WO2018121201A1 - 分布式集群服务结构及节点协同方法和装置、终端及介质 - Google Patents

分布式集群服务结构及节点协同方法和装置、终端及介质 Download PDF

Info

Publication number
WO2018121201A1
WO2018121201A1 PCT/CN2017/114855 CN2017114855W WO2018121201A1 WO 2018121201 A1 WO2018121201 A1 WO 2018121201A1 CN 2017114855 W CN2017114855 W CN 2017114855W WO 2018121201 A1 WO2018121201 A1 WO 2018121201A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
stacking
proxy
proxy node
cluster
Prior art date
Application number
PCT/CN2017/114855
Other languages
English (en)
French (fr)
Inventor
钟斌
Original Assignee
深圳云天励飞技术有限公司
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 深圳云天励飞技术有限公司 filed Critical 深圳云天励飞技术有限公司
Publication of WO2018121201A1 publication Critical patent/WO2018121201A1/zh

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
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • 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
    • H04L67/50—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

Definitions

  • the present invention relates to the field of server clusters, and in particular, to a distributed cluster service structure, a node collaboration method and device, a terminal, and a medium.
  • the embodiments of the present invention provide a distributed cluster service structure, a node collaboration method and device, a terminal, and a medium, which can ensure that the system has excellent self-organization and scalability, and at the same time, implement high service availability and service balance.
  • the first aspect of the embodiment of the present invention discloses a distributed cluster service structure, where the distributed cluster service structure is a multi-level sub-center structure, and the distributed cluster service structure includes: a cluster working node and a stacking proxy node, A stacking proxy node acts as a proxy for the cluster working node, where:
  • the cluster working node is configured to determine, by using a protocol and a process, a stacking proxy node to which the cluster working node is attached;
  • the stacking proxy node is configured to manage all cluster working nodes that are proxied by the stacking proxy node.
  • the stacking proxy node includes a root stacking proxy node, and the root stacking proxy node is configured to maintain the distributed cluster service structure.
  • the stacking proxy node is further configured to perform an election process, a topology update process, and a service route maintenance process of the root stack proxy node.
  • the interface of the stacking proxy node is a superset of the interface of the cluster working node, and the interface of the stacking proxy node includes a cluster stacking proxy interface. Interface with the worker node interface.
  • the distributed cluster service structure is a self-organizing and self-healing dynamic cluster.
  • a second aspect of the embodiments of the present invention discloses a node collaboration method based on a distributed cluster service structure, including:
  • the service routing formation process of the allocation from the cluster working node to the stacking proxy node is coordinated by the root stacking proxy node.
  • the selecting the root stacking proxy node from the multiple stacking proxy nodes included in the distributed cluster service structure includes:
  • the stacking proxy node having the strongest server management capability is determined as the root stacking proxy node.
  • the method further includes:
  • the stacking proxy node with the smallest identification number is determined as the root stacking proxy node among the plurality of stacked proxy nodes with the strongest server management capability.
  • each of the stacked proxy nodes in the distributed cluster service structure is determined, and the parent node of the stacking proxy node is a stacking proxy node with optimal server management capability and lifetime TTL distance.
  • each of the cluster working nodes in the distributed cluster service structure determines the cluster A host proxy node of the working node, the host proxy node is a stacking proxy node to which the cluster working node is attached, and the host proxy node has the highest service management capability when serving multiple cluster working nodes.
  • a third aspect of the embodiments of the present invention discloses a node collaboration device of a distributed cluster service structure, where the node collaboration device includes: an election unit, configured to elect a root from multiple stacked proxy nodes included in the distributed cluster service structure. a stacking proxy node; a first execution unit, configured to perform a topology forming process of the multiple stacking proxy nodes by using the root stacking proxy node; and a second executing unit, configured to pass The root stacking proxy node coordinates the process of forming a service route from the cluster working node to the stacking proxy node.
  • a fourth aspect of the embodiments of the present invention discloses a terminal, where the terminal includes a processor, and the processor implements the node collaboration method when executing a computer program stored in a memory.
  • a fifth aspect of the embodiments of the present invention discloses a computer readable storage medium having computer instructions stored thereon, the computer instructions being implemented by a processor to implement the node cooperation method.
  • the embodiment of the invention has the following beneficial effects:
  • the distributed cluster service structure is a multi-level sub-center structure, which can meet the locality, scalability, and self-organization requirements of services in a cross-region deployment situation.
  • the stacking proxy node acts as a proxy for the cluster working node, and provides a self-organizing mechanism of the cluster through the stacking proxy node, thereby realizing high scalability and high availability of the cluster.
  • a root stacking proxy node is selected from a plurality of stacked proxy nodes included in the distributed cluster service structure, and a topology stacking process of multiple stacking proxy nodes is performed through the root stacking proxy node, and the slave clustering node is coordinated by the root stacking proxy node.
  • the coordinated work of the assigned service route formation process to the stacking agent node realizes high availability and service balance of the service.
  • FIG. 1 is a schematic structural diagram of a distributed cluster service structure according to an embodiment of the present invention.
  • FIG. 2 is a structural diagram of a relationship between a stacking proxy node and a cluster working node according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of a node collaboration method based on a distributed cluster service structure according to an embodiment of the present invention
  • FIG. 4 is a schematic structural diagram of a node cooperation apparatus according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
  • the embodiment of the invention discloses a distributed cluster service structure and a node cooperation method and device, a terminal and a medium, which can ensure the system has excellent self-organization and scalability, and at the same time realize high availability and service balance of the service. The details are described below separately.
  • FIG. 1 is a schematic structural diagram of a distributed cluster service structure according to an embodiment of the present invention.
  • the distributed cluster service structure is a multi-level sub-center structure, and the distributed cluster service node includes a cluster working node and a stacking proxy node, for example, a root stacking proxy node, and a cluster working node in layer 0. , stacking proxy node 0 and stacking proxy node 1, stacking proxy node 2 in layer 1, cluster working node 2, cluster working node 3, and stacking proxy node 3, cluster working node (N-1) in layer 2, The cluster working node 1, the cluster working node N, and the stacking proxy node 4.
  • the stacking proxy node acts as a proxy for the cluster working node.
  • the cluster working node is configured to determine, by using a protocol and a process, a stacking proxy node to which the cluster working node is attached;
  • the stacking proxy node is configured to manage all cluster working nodes that are proxied by the stacking proxy node.
  • the stacking proxy node to which the cluster working node 0 is attached is the root stacking proxy node
  • the stacking proxy node to which the cluster working node 1 is attached is the stacking proxy node 2
  • the stacking proxy node to which the cluster working node 2 is attached is the stacking proxy node.
  • the stacking proxy node to which the cluster working node 3 is attached is the stacking proxy node 1
  • the stacking proxy node to which the cluster working node (N-1) is attached is the stacking proxy node 2
  • the stacking proxy nodes to which the cluster working node N is attached are stacked. Proxy node 3.
  • the stacking proxy node can manage all the cluster working nodes that it is proxying. In addition, the stacking proxy node can also manage the stacking sub-proxy nodes. For example, the stacking agent node 1 can manage the cluster working node 3 and the stacking agent node 3.
  • the system can access the entire cluster through the root stacking proxy node.
  • the root stacking proxy node is used to maintain the distributed cluster service structure.
  • the distributed cluster service structure is a self-organizing and self-healing dynamic cluster.
  • the stacking proxy node is further configured to perform an election process, a topology update process, and a service route maintenance process of the root stack proxy node. It can achieve high availability and business balance of services.
  • the interface of the stacking proxy node is a superset of the interface of the cluster working node, where
  • the interface of the stacking proxy node includes a cluster stacking proxy interface and a working node interface proxy interface.
  • FIG. 2 is a structural diagram of a relationship between a stacking proxy node and a cluster working node according to an embodiment of the present invention.
  • the interface of the stacking proxy node includes a cluster stacking proxy interface and a working node interface proxy interface, and the stacking proxy node may perform an election process of the root stacking proxy node, a topology update process, and a service routing maintenance process.
  • the interface of the cluster working node includes a working node interface. It can be seen that the interface of the stacking proxy node is a superset of the interface of the cluster working node.
  • the system can be self-organizing, scalable, and business balanced.
  • the distributed cluster service structure is a multi-level sub-center structure, which can meet the locality, scalability, and self-organization requirements of services in a cross-region deployment situation.
  • the stacking proxy node acts as a proxy for the cluster working node, and provides a self-organizing mechanism of the cluster through the stacking proxy node, thereby realizing high scalability and high availability of the cluster.
  • a root stacking proxy node is selected from a plurality of stacked proxy nodes included in the distributed cluster service structure, and a topology stacking process of multiple stacking proxy nodes is performed through the root stacking proxy node, and the slave clustering node is coordinated by the root stacking proxy node.
  • the coordinated work of the assigned service route formation process to the stacking agent node realizes high availability and service balance of the service.
  • the distributed cluster service structure is stackable, and can utilize the locality of data processing and service access, and eliminate unnecessary IO and information interaction overhead of unnecessary network/storage/control.
  • a unified access interface is designed between the subsystems to support hybrid stacking between different components at different levels, simplifying system deployment and maintenance.
  • FIG. 3 is a schematic flowchart diagram of a node collaboration method based on a distributed cluster service structure according to an embodiment of the present invention. As shown in FIG. 3, the method can include the following steps.
  • 301 Elect a root stack proxy node from a plurality of stacked proxy nodes included in the distributed cluster service structure.
  • all the stacked proxy nodes are peer-to-peer relationships. This peer-to-peer relationship is reflected in: in the initial process of the cluster, any stacking proxy node has the possibility of becoming a root stacking proxy node. When the current root stack proxy node exits the service abnormally, during the update process of the stack cluster, the remaining normal stack proxy nodes may have the potential to become the new root stack proxy node again.
  • Stacking proxy nodes include:
  • the stacking proxy node having the strongest server management capability is determined as the root stacking proxy node.
  • the server management capabilities (ie, Capability) of the plurality of stacked proxy nodes included in the distributed cluster service structure may be acquired in advance.
  • factors such as CPU/GPU/memory/network IO/storage IO/storage capacity and so on all characterize the actual service capabilities of a server node.
  • the following formula expresses the capabilities of the node:
  • Capability(x) ⁇ SubCapality(n) ⁇ Weight(n)
  • Capability(x) represents the server management capability of the xth stacking proxy node
  • SubCapality(n) is the value of each subkey
  • Weight(n) is the weight of each subkey (ie, the degree of importance).
  • the root stacking proxy node broadcasts the first message to the system according to the first period.
  • the non-root stacking proxy node in the system broadcasts the second message to the system according to the second period.
  • the first period and the second period may be any time specified by the system, and the first period and the second period may be the same or different.
  • the first message may be a root node notification message (ie, a RootNotify message)
  • the second message may be For the capability notification message (ie, CapabilityNotify message)
  • both the first message and the second message carry the identity number of the stacking proxy node and the server management capability value.
  • N first first period of message monitoring is performed.
  • the new stacking proxy node may collect two kinds of messages: the first message and the second message. If the new stacking agent node collects the first message during the first period, it indicates that the election of the root stacking proxy node in the system has been completed. At this point, the new stacking proxy node chooses to temporarily accept the election result, joins the system, and starts sending its second message.
  • the current root stacking proxy node broadcasts a re-election notification (ie, a RerootRequest message) to the system according to the third period to query Whether other stacking proxy nodes in the system need to re-elect the root stacking proxy node.
  • a re-election notification ie, a RerootRequest message
  • the new stacking proxy node After receiving the re-election notification, the new stacking proxy node determines that the new stacking proxy node is more suitable to be the new root stacking proxy node, and responds to the current root stacking proxy node in response to the re-election notification, and the current root-stacking proxy node receives the re-election
  • the first message is stopped, and the second message is sent, when there is no stacking proxy node in the system to send the first message.
  • the new stacking proxy node with the highest server management capability stops sending the second message and starts to send the first message.
  • the third period may be any time specified by the system.
  • the method further includes:
  • the stacking proxy node with the smallest identification number is determined as the root stacking proxy node among the plurality of stacked proxy nodes with the strongest server management capability.
  • the stacking proxy node with the smallest identification number will be determined as the root stacking proxy node.
  • the topology forming process of the stack proxy node needs to be started.
  • the root stacking proxy node broadcasts topology establishment information (ie, TopologyBuild message) to the system, and other stacking proxy nodes broadcast topology waiting information (ie, TopologyWait message) to the system.
  • topology establishment information ie, TopologyBuild message
  • other stacking proxy nodes broadcast topology waiting information (ie, TopologyWait message) to the system.
  • topology establishment information ie, TopologyBuild message
  • topology waiting information ie, TopologyWait message
  • topology establishment information may be sent to nodes other than the root stack proxy node and the parent node.
  • each of the stacked proxy nodes in the distributed cluster service structure determines a parent node of the stacking proxy node, and a parent node of the stacking proxy node A stacking proxy node that is optimal for server management capabilities and lifetime TTL distance.
  • the root stacking proxy node broadcasts route notification information (ie, RouteNotice message) to the system to obtain information of all cluster working nodes.
  • the root stacking proxy node sends a routing token (ie, a RouteToken token) to any one of the cluster working nodes.
  • a routing token ie, a RouteToken token
  • the cluster working node calculates the remaining servers of the stacked proxy node from the agent information collected by itself. Management ability value, the formula is as follows:
  • Capabilityleft Capability- ⁇ UsedCapability(n)
  • the meaning of the formula is the value of the server management capability of the stacking proxy node itself minus the value of the server management capability required by the clustering node after serving a certain cluster working node, that is, the remaining Capability.
  • the highest compute is the host proxy node for the cluster worker node.
  • each of the cluster working nodes in the distributed cluster service structure determines a host proxy node of the cluster working node, and the host proxy node works for the cluster.
  • a stacking proxy node to which a node is attached, and the host proxy node has the highest service management capability when serving multiple cluster working nodes.
  • the distributed cluster service structure is a multi-level sub-center structure, which can meet the locality, scalability, and self-organization requirements of services in a cross-region deployment situation.
  • the stacking proxy node acts as a proxy for the cluster working node, and provides a self-organizing mechanism of the cluster through the stacking proxy node, thereby realizing high scalability and high availability of the cluster.
  • a root stacking proxy node is selected from a plurality of stacked proxy nodes included in the distributed cluster service structure, and a topology stacking process of multiple stacking proxy nodes is performed through the root stacking proxy node, and the slave clustering node is coordinated by the root stacking proxy node.
  • the coordinated work of the assigned service route formation process to the stacking agent node realizes high availability and service balance of the service.
  • the distributed cluster service structure is stackable, and can utilize the locality of data processing and service access, and eliminate unnecessary IO and information interaction overhead of unnecessary network/storage/control.
  • a unified access interface is designed between the subsystems to support hybrid stacking between different components at different levels, simplifying system deployment and maintenance.
  • FIG. 4 is a schematic structural diagram of a node cooperation apparatus according to an embodiment of the present invention.
  • the node collaboration device may be used to perform the node collaboration method based on the distributed cluster service structure described in FIG. 2 .
  • the node cooperation device includes:
  • the election unit 401 is configured to elect a root stacking proxy node from multiple stacked proxy nodes included in the distributed cluster service structure;
  • the first execution unit 402 is configured to perform a topology forming process of the multiple stacked proxy nodes by using the root stacking proxy node;
  • the second executing unit 403 is configured to coordinate, by the root stacking proxy node, a service routing forming process for performing allocation from the cluster working node to the stacking proxy node.
  • the election unit is configured from multiple stacked proxy nodes included in the distributed cluster service structure.
  • the way to elect a root stack proxy node is as follows:
  • the stacking proxy node having the strongest server management capability is determined as the root stacking proxy node.
  • the manner in which the election unit elects a root stacking proxy node from multiple stacking proxy nodes included in the distributed cluster service structure is specifically:
  • the stacking proxy node with the smallest identification number is determined as the root stacking proxy node among the plurality of stacked proxy nodes with the strongest server management capability.
  • each of the stacked proxy nodes in the distributed cluster service structure determines a parent node of the stacking proxy node, and the stacking proxy node
  • the parent node is a stacking proxy node with optimal server management capabilities and a lifetime TTL distance.
  • each of the cluster working nodes in the distributed cluster service structure determines a host proxy node of the cluster working node, and the host proxy node works for the cluster.
  • a stacking proxy node to which a node is attached, and the host proxy node has the highest service management capability when serving multiple cluster working nodes.
  • the distributed cluster service structure is a multi-level sub-center structure, which can meet the locality, scalability, and self-organization requirements of services in a cross-region deployment situation.
  • the stacking proxy node acts as a proxy for the cluster working node, and provides a self-organizing mechanism of the cluster through the stacking proxy node, thereby realizing high scalability and high availability of the cluster.
  • a root stacking proxy node is selected from a plurality of stacked proxy nodes included in the distributed cluster service structure, and a topology stacking process of multiple stacking proxy nodes is performed through the root stacking proxy node, and the slave clustering node is coordinated by the root stacking proxy node.
  • the coordinated work of the assigned service route formation process to the stacking agent node realizes high availability and service balance of the service.
  • the distributed cluster service structure is stackable, and can utilize the locality of data processing and service access, and eliminate unnecessary IO and information interaction overhead of unnecessary network/storage/control.
  • a unified access interface is designed between the subsystems to support hybrid stacking between different components at different levels, simplifying system deployment and maintenance.
  • the disclosed apparatus may be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of cells is only a logical function division. In actual implementation, there may be another division manner.
  • multiple units or components may be combined or may be integrated into Another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be electrical or otherwise.
  • the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, can be stored in a computer readable memory.
  • the technical solution of the present invention may contribute to the prior art or all or part of the technical solution may be embodied in the form of a software product stored in a memory.
  • a number of instructions are included to cause a computer device (which may be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention.
  • the foregoing memory includes: U disk, read-only memory (ROM), random access memory (RAM), random access memory (RAM), mobile hard disk, magnetic disk or optical disk, and the like, which can store program code.
  • the terminal in FIG. 5 also includes a processor for performing calculations stored in the memory Machine program to implement the node collaboration method.
  • ROM Read-Only Memory
  • RAM Random Access Memory

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer And Data Communications (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明实施例公开了一种分布式集群服务结构及节点协同方法和装置、终端及介质。其中，所述分布式集群服务结构为多层次分中心结构，所述分布式集群服务结构包括：集群工作节点和堆叠代理节点，所述堆叠代理节点作为所述集群工作节点的代理，其中：所述集群工作节点，用于通过协议和流程确定所述集群工作节点所挂靠的堆叠代理节点；所述堆叠代理节点，用于管理所述堆叠代理节点所代理的所有集群工作节点。实施本发明实施例可以确保系统具有极好的自组织性、扩展性，同时，实现业务的高可用性和业务均衡。

Description

分布式集群服务结构及节点协同方法和装置、终端及介质
本申请要求于2016年12月29日提交中国专利局，申请号为201611244228.0、发明名称为“一种分布式集群服务结构及其节点协同方法”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及服务器集群领域，尤其涉及一种分布式集群服务结构及节点协同方法和装置、终端及介质。
背景技术
随着各种现代系统中业务的复杂性规模、业务的并发规模、数据处理和存储规模地爆发式的增长，现在的计算机服务系统已经越来越依赖集群式的计算和存储。
目前，由于系统中的所有服务器不能在同一个地点的场所内容纳，则必需将同一个业务体系内的服务器分离部署到多个区域的场所中，在业务规模不断变化造成服务器的规模和分布不断变化的场景下，如果直接套用现存的常见的分布式系统的方案及结构(即中心化集群结构)，将会导致以下一些问题：1、中心控制下的集群计算需要大量的网络输入输出(Input Output，IO)的支持，在以数据为主要处理对象的系统中，IO瓶颈将是系统的处理能力扩展的一个制约；2、系统的频繁变化将造成业务的维护上的巨大困难；3、网络通信的不稳定性(如时延的抖动及丢包等)，对系统的稳定带来较大的影响；4、全网内的服务器之间业务相互关联，可扩展性差；5、系统之间的接口不统一，给各系统的设计带来较大的难度。
可见，如何提供一种新的集群服务结构来简化系统的设计/运维等多方面的问题是一个亟待解决的技术课题。
发明内容
本发明实施例提供了一种分布式集群服务结构及节点协同方法和装置、终端及介质，可以确保系统具有极好的自组织性、扩展性，同时，实现业务的高可用性和业务均衡。
本发明实施例第一方面公开了一种分布式集群服务结构，所述分布式集群服务结构为多层次分中心结构，所述分布式集群服务结构包括：集群工作节点和堆叠代理节点，所述堆叠代理节点作为所述集群工作节点的代理，其中：
所述集群工作节点，用于通过协议和流程确定所述集群工作节点所挂靠的堆叠代理节点；
所述堆叠代理节点，用于管理所述堆叠代理节点所代理的所有集群工作节点。
作为一种可选的实施方式，在本发明实施例第一方面中，所述堆叠代理节点包括根堆叠代理节点，所述根堆叠代理节点用于维护所述分布式集群服务结构。
作为一种可选的实施方式，在本发明实施例第一方面中，所述堆叠代理节点还用于进行所述根堆叠代理节点的选举过程、拓扑更新过程以及业务路由的维护过程。
作为一种可选的实施方式，在本发明实施例第一方面中，所述堆叠代理节点的接口是所述集群工作节点的接口的超集，所述堆叠代理节点的接口包括集群堆叠代理接口和工作节点接口代理接口。
作为一种可选的实施方式，在本发明实施例第一方面中，所述分布式集群服务结构为一个自组织且自愈的动态集群。
本发明实施例第二方面公开了一种基于分布式集群服务结构的节点协同方法，包括：
从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点；
通过所述根堆叠代理节点进行所述多个堆叠代理节点的拓扑形成过程；
通过所述根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
作为一种可选的实施方式，在本发明实施例第二方面中，所述从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点包括：
获取分布式集群服务结构包括的多个堆叠代理节点的服务器管理能力；
从所述多个堆叠代理节点中，将服务器管理能力最强的堆叠代理节点确定为根堆叠代理节点。
作为一种可选的实施方式，在本发明实施例第二方面中，所述方法还包括：
若服务器管理能力最强的堆叠代理节点存在多个，则从服务器管理能力最强的多个堆叠代理节点中，将身份标识号最小的堆叠代理节点确定为根堆叠代理节点。
作为一种可选的实施方式，在本发明实施例第二方面中，在所述多个堆叠代理节点的拓扑形成过程结束后，所述分布式集群服务结构中的每个所述堆叠代理节点均确定所述堆叠代理节点的父节点，所述堆叠代理节点的父节点为服务器管理能力以及存活时间TTL距离最优的堆叠代理节点。
作为一种可选的实施方式，在本发明实施例第二方面中，在所述业务路由形成过程结束后，所述分布式集群服务结构中的每个所述集群工作节点均确定所述集群工作节点的宿主代理节点，所述宿主代理节点为所述集群工作节点所挂靠的堆叠代理节点，并且所述宿主代理节点在服务于多个集群工作节点时剩余的服务管理能力最高。
本发明实施例第三方面公开了一种分布式集群服务结构的节点协同装置，所述节点协同装置包括：选举单元，用于从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点；第一执行单元，用于通过所述根堆叠代理节点进行所述多个堆叠代理节点的拓扑形成过程；第二执行单元，用于通过 所述根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
本发明实施例第四方面公开了一种终端，所述终端包括处理器，所述处理器用于执行存储器中存储的计算机程序时实现所述节点协同方法。
本发明实施例第五方面公开了一种计算机可读存储介质，其上存储有计算机指令，所述计算机指令被处理器执行时实现所述节点协同方法。
与现有技术相比，本发明实施例具备以下有益效果：
本发明实施例中，分布式集群服务结构为多层次分中心结构，可以满足在跨区域部署情况的业务的局部性、扩展性以及自组织性地需求。同时，通过在集群结构上加入堆叠代理节点，该堆叠代理节点作为集群工作节点的代理，并通过堆叠代理节点提供集群的自组织机制，实现集群的高扩展性和高可用性。此外，从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点，通过根堆叠代理节点进行多个堆叠代理节点的拓扑形成过程，通过根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程等的协同工作，实现了业务的高可用性和业务均衡。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例公开的一种分布式集群服务结构的结构示意图；
图2是本发明实施例公开的一种堆叠代理节点与集群工作节点的关系结构图；
图3是本发明实施例公开的一种基于分布式集群服务结构的节点协同方法的流程示意图；
图4是本发明实施例公开的一种节点协同装置的结构示意图。
图5是本发明实施例公开的一种终端的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选 地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
本发明实施例公开了一种分布式集群服务结构及节点协同方法和装置、终端及介质，可以确保系统具有极好的自组织性、扩展性，同时，实现业务的高可用性和业务均衡。以下分别进行详细说明。
请参阅图1，图1是本发明实施例公开的一种分布式集群服务结构的结构示意图。如图1所示，该分布式集群服务结构为多层次分中心结构，该分布式集群服务结包括集群工作节点和堆叠代理节点，比如：根堆叠代理节点，第0层中的集群工作节点0、堆叠代理节点0以及堆叠代理节点1，第1层中的堆叠代理节点2、集群工作节点2、集群工作节点3以及堆叠代理节点3，第2层中的集群工作节点(N-1)、集群工作节点1、集群工作节点N以及堆叠代理节点4。其中，堆叠代理节点作为集群工作节点的代理。
所述集群工作节点，用于通过协议和流程确定所述集群工作节点所挂靠的堆叠代理节点；
所述堆叠代理节点，用于管理所述堆叠代理节点所代理的所有集群工作节点。
举例来说，集群工作节点0所挂靠的堆叠代理节点为根堆叠代理节点，集群工作节点1所挂靠的堆叠代理节点为堆叠代理节点2，集群工作节点2所挂靠的堆叠代理节点为堆叠代理节点0，集群工作节点3所挂靠的堆叠代理节点为堆叠代理节点1，集群工作节点(N-1)所挂靠的堆叠代理节点为堆叠代理节点2，集群工作节点N所挂靠的堆叠代理节点为堆叠代理节点3。
其中，堆叠代理节点可以管理是其所代理的所有集群工作节点，此外，堆叠代理节点还可以管理堆叠子代理节点。比如：堆叠代理节点1可以管理集群工作节点3和堆叠代理节点3。
其中，系统可以通过根堆叠代理节点访问整个集群，同时，根堆叠代理节点用于维护分布式集群服务结构。
其中，该分布式集群服务结构为一个自组织且自愈的动态集群。
可选的，所述堆叠代理节点还用于进行所述根堆叠代理节点的选举过程、拓扑更新过程以及业务路由的维护过程。可以实现业务的高可用性和业务均衡。
可选的，所述堆叠代理节点的接口是所述集群工作节点的接口的超集，所 述堆叠代理节点的接口包括集群堆叠代理接口和工作节点接口代理接口。
请一并参见图2，图2是本发明实施例公开的一种堆叠代理节点与集群工作节点的关系结构图。如图2所示，堆叠代理节点的接口包括集群堆叠代理接口和工作节点接口代理接口，堆叠代理节点可以进行根堆叠代理节点的选举过程、拓扑更新过程以及业务路由的维护过程。集群工作节点的接口包括工作节点接口，可见，堆叠代理节点的接口是所述集群工作节点的接口的超集。
通过提供对集群工作节点的一个超级接口，可以实现系统的自组织性、可扩展性以及业务均衡性的特点。
本发明实施例中，分布式集群服务结构为多层次分中心结构，可以满足在跨区域部署情况的业务的局部性、扩展性以及自组织性地需求。同时，通过在集群结构上加入堆叠代理节点，该堆叠代理节点作为集群工作节点的代理，并通过堆叠代理节点提供集群的自组织机制，实现集群的高扩展性和高可用性。此外，从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点，通过根堆叠代理节点进行多个堆叠代理节点的拓扑形成过程，通过根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程等的协同工作，实现了业务的高可用性和业务均衡。
此外，该分布式集群服务结构具备堆叠性，可以利用数据处理及业务访问的局部性，消除不必要的网络/存储/控制等各方面不必要的IO及信息交互开销。各子系统之间设计统一的访问接口，可以支持不限层次的不同组件之间的混合堆叠，简化了系统的部署和维护。
请参见图3，图3是本发明实施例公开的一种基于分布式集群服务结构的节点协同方法的流程示意图。如图3所示，该方法可以包括以下步骤。
301、从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点。
本发明实施例中，在堆叠结构形成之前，所有堆叠代理节点是对等关系，这种对等关系体现在：在集群初始过程中，任意一个堆叠代理节点都有成为根堆叠代理节点的可能，在当前根堆叠代理节点异常退出服务时，在堆叠集群的更新过程中，剩余的正常的堆叠代理节点都有重新成为新的根堆叠代理节点的可能。
具体的，所述从分布式集群服务结构包括的多个堆叠代理节点中选举出根 堆叠代理节点包括：
获取分布式集群服务结构包括的多个堆叠代理节点的服务器管理能力；
从所述多个堆叠代理节点中，将服务器管理能力最强的堆叠代理节点确定为根堆叠代理节点。
在该实施例中，可以预先获取分布式集群服务结构包括的多个堆叠代理节点的服务器管理能力(即Capability)。在实际的系统中，CPU/GPU/内存/网络IO/存储IO/存储容量等等因素都表征了一个服务器节点的实际服务能力。如下公式表达节点的能力：
Capability(x)＝ΣSubCapality(n)×Weight(n)
其中，Capability(x)表示第x堆叠代理节点的服务器管理能力，SubCapality(n)为每一个子项的值，Weight(n)为每一个子项的权重(即重要程度)。
其中，假设系统中存在根堆叠代理节点，则根堆叠代理节点会按照第一周期向系统广播第一消息，同时，系统中的非根堆叠代理节点会按照第二周期向系统广播第二消息。其中，上述第一周期和第二周期可以为系统规定的任意时间，第一周期和第二周期可以相同也可以不同，第一消息可以为根节点通知消息(即RootNotify消息)，第二消息可以为能力通知消息(即CapabilityNotify消息)，第一消息和第二消息均携带有堆叠代理节点的身份标识号以及服务器管理能力值。
当一个新的堆叠代理节点的初始化完毕后，首先作N个第一周期的消息监听。在消息的监听过程中，该新的堆叠代理节点可能会收集到两种消息：第一消息和第二消息。如果在第一周期内，该新的堆叠代理节点收集到第一消息，表明系统内根堆叠代理节点的选举已经完成。此时这个新的堆叠代理节点选择暂时接受选举结果，并加入到此系统中，同时开始发送自己的第二消息。在该新的堆叠代理节点的服务器管理能力高于当前根堆叠代理节点的服务器管理能力的情况下，当前根堆叠代理节点会按照第三周期向系统广播重选举通知(即RerootRequest消息)，以询问系统内的其他堆叠代理节点是否需要重新选举根堆叠代理节点。该新的堆叠代理节点接收到重选举通知之后，确定该新的堆叠代理节点更适合成为新的根堆叠代理节点，向当前根堆叠代理节点响应重选举通知，当前根堆叠代理节点接收到针对重选举通知的响应消息后，停止发送第一消息，而开始发送第二消息，当系统中没有任何的堆叠代理节点发送第一消息 时，服务器管理能力最高的该新的堆叠代理节点停止发送第二消息，而开始发送第一消息。其中，上述第三周期可以为系统规定的任意时间。
作为一种可选的实施方式，所述方法还包括：
若服务器管理能力最强的堆叠代理节点存在多个，则从服务器管理能力最强的多个堆叠代理节点中，将身份标识号最小的堆叠代理节点确定为根堆叠代理节点。
其中，如果服务器管理能力最强的堆叠代理节点存在多个(即至少2个)，则身份标识号最小的堆叠代理节点将确定为根堆叠代理节点。
302、通过根堆叠代理节点进行多个堆叠代理节点的拓扑形成过程。
本发明实施例中，在根堆叠代理节点的选举结束后，需要启动堆叠代理节点的拓扑形成过程。根堆叠代理节点向系统广播拓扑建立信息(即TopologyBuild消息)，其他的堆叠代理节点向系统广播拓扑等待信息(即TopologyWait消息)。其他的堆叠代理节点接收到拓扑建立信息之后，如果该堆叠代理节点的拓扑处于正在建立状态(即Building状态)，则缓存拓扑建立信息，在超过预设的时间后，将服务管理能力及存活时间(Time to Live，TTL)距离最优的节点确定作为父节点，并进入到拓扑完成状态，停止发送拓扑等待信息。如果该堆叠代理节点的拓扑处于完成状态(即Builed状态)，则可以向除了根堆叠代理节点以及父节点之外的节点发送拓扑建立信息。
在所述多个堆叠代理节点的拓扑形成过程结束后，所述分布式集群服务结构中的每个所述堆叠代理节点均确定所述堆叠代理节点的父节点，所述堆叠代理节点的父节点为服务器管理能力以及存活时间TTL距离最优的堆叠代理节点。
303、通过根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
本发明实施例中，根堆叠代理节点向系统广播路由通知信息(即RouteNotice消息)，以获取所有集群工作节点的信息。根堆叠代理节点向任一个集群工作节点发送路由令牌(即RouteToken令牌)，集群工作节点接收到该RouteToken令牌之后，从自己收集到的代理信息中，分别计算该堆叠代理节点的剩余服务器管理能力值，计算公式如下：
Capabilityleft＝Capability-ΣUsedCapability(n)
其中，该公式的含义为该堆叠代理节点本身具有的服务器管理能力值减去当其服务于某个集群工作节点后，其所需要的服务器管理能力值，即剩余的Capability的值。计算最高的为该集群工作节点的宿主代理节点。
其中，在所述业务路由形成过程结束后，所述分布式集群服务结构中的每个所述集群工作节点均确定所述集群工作节点的宿主代理节点，所述宿主代理节点为所述集群工作节点所挂靠的堆叠代理节点，并且所述宿主代理节点在服务于多个集群工作节点时剩余的服务管理能力最高。
本发明实施例中，分布式集群服务结构为多层次分中心结构，可以满足在跨区域部署情况的业务的局部性、扩展性以及自组织性地需求。同时，通过在集群结构上加入堆叠代理节点，该堆叠代理节点作为集群工作节点的代理，并通过堆叠代理节点提供集群的自组织机制，实现集群的高扩展性和高可用性。此外，从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点，通过根堆叠代理节点进行多个堆叠代理节点的拓扑形成过程，通过根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程等的协同工作，实现了业务的高可用性和业务均衡。
此外，该分布式集群服务结构具备堆叠性，可以利用数据处理及业务访问的局部性，消除不必要的网络/存储/控制等各方面不必要的IO及信息交互开销。各子系统之间设计统一的访问接口，可以支持不限层次的不同组件之间的混合堆叠，简化了系统的部署和维护。
请参见图4，图4是本发明实施例公开的一种节点协同装置的结构示意图。其中，该节点协同装置可以用于执行图2所描述的基于分布式集群服务结构的节点协同方法，具体请参照图2中的相关描述，在此不再赘述。如图4所示，该节点协同装置包括：
选举单元401，用于从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点；
第一执行单元402，用于通过所述根堆叠代理节点进行所述多个堆叠代理节点的拓扑形成过程；
第二执行单元403，用于通过所述根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
可选的，所述选举单元从分布式集群服务结构包括的多个堆叠代理节点中 选举出根堆叠代理节点的方式具体为：
获取分布式集群服务结构包括的多个堆叠代理节点的服务器管理能力；
从所述多个堆叠代理节点中，将服务器管理能力最强的堆叠代理节点确定为根堆叠代理节点。
可选的，所述选举单元从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点的方式具体为：
若服务器管理能力最强的堆叠代理节点存在多个，则从服务器管理能力最强的多个堆叠代理节点中，将身份标识号最小的堆叠代理节点确定为根堆叠代理节点。
其中，在所述多个堆叠代理节点的拓扑形成过程结束后，所述分布式集群服务结构中的每个所述堆叠代理节点均确定所述堆叠代理节点的父节点，所述堆叠代理节点的父节点为服务器管理能力以及存活时间TTL距离最优的堆叠代理节点。
其中，在所述业务路由形成过程结束后，所述分布式集群服务结构中的每个所述集群工作节点均确定所述集群工作节点的宿主代理节点，所述宿主代理节点为所述集群工作节点所挂靠的堆叠代理节点，并且所述宿主代理节点在服务于多个集群工作节点时剩余的服务管理能力最高。
本发明实施例中，分布式集群服务结构为多层次分中心结构，可以满足在跨区域部署情况的业务的局部性、扩展性以及自组织性地需求。同时，通过在集群结构上加入堆叠代理节点，该堆叠代理节点作为集群工作节点的代理，并通过堆叠代理节点提供集群的自组织机制，实现集群的高扩展性和高可用性。此外，从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点，通过根堆叠代理节点进行多个堆叠代理节点的拓扑形成过程，通过根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程等的协同工作，实现了业务的高可用性和业务均衡。
此外，该分布式集群服务结构具备堆叠性，可以利用数据处理及业务访问的局部性，消除不必要的网络/存储/控制等各方面不必要的IO及信息交互开销。各子系统之间设计统一的访问接口，可以支持不限层次的不同组件之间的混合堆叠，简化了系统的部署和维护。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述 为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，因为依据本发明，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本发明所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置，可通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性或其它的形式。
作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
请参见图5，集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储器中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储器中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本发明各个实施例方法的全部或部分步骤。而前述的存储器包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，RandomAccess Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。图5中的终端还包括处理器，所述处理器用于执行存储器中存储的计算 机程序，以实现所述节点协同方法。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序可以存储于一计算机可读存储器中，存储器可以包括：闪存盘、只读存储器(英文：Read-Only Memory，简称：ROM)、随机存取器(英文：Random Access Memory，简称：RAM)、磁盘或光盘等。
以上对本发明实施例进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上上述，本说明书内容不应理解为对本发明的限制。

Claims (13)

  1. 一种分布式集群服务结构，其特征在于，所述分布式集群服务结构为多层次分中心结构，所述分布式集群服务结构包括：集群工作节点和堆叠代理节点，所述堆叠代理节点作为所述集群工作节点的代理，其中：
    所述集群工作节点，用于通过协议和流程确定所述集群工作节点所挂靠的堆叠代理节点；
    所述堆叠代理节点，用于管理所述堆叠代理节点所代理的所有集群工作节点。
  2. 根据权利要求1所述的分布式集群服务结构，其特征在于，所述堆叠代理节点包括根堆叠代理节点，所述根堆叠代理节点用于维护所述分布式集群服务结构。
  3. 根据权利要求2所述的分布式集群服务结构，其特征在于，所述堆叠代理节点还用于进行所述根堆叠代理节点的选举过程、拓扑更新过程以及业务路由的维护过程。
  4. 根据权利要求1-3任一项所述的分布式集群服务结构，其特征在于，所述堆叠代理节点的接口是所述集群工作节点的接口的超集，所述堆叠代理节点的接口包括集群堆叠代理接口和工作节点接口代理接口。
  5. 根据权利要求1-3任一项所述的分布式集群服务结构，其特征在于，所述分布式集群服务结构为一个自组织且自愈的动态集群。
  6. 一种基于分布式集群服务结构的节点协同方法，其特征在于，包括：
    从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点；
    通过所述根堆叠代理节点进行所述多个堆叠代理节点的拓扑形成过程；
    通过所述根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
  7. 根据权利要求6所述的方法，其特征在于，所述从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点包括：
    获取分布式集群服务结构包括的多个堆叠代理节点的服务器管理能力；
    从所述多个堆叠代理节点中，将服务器管理能力最强的堆叠代理节点确定为根堆叠代理节点。
  8. 根据权利要求7所述的方法，其特征在于，所述方法还包括：
    若服务器管理能力最强的堆叠代理节点存在多个，则从服务器管理能力最强的多个堆叠代理节点中，将身份标识号最小的堆叠代理节点确定为根堆叠代理节点。
  9. 根据权利要求6-8任一项所述的方法，其特征在于，在所述多个堆叠代理节点的拓扑形成过程结束后，所述分布式集群服务结构中的每个所述堆叠代理节点均确定所述堆叠代理节点的父节点，所述堆叠代理节点的父节点为服务器管理能力以及存活时间TTL距离最优的堆叠代理节点。
  10. 根据权利要求6-8任一项所述的方法，其特征在于，在所述业务路由形成过程结束后，所述分布式集群服务结构中的每个所述集群工作节点均确定所 述集群工作节点的宿主代理节点，所述宿主代理节点为所述集群工作节点所挂靠的堆叠代理节点，并且所述宿主代理节点在服务于多个集群工作节点时剩余的服务管理能力最高。
  11. 一种基于分布式集群服务结构的节点协同装置，其特征在于，包括：
    选举单元，用于从分布式集群服务结构包括的多个堆叠代理节点中选举出根堆叠代理节点；
    第一执行单元，用于通过所述根堆叠代理节点进行所述多个堆叠代理节点的拓扑形成过程；
    第二执行单元，用于通过所述根堆叠代理节点协调进行从集群工作节点到堆叠代理节点的分配的业务路由形成过程。
  12. 一种终端，其特征在于，所述终端包括处理器，所述处理器用于执行存储器中存储的计算机程序时实现如权利要求6至10中任意一项所述节点协同方法。
  13. 一种计算机可读存储介质，其上存储有计算机指令，其特征在于：所述计算机指令被处理器执行时实现如权利要求6至10中任意一项所述节点协同方法。
PCT/CN2017/114855 2016-12-29 2017-12-06 分布式集群服务结构及节点协同方法和装置、终端及介质 WO2018121201A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611244228.0 2016-12-29
CN201611244228.0A CN106713468B (zh) 2016-12-29 2016-12-29 一种分布式集群服务系统及其节点协同方法

Publications (1)

Publication Number Publication Date
WO2018121201A1 true WO2018121201A1 (zh) 2018-07-05

Family

ID=58903881

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/114855 WO2018121201A1 (zh) 2016-12-29 2017-12-06 分布式集群服务结构及节点协同方法和装置、终端及介质

Country Status (2)

Country Link
CN (1) CN106713468B (zh)
WO (1) WO2018121201A1 (zh)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106713468B (zh) * 2016-12-29 2018-11-20 深圳云天励飞技术有限公司 一种分布式集群服务系统及其节点协同方法
CN110033078B (zh) * 2018-01-12 2024-01-12 华为技术有限公司 一种基于树状拓扑的计算系统及方法
CN108616566B (zh) * 2018-03-14 2021-02-23 华为技术有限公司 raft分布式系统选主方法、相关设备及系统
CN109714312B (zh) * 2018-11-19 2020-04-24 中国科学院信息工程研究所 一种基于外部威胁的采集策略生成方法及系统
CN111404647B (zh) * 2019-01-02 2023-11-28 中兴通讯股份有限公司 一种节点协同关系的控制方法及相关设备
CN110351375B (zh) * 2019-07-16 2022-08-12 深圳市网心科技有限公司 一种数据处理方法、装置及计算机装置、可读存储介质
CN110855505B (zh) * 2019-11-26 2022-03-04 上海英方软件股份有限公司 一种高可用规则堆叠实现系统及方法
CN113568303B (zh) * 2021-09-26 2021-12-14 成都数默科技有限公司 一种基于pid控制算法的网络流量抓包限流丢包方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101478435A (zh) * 2009-01-20 2009-07-08 杭州华三通信技术有限公司 一种堆叠系统的拓扑收集方法和双控制板设备
CN101594292A (zh) * 2008-05-30 2009-12-02 中兴通讯股份有限公司 内容发布方法、服务重定向方法及系统、节点设备
CN101841735A (zh) * 2009-03-17 2010-09-22 华为技术有限公司 框式交换机及堆叠系统以及堆叠后故障处理方法
CN102325196A (zh) * 2011-10-27 2012-01-18 上海文广互动电视有限公司 分布式集群存储系统
CN104518995A (zh) * 2013-09-26 2015-04-15 中国电信股份有限公司 基于分布式架构的交换机虚拟化系统
WO2016100240A1 (en) * 2014-12-14 2016-06-23 Zte Corporation Techniques for operating a distributed communication network
CN106713468A (zh) * 2016-12-29 2017-05-24 深圳云天励飞技术有限公司 一种分布式集群服务结构及其节点协同方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101594292A (zh) * 2008-05-30 2009-12-02 中兴通讯股份有限公司 内容发布方法、服务重定向方法及系统、节点设备
CN101478435A (zh) * 2009-01-20 2009-07-08 杭州华三通信技术有限公司 一种堆叠系统的拓扑收集方法和双控制板设备
CN101841735A (zh) * 2009-03-17 2010-09-22 华为技术有限公司 框式交换机及堆叠系统以及堆叠后故障处理方法
CN102325196A (zh) * 2011-10-27 2012-01-18 上海文广互动电视有限公司 分布式集群存储系统
CN104518995A (zh) * 2013-09-26 2015-04-15 中国电信股份有限公司 基于分布式架构的交换机虚拟化系统
WO2016100240A1 (en) * 2014-12-14 2016-06-23 Zte Corporation Techniques for operating a distributed communication network
CN106713468A (zh) * 2016-12-29 2017-05-24 深圳云天励飞技术有限公司 一种分布式集群服务结构及其节点协同方法

Also Published As

Publication number Publication date
CN106713468B (zh) 2018-11-20
CN106713468A (zh) 2017-05-24

Similar Documents

Publication Publication Date Title
WO2018121201A1 (zh) 分布式集群服务结构及节点协同方法和装置、终端及介质
US9021065B2 (en) Automated topology formation in dynamic distributed environments
CN101471885B (zh) 具有状态同步的集群的虚拟组播路由
CN102447624B (zh) 在服务器集群上实现负载均衡的方法、节点服务器及集群
CN109548082A (zh) 业务重定向方法及装置
EP3063903A1 (en) Method and system for load balancing at a data network
JP6551547B2 (ja) マルチグループｐ２ｐネットワークにおけるトラフィック認識型グループ再構成
US11140091B2 (en) Openflow protocol-based resource control method and system, and apparatus
US20130201875A1 (en) Distributed fabric management protocol
EP3132567B1 (en) Event processing in a network management system
WO2021120633A1 (zh) 一种负载均衡方法及相关设备
WO2012079312A1 (zh) 一种告警数据的同步方法及通讯网络管理系统
CN108282526B (zh) 双集群间服务器动态分配方法及系统
CN111858033A (zh) 基于集群和多进程的负载均衡方法
CN103401951A (zh) 基于对等架构的弹性云分发方法
WO2023124743A1 (zh) 区块同步
US11329880B2 (en) Automatic route reflector configuration system
WO2018036521A1 (zh) 资源调整方法、装置及系统
US11539584B2 (en) Automatic and dynamic adaptation of grouping in a data processing system
CN115514651B (zh) 基于软件定义层叠网的云边数据传输路径规划方法及系统
US11422866B2 (en) Computer network of computing resource infrastructures and method for allocating said resources to client applications
US10397099B2 (en) Spanning tree protocol enabled N-node link aggregation system
CN102647424A (zh) 一种数据传输方法及其装置
CN113055195A (zh) 基于sdon的多域控制器集群和sdon系统
JP2020061135A (ja) 企業ネットワーク内の動的コンテンツ配信

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17885643

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 24/09/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 17885643

Country of ref document: EP

Kind code of ref document: A1