corresponding to the at least one edge cluster based on each target resource information;
and identifying the at least one second virtual node by adopting the first identification information to obtain the at least one first virtual node.
3. The method of claim 2, wherein the generating at least one second virtual node corresponding to the at least one edge cluster based on each of the target resource information comprises:
determining a resource category corresponding to each target resource information based on each target resource information;
and generating a corresponding virtual node based on each resource category corresponding to each target resource information to obtain the at least one second virtual node.
4. The method of claim 1, prior to determining the resource demand information for the first to-be-scheduled container having the second identification information, the method further comprising:
determining the configuration content of the virtual application instance to be created;
creating a target virtual application based on the configuration content;
creating a first reference scheduling container based on the target virtual application;
and identifying the first reference scheduling container by adopting the second identification information to obtain the first container to be scheduled.
5. The method of any of claims 1 to 4, wherein said determining a target virtual node from said at least one first virtual node based on said resource demand information comprises:
screening at least one third virtual node from the at least one first virtual node based on the resource demand information and the service directory information of the at least one first virtual node;
evaluating each third virtual node based on the service information of the at least one third virtual node to obtain at least one evaluation value;
and determining the target virtual node from the at least one third virtual node based on the at least one evaluation value.
6. The method of claim 4, after the creating, by the target virtual node, the virtual application instance to be created based on the resource requirement information of the first container to be scheduled, the method further comprising:
monitoring the state of the first container to be dispatched;
if the state of the first container to be scheduled is a first state, updating the state of the target virtual application to be the first state; the first to-be-scheduled container indicates that the running state of the to-be-created virtual application instance on the edge cluster corresponding to the target virtual node is a target state for the first state;
if the state of the first container to be scheduled is a second state, creating a second reference scheduling container based on the target virtual application;
identifying the second reference scheduling container by using the second identification information to obtain a second container to be scheduled;
scheduling the second container to be scheduled to the target virtual node;
and creating the virtual application instance to be created by the target virtual node based on the resource demand information of the second container to be scheduled.
7. The method of claim 6, wherein after updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the method further comprises:
if the state of the target virtual application is detected to be the first state, deleting the target virtual application;
and if the target virtual application is detected to be deleted, deleting the first container to be scheduled.
8. The method of claim 6, wherein after updating the state of the target virtual application to the first state if the state of the first container to be scheduled is the first state, the method further comprises:
and managing and controlling the target virtual node in a target communication mode so as to realize the operation and maintenance management of the virtual application instance to be created.
9. An application scheduling platform, the application scheduling platform comprising at least: an edge system and a cloud end system; wherein:
the edge system is used for managing at least one edge cluster managed by the application scheduling platform;
the cloud system is configured to implement the steps of the application scheduling method according to any one of claims 1 to 8, based on at least one edge cluster managed by the edge system.
10. A storage medium having stored thereon an application scheduler which, when executed by a processor, carries out the steps of the application scheduling method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110492267.7A CN113312161A (en) | 2021-05-06 | 2021-05-06 | Application scheduling method, platform and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110492267.7A CN113312161A (en) | 2021-05-06 | 2021-05-06 | Application scheduling method, platform and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113312161A true CN113312161A (en) | 2021-08-27 |
Family
ID=77371739
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110492267.7A Pending CN113312161A (en) | 2021-05-06 | 2021-05-06 | Application scheduling method, platform and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113312161A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113553140A (en) * | 2021-09-17 | 2021-10-26 | 阿里云计算有限公司 | Resource scheduling method, equipment and system |
| CN113746676A (en) * | 2021-09-01 | 2021-12-03 | 京东科技信息技术有限公司 | Container cluster-based network card management method, device, equipment, medium and product |
| CN113839814A (en) * | 2021-09-22 | 2021-12-24 | 银河麒麟软件(长沙)有限公司 | Decentralized Kubernetes cluster federal implementation method and system |
| CN114745377A (en) * | 2022-02-14 | 2022-07-12 | 优刻得科技股份有限公司 | Edge cloud cluster service system and implementation method |
| CN117931379A (en) * | 2024-03-21 | 2024-04-26 | 北京火山引擎科技有限公司 | Cluster deployment method, device, system, equipment and medium for edge computing |
| CN114745377B (en) * | 2022-02-14 | 2024-06-07 | 优刻得科技股份有限公司 | Edge cloud cluster service system and implementation method |
-
2021
- 2021-05-06 CN CN202110492267.7A patent/CN113312161A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113746676A (en) * | 2021-09-01 | 2021-12-03 | 京东科技信息技术有限公司 | Container cluster-based network card management method, device, equipment, medium and product |
| CN113746676B (en) * | 2021-09-01 | 2023-09-01 | 京东科技信息技术有限公司 | Network card management method, device, equipment, medium and product based on container cluster |
| CN113553140A (en) * | 2021-09-17 | 2021-10-26 | 阿里云计算有限公司 | Resource scheduling method, equipment and system |
| CN113553140B (en) * | 2021-09-17 | 2022-03-18 | 阿里云计算有限公司 | Resource scheduling method, equipment and system |
| CN113839814A (en) * | 2021-09-22 | 2021-12-24 | 银河麒麟软件(长沙)有限公司 | Decentralized Kubernetes cluster federal implementation method and system |
| CN113839814B (en) * | 2021-09-22 | 2022-02-18 | 银河麒麟软件(长沙)有限公司 | Decentralized Kubernetes cluster federal implementation method and system |
| CN114745377A (en) * | 2022-02-14 | 2022-07-12 | 优刻得科技股份有限公司 | Edge cloud cluster service system and implementation method |
| CN114745377B (en) * | 2022-02-14 | 2024-06-07 | 优刻得科技股份有限公司 | Edge cloud cluster service system and implementation method |
| CN117931379A (en) * | 2024-03-21 | 2024-04-26 | 北京火山引擎科技有限公司 | Cluster deployment method, device, system, equipment and medium for edge computing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113312161A (en) | Application scheduling method, platform and storage medium | |
| US10735345B2 (en) | Orchestrating computing resources between different computing environments | |
| CN107426023B (en) | Cloud platform log collection and forwarding method, system, equipment and storage medium | |
| CN105049268A (en) | Distributed computing resource allocation system and task processing method | |
| US20090282413A1 (en) | Scalable Scheduling of Tasks in Heterogeneous Systems | |
| EP3285187B1 (en) | Optimized merge-sorting of data retrieved from parallel storage units | |
| CN107807815B (en) | Method and device for processing tasks in distributed mode | |
| CN103458270A (en) | Video transcoding system and method | |
| CN112181648B (en) | Configuration-driven lightweight hybrid infrastructure platform and data processing method | |
| CN112835714B (en) | Container arrangement method, system and medium for CPU heterogeneous clusters in cloud edge environment | |
| CN109284180B (en) | Task scheduling method and device, electronic equipment and storage medium | |
| CN111459641B (en) | Method and device for task scheduling and task processing across machine room | |
| CN112565317B (en) | Hybrid cloud system, data processing method and device thereof, and storage medium | |
| EP3018581B1 (en) | Data staging management system | |
| CN111552558A (en) | Scheduling method and device of heterogeneous cloud resources | |
| CN111399764A (en) | Data storage method, data reading device, data storage equipment and data storage medium | |
| CN113468111A (en) | Log monitoring management system and method for container cloud | |
| US11726819B2 (en) | Tool for viewing jobs managed by heterogeneous job schedulers | |
| CN115827250A (en) | Data storage method, device and equipment | |
| CN113535321A (en) | Virtualized container management method, system and storage medium | |
| CN114205641A (en) | Video data processing method and device | |
| CN109684130A (en) | The method and device of data backup between a kind of computer room | |
| CN106843890B (en) | Sensor network, node and operation method thereof based on intelligent decision | |
| US7127446B1 (en) | File system based task queue management | |
| CN110750362A (en) | Method and apparatus for analyzing biological information, 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 |