CN113037794A - Computing resource allocation scheduling method, device and system - Google Patents

Computing resource allocation scheduling method, device and system Download PDF

Info

Publication number
CN113037794A
CN113037794A CN201911355811.2A CN201911355811A CN113037794A CN 113037794 A CN113037794 A CN 113037794A CN 201911355811 A CN201911355811 A CN 201911355811A CN 113037794 A CN113037794 A CN 113037794A
Authority
CN
China
Prior art keywords
application
node
resource
proxy service
event
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.)
Granted
Application number
CN201911355811.2A
Other languages
Chinese (zh)
Other versions
CN113037794B (en
Inventor
刘虢
余万水
黄浩
刘洪政
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mashang Consumer Finance Co Ltd
Original Assignee
Mashang Consumer Finance Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Mashang Consumer Finance Co Ltd filed Critical Mashang Consumer Finance Co Ltd
Priority to CN201911355811.2A priority Critical patent/CN113037794B/en
Publication of CN113037794A publication Critical patent/CN113037794A/en
Application granted granted Critical
Publication of CN113037794B publication Critical patent/CN113037794B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/02—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP]
    • H04L67/025—Protocols based on web technology, e.g. hypertext transfer protocol [HTTP] for remote control or remote monitoring of applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/2866—Architectures; Arrangements
    • H04L67/30—Profiles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the invention provides a method, a device and a system for allocating and scheduling computing resources, wherein the method comprises the following steps: the resource manager judges whether a computing resource configuration event for modifying the application is monitored; if the event of modifying the computing resource configuration of the application is monitored, the resource manager calls a proxy service on a node so as to modify the computing resource configuration of the application through the proxy service; the resource manager acquires a modification result returned by the proxy service for modifying the computing resource configuration of the application; the resource manager operates the data storage system according to the modification result to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof, so that the re-allocation of the computing resources is realized under the condition of not restarting the application example, and the use experience of a user is not influenced.

Description

Computing resource allocation scheduling method, device and system
Technical Field
The embodiment of the invention relates to the technical field of computers, in particular to a method, a device and a system for computing resource configuration scheduling.
Background
Referring to fig. 1, a Vertical Pod auto scaler (VPA) scheme of kubernets (a container cluster management system) is illustrated, which inherits the basic flow of kubernets cluster management that applies resource definition change trigger redeployment (restart).
The method comprises the steps that a recommender in a VPA controller obtains resource usage information from a cluster index server, continuously calculates a recommended value and sends the recommended value to a VPA access controller, the access controller judges whether a condition for triggering and modifying resource definition is met or not according to VPA configuration, the condition for triggering and modifying the resource definition is judged to be met and then the original resource definition is covered, and an updater in the VPA controller monitors that an application instance is changed and restarted to enable configuration change to take effect.
Implementation of the VPA scheme of Kubernetes requires a restart of the application instance. Restarting an application instance may have an impact on the user using the application, for example, restarting an application instance in a production environment may cause a reduction in quality of service, and a test environment may affect the work under test and affect the test results.
Disclosure of Invention
An object of the embodiments of the present invention is to provide a method, an apparatus, and a system for scheduling computing resource allocation, which solve the problem that modifying application computing resource allocation by using a container cluster management system will cause restarting of an application instance.
In a first aspect, an embodiment of the present invention provides a method for computing resource configuration scheduling, which is applied to a resource manager, and includes:
judging whether a computing resource configuration event of an application is monitored to be modified;
if the event of modifying the computing resource configuration of the application is monitored, calling a proxy service on a node to modify the computing resource configuration of the application through the proxy service;
obtaining a modification result returned by the proxy service for modifying the computing resource configuration of the application;
and operating the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof.
Optionally, if the event of modifying the computing resource configuration of the application is monitored, invoking a proxy service on a node to modify the computing resource configuration of the application through the proxy service, including:
determining the name of each first node deployed by the application instance if a computing resource configuration event modifying the application is monitored;
counting currently available computing resources for scheduling on each first node through an application program interface server API server;
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
Optionally, if the event of modifying the computing resource configuration of the application is monitored, invoking a proxy service on a node to modify the computing resource configuration of the application through the proxy service, including:
determining a type of a computing resource configuration event that modifies the application if the event is heard;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
and if the current resources available for scheduling of each node are larger than the added value of the resource request, calling a proxy service on each node, and modifying the control group resource limit value of the application instance container on each node through the proxy service.
Optionally, the operating the data storage system according to the modification result to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof includes:
and calling a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifying the value of the resource request quantity of the application in the key-value-pair data storage system so as to keep the control group resource limit value in each node consistent with the resource limit value stored in the key-value-pair data storage system.
In a second aspect, an embodiment of the present invention provides a resource manager, including:
the judging module is used for judging whether a computing resource configuration event of the modified application is monitored or not;
the calling module is used for calling the proxy service on the node to modify the computing resource configuration of the application through the proxy service if the computing resource configuration event of the application is monitored to be modified;
an obtaining module, configured to obtain a modification result returned by the proxy service for modifying the computing resource configuration of the application;
and the synchronous updating module is used for operating the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result.
Optionally, the invoking module is further configured to:
determining the name of each first node deployed by the application instance if a computing resource configuration event modifying the application is monitored;
counting currently available computing resources for scheduling on each first node through an API server;
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
Optionally, the invoking module is further configured to:
determining a type of a computing resource configuration event that modifies the application if the event is heard;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the current resource available for scheduling of each node is larger than the added value of the resource request, then the proxy service on each node is invoked, and the proxy service modifies the group resource limit value controlled by the application instance container on each node.
Optionally, the synchronization update module is further configured to: and calling a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifying the value of the resource request quantity of the application in the key-value-pair data storage system so as to keep the control group resource limit value in each node consistent with the resource limit value stored in the key-value-pair data storage system.
In a third aspect, an embodiment of the present invention provides a computing resource allocation scheduling system, including: the system comprises a monitoring system, a control system, a container scheduling platform and a resource manager:
the monitoring system sends a request message to a management and control system, wherein the request message is used for requesting to modify the computing resource configuration of the application;
the management and control system informs an API server of the container scheduling platform to modify the computing resource configuration of the application according to the request message;
if the resource manager monitors an event for modifying the computing resource configuration of the application, the resource manager calls proxy service on a node and modifies the computing resource configuration of the application through the proxy service;
and the resource manager operates the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof, wherein the data storage system is used for storing the information of the container scheduling platform.
Optionally, the resource manager determines a name of each first node deployed by the application instance;
the resource manager counts the computing resources currently available for scheduling on each first node through the API server;
if the current available resource for scheduling of each first node is greater than the added value of the resource request, the resource manager invokes a proxy service on each first node, which modifies the group resource limit value controlled by the application instance container on each first node.
Optionally, the proxy service determines process identifiers of the containers running on the nodes according to the identifiers of the applications;
and finding a path corresponding to the corresponding process in the control group through the process identifier, and modifying the content of the control file of the corresponding control group under the path so as to enable the changed resource limit value of the control group to take effect on each node.
Optionally, the resource manager calls a key-value-pair data storage system interface according to a modification result of the computing resource configuration of the application, and modifies a value of the resource request amount of the application in the key-value-pair data storage system, so that the control group resource limit value in each node is consistent with the resource limit value stored in the key-value-pair data storage system.
Optionally, if the resource manager listens for a computing resource configuration event that modifies the application, determining the type of the event;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the current resource available for scheduling of each node is greater than the added value of the resource request, the resource manager invokes a proxy service on each node that modifies the group resource limit value controlled by the application instance container on each node.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements steps including the method for scheduling a computing resource configuration according to the first aspect.
In the embodiment of the invention, the resource limit configuration modification of the control group is realized by calling the proxy service on the node through the resource manager, and simultaneously, the data is written into the data storage system for cluster information storage through the interface, so that the re-allocation of the computing resources is realized under the condition of not restarting the application example, and the use experience of a user is not influenced.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a schematic diagram of the longitudinal expansion of a prior Kubernets;
FIG. 2 is a flowchart of a computing resource allocation scheduling method according to an embodiment of the present invention;
FIG. 3 is a second flowchart of a computing resource allocation scheduling method according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a computing resource allocation schedule according to an embodiment of the present invention;
FIG. 5 is a second illustration of a computing resource allocation schedule according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating a computing resource allocation scheduling system according to an embodiment of the present invention;
FIG. 7 is a diagram of a resource manager according to an embodiment of the invention.
Detailed Description
To facilitate understanding of embodiments of the present invention, the following technical terms are introduced.
(1) Kubernetes: the cross-host cluster open-source container scheduling platform can automatically deploy, expand and operate application containers and provides a container-centered infrastructure.
(2) Kubernets CRD (Custom Resource Definition): the Application Programming Interface (API) is expanded by self-defining the secondary development capability of the resources, a new resource type can be added into the Kubernets API through the CRD without modifying a Kubernets source code to create a self-defined API server (server), and the expansion capability of the Kubernets is greatly improved by the function.
(3) API Server: the kubernets API Server provides HyperText Transfer Protocol (HTTP) representation State Transfer (Rest) interfaces such as addition, deletion, modification, query and monitoring (watch) of various resource objects (e.g., pod, etc.) of kubernets, and the interfaces are data buses and data centers of the whole cluster.
(4) Pod: the smallest unit that can be created and managed in the kubernets system is the smallest resource object model created or deployed by a user in the resource object model, and is also a resource object for running a containerized application on kubernets, other resource objects are all used to support or extend the functions of Pod objects, such as a controller object used to manage Pod objects, a Service or Ingress resource object used to expose Pod reference objects, a PersistentVolume resource object used to provide storage for pods, etc., k8s (an open development platform) does not directly process the pods, and the pods can directly process the pods, the pods are composed of one or more pods, and each Pod can set a quota of computing resources including CPUs and memories.
(5) Cgroups: the control groups (control groups) are a function of the Linux kernel, and are used to limit, control and separate resources (such as CPU, memory, and disk input/output) of a process group.
(6) A container: (container) a portable, lightweight operating system level virtualization technique. It uses namespace (namespace) to isolate different software running environments and by mirroring the running environment of self-contained software, so that the container can be conveniently run anywhere. Compared with the traditional virtual machine technology, because the virtual operating system does not exist, the occupied resource is smaller, and the running efficiency is higher.
(7) Etcd: a highly available, distributed key-value pair data storage system. Kubernetes uses etcd as a back-end store to store state information of all network configurations and objects of the cluster.
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "comprises," "comprising," or any other variation thereof, in the description and claims of this application, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Furthermore, the use of "and/or" in the specification and claims means that at least one of the connected objects, such as a and/or B, means that three cases, a alone, B alone, and both a and B, exist.
In the embodiments of the present invention, words such as "exemplary" or "for example" are used to mean serving as examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "e.g.," an embodiment of the present invention is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.
Referring to fig. 2, an embodiment of the present invention provides a method for scheduling computing resource allocation, where the method includes: step 201, step 202, step 203 and step 204.
Step 201: the monitoring system sends a request message to the management and control system, wherein the request message is used for requesting to modify the computing resource configuration of the application;
in the embodiment of the present invention, the monitoring system is configured to monitor whether it is necessary to trigger modification of the computing resource configuration of the application, for example, when the monitoring system adds a corresponding configuration, and when a certain index exceeds a threshold, the management and control system is invoked to request modification of the computing resource of the application instance container.
In the embodiment of the present invention, the policing system is configured to manage (for example, increase resource limitation or decrease resource limitation) a computing resource on a node (which is a smallest computing hardware unit in the container cluster management system, and may be a physical machine or a virtual machine, for example).
For example, if a monitoring index (for example, the usage rate of a CPU or a memory, etc.) configured in the monitoring system exceeds a threshold, the request message is sent to the management and control system.
In the embodiment of the present invention, the request message may be used to request to decrease the computing resource configuration of the application, or may also request to increase the computing resource configuration of the application.
Step 202: the management and control system informs an application program interface server (API server) of the container scheduling platform to modify the computing resource configuration of the application according to the request message;
in the embodiment of the present invention, the container scheduling platform is used for managing containerized applications on multiple hosts in a cloud platform, for example, the container scheduling platform may be kubernets, but is not limited thereto.
Step 203: if the resource manager monitors the event of modifying the computing resource configuration of the application, the proxy service on the node is called, and the computing resource configuration of the application is modified through the proxy service;
in the embodiment of the present invention, the resource manager may be a custom resource manager, which is used for scheduling cluster resources, and the resource manager may count, through an API server interface, computing resources (for example, CPU resources or memory resources) currently available for scheduling on the application instance deployment node, for example, the resource manager may be a kubernets CRD, but is not limited thereto.
Step 204: and the resource manager operates the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof.
In this embodiment of the present invention, the data storage system is configured to store information of the container scheduling platform, for example, the information may include information (e.g., a limit value) related to a computing resource of the application instance deployment node, for example, the data storage system may be Etcd, but is not limited thereto.
In some embodiments, if a resource manager hears, through the API server, a computing resource configuration event that modifies the application, the resource manager determines a name of each first node deployed by the application instance; the resource manager counts the computing resources currently available for scheduling on each first node through the API server; if the current available resource for scheduling of each first node is greater than the added value of the resource request, the resource manager invokes a proxy service on each first node, which modifies the group resource limit value controlled by the application instance container on each first node.
In some embodiments, if a resource manager hears, through the API server, a computing resource configuration event that modifies the application, then determining a type of the event;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the currently available resources for scheduling of each of the nodes is greater than the added value of the resource request, the resource manager invokes a proxy service on each of the nodes that modifies the group resource limit value controlled on the node by the application instance container.
It is understood that the resource currently available for scheduling by the node is the amount of allocable resource on the node minus the amount of resource requested by each instance of all deployed applications.
In some embodiments, said proxy service modifying application instance container controls a group resource limit value at each of said nodes, comprising: the proxy service determines process identifiers of the containers running on the nodes according to the identifiers of the applications; and finding a path corresponding to the corresponding process in the control group through the process identifier, and modifying the content of the control file of the corresponding control group under the path so as to enable the changed resource limit value of the control group to take effect on each node.
In some embodiments, the resource manager performs a synchronous update on the data storage system according to a modification result of the computing resource configuration of the application, including: the resource manager calls a key-value-pair data storage system interface according to a modification result of the computing resource configuration of the application, and modifies values of resource request amounts of the application in the key-value-pair data storage system (for example, modifies resource limit values of a CPU (central processing unit) and a memory occupied by the application), so that a control group resource limit value in the node is consistent with a resource limit value stored in the key-value-pair data storage system. Therefore, the resource limit of the CPU and the memory occupied by the application can be automatically modified. For a single application, automatic longitudinal scalability is achieved. For the whole cluster, the utilization rate of the computing resources of the whole cluster can be effectively improved.
In the embodiment of the invention, the resource limit configuration modification of the control group is realized by calling the proxy service on the node through the resource manager, and simultaneously, the data is written into the data storage system for cluster information storage through the interface, so that the transparent reallocation of the computing resources is realized under the condition of not restarting the application example, and the use experience of a user is not influenced.
Referring to fig. 3, an embodiment of the present invention provides a method for scheduling computing resource allocation, where an execution subject of the method is a resource manager, and the method includes: step 301, step 302, step 303 and step 304.
Step 301: judging whether a computing resource configuration event for modifying the application is monitored, and executing a step 302 if the computing resource configuration event for modifying the application is monitored; otherwise, the flow ends.
Step 302: invoking a proxy service on a node to modify a computing resource configuration of the application through the proxy service;
step 303: obtaining a modification result returned by the proxy service for modifying the computing resource configuration of the application;
step 304: and operating the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof.
In some embodiments, in step 302, if a computing resource configuration event that modifies the application is monitored, a name of each first node deployed by the application instance is determined;
counting currently available computing resources for scheduling on each first node through an application program interface server (API server);
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
In some embodiments, in step 302, if a computing resource configuration event that modifies the application is monitored, a type of the event is determined;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event of the computing resource limit of the application, confirming a node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the APIserver;
and if the current resources available for scheduling of each node are larger than the added value of the resource request, calling a proxy service on each node, and modifying the control group resource limit value of the application instance container on each node through the proxy service.
In some embodiments, in step 304, a key-value pair data storage system interface is invoked according to a modification result of the computing resource configuration of the application, and a value of a resource request amount of the application in the key-value pair data storage system is modified, so that a control group resource limit value in each node is consistent with a resource limit value stored in the key-value pair data storage system.
In the embodiment of the invention, the resource limit configuration modification of the control group is realized by calling the proxy service on the node through the resource manager, and simultaneously, the data is written into the data storage system for cluster information storage through the interface, so that the transparent reallocation of the computing resources is realized under the condition of not restarting the application example, and the use experience of a user is not influenced.
Referring to fig. 4, when the computing resource of the application needs to be modified, a corresponding configuration may be added to the monitoring system, and when a certain index exceeds a threshold, the management and control system is invoked to request modification of the computing resource of the application instance container.
And after receiving the request, the management and control system calls an API server of a container scheduling platform Kubernetes to inform modification.
The resource manager exists in the form of a kubernets CRD and listens for the above modifications. After monitoring a resource application modification event, a resource manager confirms the names of nodes deployed by each application instance from a resource definition, then counts the currently available scheduling computing resources (such as cpu or memory) on the computing nodes through an API server interface, and directly abandons if the available scheduling resources of any host in the nodes are smaller than the added value of the modified resource request; otherwise, a pre-installed proxy service (or called a proxy program) on each node is called, and the corresponding cgroup resource limit value of the specified container on the node (or called a physical machine) is modified through the proxy service.
If the modification operation of each node is successful, the resource manager calls an etcd interface to modify the value of the resource request quantity corresponding to the POD in the etcd, so that the cgroup resource limit value in the physical machine is consistent with the value stored in the etcd by the corresponding object in the cluster.
Referring to fig. 5, the trigger management and control system modifies the computing resource configuration: request/limit (limit) configuration supporting manual modification of the cpu and memory of the application; the minimum resource requirement of the container is used as the judgment of resource allocation in the container scheduling.
Regarding the request:
(1) the container is allowed to be scheduled to the node only when the resource quantity which can be allocated on the current node is equal to the request;
(2) the request parameter does not limit the maximum available resources of the container.
Regarding Limit:
(1) maximum value of available resources of the container;
(2) the setting of "0" means that the resource to be used is not limited and can be used indefinitely.
Meanwhile, a monitoring notification item can be configured in the monitoring system, for example, when the average cpu utilization rate of a certain application is lower than 1% in a certain period of time, the management and control system is notified, and the value of request/limit of the cpu in the original resource definition of the application is changed to be smaller. The management and control system can determine a recommended value according to the preset configuration and algorithm and inform the recommended value to the API Server of Kubernets.
Monitoring and modifying: a user-defined resource manager is realized in a Kubernets CRD mode, and the user-defined resource manager monitors the event of application computing resource change through an API Server of a Kubernets master control node.
Check feasibility of resource change: if the change is a reduction in resource constraints, go directly to the next step to modify. If the resource limit is increased, whether expansion room exists on each computing node needs to be verified firstly, so that the computing node where each instance of the application is located is confirmed according to the deployment situation of the application in the event in the cluster, the API Server is inquired, the resource quantity applied by all deployed Pod is subtracted from the node allocable resource quantity, and when the value is larger than the increment of the change, the expansion is feasible, and the next actual modification is allowed.
Modifying node (physical machine) configuration through proxy: the resource manager calls the agent service on the node, the agent service determines the process ID number (PID) of the container running on the node (physical machine) according to the application identifier, then finds the corresponding path of the process in the cgroup through the PID, and modifies the content of the corresponding cgroup control file under the path, so that the changed value takes effect on the node (physical machine).
Although the configuration is valid at this time, the kubernets cluster is unaware of the change and needs to continuously modify the data of the relevant pod in the cluster object information store to match the actual situation with the metadata of the cluster management.
The modification is synchronized to etcd: and when the change of each computing node in the previous step is successful, the user-defined resource manager modifies the metadata information of all the Pods applying the change through the etcd interface. Thereby making the validated configuration consistent with the cluster-managed metadata to avoid subsequent negative impact on cluster scheduling.
By combining the monitoring system with the predefined configuration and algorithm, the resource limit of the CPU and the memory occupied by the application can be automatically modified. For a single application, automatic longitudinal scalability is achieved. For the whole cluster, the utilization rate of the computing resources of the whole cluster can be effectively improved.
The method avoids the basic process that the original application resource definition is modified by Kubernetes to cause restarting, realizes the resource limitation configuration modification of the control group by directly calling a user-defined proxy on a physical machine through a user-defined resource manager, and simultaneously writes data into a data storage system for cluster information storage through an interface, realizes the transparent re-allocation of computing resources under the condition of not restarting an application example, and does not influence the use experience of a user.
Referring to fig. 6, an embodiment of the present invention further provides a computing resource configuration scheduling system (alternatively referred to as a cloud computing platform), including: monitoring system 601, management and control system 602, container scheduling platform 603, resource manager 604, data storage system 605 and node 606, optionally, management and control system 602 and resource manager 604 may be deployed on a cloud management and control system;
the monitoring system 601 sends a request message to the management and control system 602, where the request message is used to request to modify the computing resource configuration of the application;
the management and control system 602 notifies the API server of the container scheduling platform 603 to modify the computing resource configuration of the application according to the request message;
if the resource manager 604 monitors an event for modifying the computing resource configuration of the application, invoking a proxy service on a node 606, and modifying the computing resource configuration of the application through the proxy service;
the resource manager 604 operates the data storage system according to the modification result to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof, and the data storage system 605 is used for storing the information of the container scheduling platform.
In some embodiments, the resource manager 604 determines the name of each first node of the application instance deployment; the resource manager 604 counts the currently available computing resources for scheduling on each first node through the API server; if the currently available resources for scheduling at each of the first nodes is greater than the incremental value of the resource request, the resource manager 604 invokes a proxy service at each of the first nodes that modifies the group resource limit value controlled at each of the first nodes by the application instance container.
In some embodiments, the proxy service determines process identities for containers running on each of the nodes based on the identity of the application; and finding a path corresponding to the corresponding process in the control group through the process identifier, and modifying the content of the control file of the corresponding control group under the path so as to enable the changed resource limit value of the control group to take effect on each node.
In some embodiments, the resource manager 604 invokes a key-value pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifies the value of the resource request amount of the application in the key-value pair data storage system, so that the control group resource limit value in each of the nodes and the resource limit value stored in the key-value pair data storage system are consistent.
In some embodiments, if the resource manager 604 listens for a computing resource configuration event that modifies the application, then the type of the event is determined;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the currently available resources for scheduling for each of the nodes is greater than the incremental value of the resource request, the resource manager 604 invokes a proxy service on each of the nodes that modifies the group resource limit value controlled by the application instance container on each of the nodes.
The computing resource allocation scheduling system provided in the embodiment of the present invention may implement the method embodiments described above, and the implementation principle and technical effect are similar, which are not described herein again.
Referring to fig. 7, an embodiment of the present invention further provides a resource manager, where the resource manager 700 includes:
a determining module 701, configured to determine whether a computing resource configuration event of an application is monitored;
a calling module 702, configured to, if a computing resource configuration event of the application is monitored to be modified, call a proxy service on a node to modify the computing resource configuration of the application through the proxy service;
an obtaining module 703, configured to obtain a modification result returned by the proxy service, where the modification result modifies the computing resource configuration of the application;
and a synchronous updating module 704, configured to operate the data storage system according to the modification result, so as to synchronously update and store the configuration scheduling information that is consistent with the current modification and the modification result.
In some embodiments, the invoking module 702 is further configured to:
determining the name of each first node deployed by the application instance if a computing resource configuration event modifying the application is monitored;
counting currently available computing resources for scheduling on each first node through an API server;
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
In some embodiments, the invoking module 702 is further configured to:
determining a type of a computing resource configuration event that modifies the application if the event is heard;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the current resource available for scheduling of each node is larger than the added value of the resource request, then the proxy service on each node is invoked, and the proxy service modifies the group resource limit value controlled by the application instance container on each node.
In some embodiments, the synchronization update module 704 is further configured to: and calling a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifying the value of the resource request quantity of the application in the key-value-pair data storage system so as to keep the control group resource limit value in each node consistent with the resource limit value stored in the key-value-pair data storage system.
The resource manager provided in the embodiment of the present invention may execute the method embodiments described above, and the implementation principle and the technical effect are similar, which are not described herein again.
The 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 computer program implements each process of the foregoing method for scheduling a computing resource allocation, and can achieve the same technical effect, and in order to avoid repetition, the details are not repeated here. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.
The steps of a method or algorithm described in connection with the disclosure herein may be embodied in hardware or may be embodied in software instructions executed by a processor. The software instructions may consist of corresponding software modules that may be stored in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, a removable hard disk, a compact disk, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an ASIC. In addition, the ASIC may be carried in a core network interface device. Of course, the processor and the storage medium may reside as discrete components in a core network interface device.
Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in this invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
It will be apparent to those skilled in the art that various changes and modifications may be made in the embodiments of the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to encompass such modifications and variations.

Claims (14)

1. A computing resource allocation scheduling method is applied to a resource manager and is characterized by comprising the following steps:
judging whether a computing resource configuration event of an application is monitored to be modified;
if the event of modifying the computing resource configuration of the application is monitored, calling a proxy service on a node to modify the computing resource configuration of the application through the proxy service;
obtaining a modification result returned by the proxy service for modifying the computing resource configuration of the application;
and operating the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof.
2. The method of claim 1, wherein invoking a proxy service on a node to modify the computing resource configuration of the application via the proxy service if the monitoring of the event modifying the computing resource configuration of the application comprises:
determining the name of each first node deployed by the application instance if a computing resource configuration event modifying the application is monitored;
counting currently available computing resources for scheduling on each first node through an application program interface server API server;
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
3. The method of claim 1, wherein invoking a proxy service on a node to modify the computing resource configuration of the application via the proxy service if the monitoring of the event modifying the computing resource configuration of the application comprises:
determining a type of a computing resource configuration event that modifies the application if the event is heard;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
and if the current resources available for scheduling of each node are larger than the added value of the resource request, calling a proxy service on each node, and modifying the control group resource limit value of the application instance container on each node through the proxy service.
4. The method according to claim 1, wherein the operating the data storage system according to the modification result to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof comprises:
and calling a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifying the value of the resource request quantity of the application in the key-value-pair data storage system so as to keep the control group resource limit value in each node consistent with the resource limit value stored in the key-value-pair data storage system.
5. A resource manager, comprising:
the judging module is used for judging whether a computing resource configuration event of the modified application is monitored or not;
the calling module is used for calling the proxy service on the node to modify the computing resource configuration of the application through the proxy service if the computing resource configuration event of the application is monitored to be modified;
an obtaining module, configured to obtain a modification result returned by the proxy service for modifying the computing resource configuration of the application;
and the synchronous updating module is used for operating the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result.
6. The resource manager of claim 5, wherein the calling module is further configured to:
determining the name of each first node deployed by the application instance if a computing resource configuration event modifying the application is monitored;
counting currently available computing resources for scheduling on each first node through an API server;
if the current resource available for scheduling of each first node is larger than the added value of the resource request, invoking the proxy service on each first node, and modifying the control group resource limit value of the application instance container on each first node through the proxy service.
7. The resource manager of claim 5, wherein the calling module is further configured to:
determining a type of a computing resource configuration event that modifies the application if the event is heard;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the current resource available for scheduling of each node is larger than the added value of the resource request, then the proxy service on each node is invoked, and the proxy service modifies the group resource limit value controlled by the application instance container on each node.
8. The resource manager of claim 5, wherein the synchronization update module is further configured to: and calling a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifying the value of the resource request quantity of the application in the key-value-pair data storage system so as to keep the control group resource limit value in each node consistent with the resource limit value stored in the key-value-pair data storage system.
9. A computing resource allocation scheduling system, comprising: the system comprises a monitoring system, a control system, a container scheduling platform and a resource manager:
the monitoring system sends a request message to a management and control system, wherein the request message is used for requesting to modify the computing resource configuration of the application;
the management and control system informs an API server of the container scheduling platform to modify the computing resource configuration of the application according to the request message;
if the resource manager monitors an event for modifying the computing resource configuration of the application, the resource manager calls proxy service on a node and modifies the computing resource configuration of the application through the proxy service;
and the resource manager operates the data storage system according to the modification result so as to synchronously update and store the configuration scheduling information consistent with the current modification and the modification result thereof, wherein the data storage system is used for storing the information of the container scheduling platform.
10. The system of claim 9, wherein:
the resource manager determines the name of each first node deployed by the application instance;
the resource manager counts the computing resources currently available for scheduling on each first node through the API server;
if the current available resource for scheduling of each first node is greater than the added value of the resource request, the resource manager invokes a proxy service on each first node, which modifies the group resource limit value controlled by the application instance container on each first node.
11. The system of claim 10, wherein:
the proxy service determines process identifiers of the containers running on the nodes according to the identifiers of the applications;
and finding a path corresponding to the corresponding process in the control group through the process identifier, and modifying the content of the control file of the corresponding control group under the path so as to enable the changed resource limit value of the control group to take effect on each node.
12. The system of claim 11, wherein:
and the resource manager calls a key-value-pair data storage system interface according to the modification result of the computing resource configuration of the application, and modifies the value of the resource request quantity of the application in the key-value-pair data storage system, so that the resource limit value of the control group in each node is consistent with the resource limit value stored in the key-value-pair data storage system.
13. The system of claim 12, wherein:
if the resource manager monitors a computing resource configuration event for modifying the application, determining the type of the event;
if the type of the event is the reduction event of the computing resource limit of the application, calling proxy service on each node, and modifying the computing resource configuration of the application through the proxy service;
if the type of the event is an increasing event limited by the computing resources of the application, confirming the node where each instance of the application is located according to the deployment condition of the application in the cluster, and counting the computing resources currently available for scheduling on each node through the API server;
if the current resource available for scheduling of each node is greater than the added value of the resource request, the resource manager invokes a proxy service on each node that modifies the group resource limit value controlled by the application instance container on each node.
14. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out steps comprising the method for computing resource configuration scheduling according to any one of claims 1 to 4.
CN201911355811.2A 2019-12-25 2019-12-25 Method, device and system for computing resource allocation scheduling Active CN113037794B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911355811.2A CN113037794B (en) 2019-12-25 2019-12-25 Method, device and system for computing resource allocation scheduling

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911355811.2A CN113037794B (en) 2019-12-25 2019-12-25 Method, device and system for computing resource allocation scheduling

Publications (2)

Publication Number Publication Date
CN113037794A true CN113037794A (en) 2021-06-25
CN113037794B CN113037794B (en) 2023-04-18

Family

ID=76458074

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911355811.2A Active CN113037794B (en) 2019-12-25 2019-12-25 Method, device and system for computing resource allocation scheduling

Country Status (1)

Country Link
CN (1) CN113037794B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113821306A (en) * 2021-09-24 2021-12-21 湖北亿咖通科技有限公司 Resource isolation method and device and terminal equipment
CN114138421A (en) * 2021-12-08 2022-03-04 兴业银行股份有限公司 Intelligent resource control system and method for kubernets
CN114666102A (en) * 2022-03-03 2022-06-24 华云数据控股集团有限公司 Resource permission method and system
CN115145690A (en) * 2022-06-30 2022-10-04 小米汽车科技有限公司 Cluster resource expansion method and device and container cluster management system
WO2023098614A1 (en) * 2021-11-30 2023-06-08 华为技术有限公司 Cloud instance capacity expansion/reduction method and related device therefor
CN116302339A (en) * 2023-03-09 2023-06-23 上海道客网络科技有限公司 Container group in-situ expansion and contraction method and system based on container cloud platform
CN116974703A (en) * 2023-09-22 2023-10-31 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) Kubernetes application resource management method and system
CN117032950A (en) * 2023-07-10 2023-11-10 企迈科技有限公司 Real-time data transparent transmission method and system based on log

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030041093A1 (en) * 1998-05-29 2003-02-27 Stanley Yamane Web server content replication
CN103618644A (en) * 2013-11-26 2014-03-05 曙光信息产业股份有限公司 Distributed monitoring system based on hadoop cluster and method thereof
CN105765556A (en) * 2013-11-25 2016-07-13 亚马逊科技公司 Customer-directed networking limits in distributed systems
CN106603631A (en) * 2016-11-17 2017-04-26 北京奇艺世纪科技有限公司 Distributed message system and expansion method thereof
CN106790565A (en) * 2016-12-27 2017-05-31 中国电子科技集团公司第五十二研究所 A kind of network attached storage group system
US10084826B1 (en) * 2018-05-14 2018-09-25 Xage Security, Inc. Protocol agnostic security by using out-of-band health check
CN108829509A (en) * 2018-05-03 2018-11-16 山东汇贸电子口岸有限公司 Distributed container cluster framework resources management method based on domestic CPU and operating system
CN109739640A (en) * 2018-12-13 2019-05-10 北京计算机技术及应用研究所 A kind of container resource management system based on Shen prestige framework
WO2019149072A1 (en) * 2018-01-30 2019-08-08 华为技术有限公司 Synchronous signal configuration method and apparatus
CN110417870A (en) * 2019-07-05 2019-11-05 北京达佳互联信息技术有限公司 configuration file management method, device, equipment and storage medium

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030041093A1 (en) * 1998-05-29 2003-02-27 Stanley Yamane Web server content replication
CN105765556A (en) * 2013-11-25 2016-07-13 亚马逊科技公司 Customer-directed networking limits in distributed systems
CN103618644A (en) * 2013-11-26 2014-03-05 曙光信息产业股份有限公司 Distributed monitoring system based on hadoop cluster and method thereof
CN106603631A (en) * 2016-11-17 2017-04-26 北京奇艺世纪科技有限公司 Distributed message system and expansion method thereof
CN106790565A (en) * 2016-12-27 2017-05-31 中国电子科技集团公司第五十二研究所 A kind of network attached storage group system
WO2019149072A1 (en) * 2018-01-30 2019-08-08 华为技术有限公司 Synchronous signal configuration method and apparatus
CN108829509A (en) * 2018-05-03 2018-11-16 山东汇贸电子口岸有限公司 Distributed container cluster framework resources management method based on domestic CPU and operating system
US10084826B1 (en) * 2018-05-14 2018-09-25 Xage Security, Inc. Protocol agnostic security by using out-of-band health check
CN109739640A (en) * 2018-12-13 2019-05-10 北京计算机技术及应用研究所 A kind of container resource management system based on Shen prestige framework
CN110417870A (en) * 2019-07-05 2019-11-05 北京达佳互联信息技术有限公司 configuration file management method, device, equipment and storage medium

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MUHAMMAD ALI MUMTAZ: "Analysis of Timing Synchronization Techniques in OFDM for SDR waveform: Performance Comparison", 《2018 21ST INTERNATIONAL SYMPOSIUM ON WIRELESS PERSONAL MULTIMEDIA COMMUNICATIONS (WPMC)》 *
翟雅荣: "基于容器的分布式融合通信服务器的设计与实现", 《信息科技辑》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113821306A (en) * 2021-09-24 2021-12-21 湖北亿咖通科技有限公司 Resource isolation method and device and terminal equipment
CN113821306B (en) * 2021-09-24 2023-09-05 亿咖通(湖北)技术有限公司 Resource isolation method, device and terminal equipment
WO2023098614A1 (en) * 2021-11-30 2023-06-08 华为技术有限公司 Cloud instance capacity expansion/reduction method and related device therefor
CN114138421A (en) * 2021-12-08 2022-03-04 兴业银行股份有限公司 Intelligent resource control system and method for kubernets
CN114666102A (en) * 2022-03-03 2022-06-24 华云数据控股集团有限公司 Resource permission method and system
CN114666102B (en) * 2022-03-03 2024-02-06 华云数据控股集团有限公司 Resource licensing method and system
CN115145690A (en) * 2022-06-30 2022-10-04 小米汽车科技有限公司 Cluster resource expansion method and device and container cluster management system
CN115145690B (en) * 2022-06-30 2023-08-04 小米汽车科技有限公司 Cluster resource expansion method and device and container cluster management system
CN116302339A (en) * 2023-03-09 2023-06-23 上海道客网络科技有限公司 Container group in-situ expansion and contraction method and system based on container cloud platform
CN117032950A (en) * 2023-07-10 2023-11-10 企迈科技有限公司 Real-time data transparent transmission method and system based on log
CN116974703A (en) * 2023-09-22 2023-10-31 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) Kubernetes application resource management method and system
CN116974703B (en) * 2023-09-22 2024-01-02 山东未来网络研究院(紫金山实验室工业互联网创新应用基地) Kubernetes application resource management method and system

Also Published As

Publication number Publication date
CN113037794B (en) 2023-04-18

Similar Documents

Publication Publication Date Title
CN113037794B (en) Method, device and system for computing resource allocation scheduling
US11704144B2 (en) Creating virtual machine groups based on request
US11553034B2 (en) Server computer management system for supporting highly available virtual desktops of multiple different tenants
CN108293041B (en) Distributed system, resource container allocation method, resource manager and application controller
US20170031622A1 (en) Methods for allocating storage cluster hardware resources and devices thereof
CN105159775A (en) Load balancer based management system and management method for cloud computing data center
WO2017161979A1 (en) Cloud platform-based method and device for managing servers
CN109992373B (en) Resource scheduling method, information management method and device and task deployment system
US20220283846A1 (en) Pod deployment method and apparatus
EP3442201B1 (en) Cloud platform construction method and cloud platform
CN113204353B (en) Big data platform assembly deployment method and device
CN113382077B (en) Micro-service scheduling method, micro-service scheduling device, computer equipment and storage medium
CN111092921A (en) Data acquisition method, device and storage medium
JP2019008454A (en) Information processing system and resource allocation method
CN110569104A (en) Management method and computer storage medium for task training in deep learning system
CN109005071B (en) Decision deployment method and scheduling equipment
CN110968406B (en) Method, device, storage medium and processor for processing task
CN110308914B (en) Upgrade processing method, device, equipment, system and computer readable storage medium
CN114257505B (en) Server node configuration method, device, equipment and storage medium
CN114465958B (en) Input and output control method, device and medium
CN113873052B (en) Domain name resolution method, device and equipment of Kubernetes cluster
CN113900807A (en) Port allocation method, device and system
CN117435302A (en) Container capacity adjustment method, device, electronic equipment and storage medium
CN114584625A (en) Message processing method and device, electronic equipment and storage medium
KR20220064785A (en) Apparatus and method for neural network group connection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant