WO2025123783A1 - 云盘管控方法、系统、电子设备及存储介质 - Google Patents

云盘管控方法、系统、电子设备及存储介质 Download PDF

Info

Publication number
WO2025123783A1
WO2025123783A1 PCT/CN2024/114916 CN2024114916W WO2025123783A1 WO 2025123783 A1 WO2025123783 A1 WO 2025123783A1 CN 2024114916 W CN2024114916 W CN 2024114916W WO 2025123783 A1 WO2025123783 A1 WO 2025123783A1
Authority
WO
WIPO (PCT)
Prior art keywords
cloud disk
target
cluster
storage
management
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/CN2024/114916
Other languages
English (en)
French (fr)
Inventor
张烁
吴东
佟光勋
张小路
吴均平
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.)
Alibaba Cloud Computing Ltd
Original Assignee
Alibaba Cloud Computing 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 Alibaba Cloud Computing Ltd filed Critical Alibaba Cloud Computing Ltd
Publication of WO2025123783A1 publication Critical patent/WO2025123783A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

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
    • 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]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/10—File systems; File servers
    • G06F16/18—File system types
    • G06F16/182—Distributed file systems
    • G06F16/1824—Distributed file systems implemented using Network-attached Storage [NAS] architecture
    • G06F16/183—Provision of network file services by network file servers, e.g. by using NFS, CIFS

Definitions

  • the present disclosure relates to the field of distributed storage technology, and in particular to a cloud disk management method, system, electronic device and storage medium.
  • Block storage system is a low-latency, persistent and highly reliable block-level random distributed storage system.
  • data is divided into blocks, and each block is operated as an independent hard disk.
  • This method provides high performance and flexibility, and is particularly suitable for processing large amounts of data, such as databases and large-scale virtual machine environments.
  • cloud disk provides users with block storage services at the data block level. It virtualizes the storage space of physical hard disks and provides it to users through the network. Users can use cloud disks like local hard disks to read and write data. Therefore, in the block storage system, it is very important to achieve efficient management and control of cloud disks and reduce the fault domain of block storage systems.
  • the embodiments of the present disclosure provide a cloud disk management method, system, electronic device and storage medium to alleviate or solve one or more technical problems existing in the prior art.
  • an embodiment of the present disclosure provides a cloud disk management and control method, comprising: obtaining cloud disk identification information of a target cloud disk; the cloud disk identification information is used to identify the target cloud disk and a target storage management and control node and a target cluster corresponding to the target cloud disk; the target cluster comprises a plurality of storage management and control nodes, the plurality of storage management and control nodes including the target storage management and control node; the target storage management and control node is used to create the target cloud disk on at least one storage service node of the target cluster and generate the cloud disk identification information; and the target cluster, the target storage management and control node and the target cloud disk are determined according to the cloud disk identification information.
  • an embodiment of the present disclosure provides a cloud disk management and control method, comprising: in response to a cloud disk creation request, creating a target cloud disk on at least one storage service node of a target cluster; the target cluster comprises multiple storage management and control nodes, the multiple storage management and control nodes include the target storage management and control node; generating cloud disk identification information of the target cloud disk; the cloud disk identification information is used to identify the target cloud disk, the target storage management and control node and the target cluster.
  • an embodiment of the present disclosure provides a cloud disk management and control system, including: a cluster management and control node, used to execute the method provided by the embodiment of the first aspect of the present disclosure; at least one cluster, the cluster including multiple storage management and control nodes and multiple storage service nodes, the storage management and control node is used to execute the method provided by the embodiment of the second aspect of the present disclosure.
  • an embodiment of the present disclosure provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor implements the method provided in any embodiment of the present disclosure when executing the computer program.
  • an embodiment of the present disclosure provides a computer-readable storage medium, in which a computer program is stored.
  • the computer program is executed by a processor, the method provided by any embodiment of the present disclosure is implemented.
  • an embodiment of the present disclosure provides a computer program product, including a computer program, which implements the method provided by any embodiment of the present disclosure when executed by a processor.
  • the storage management and control nodes in a storage cluster are horizontally expanded, that is, a storage cluster includes multiple storage management and control nodes, and each storage management and control node is responsible for the management and control of several groups of cloud disks (for example, a collection of multiple cloud disks).
  • each storage management and control node works at the same time, avoiding the waste of resources caused by a one-master and multiple-standby architecture, improving resource utilization, and supporting large-scale clusters; on the other hand, when a storage management and control node fails, only a part of the cloud disks in the storage cluster are affected, so the explosion radius is much smaller than the explosion radius of the centralized BM architecture.
  • an effective cluster resource scheduling solution can be provided so that cloud disk scheduling is not affected by architecture updates.
  • FIG1 shows an architecture diagram of a cloud disk management and control system provided in the related art
  • FIG2 shows an architecture diagram of a cloud disk management and control system provided by an embodiment of the present disclosure
  • FIG3 shows an application example diagram of cloud disk scheduling provided by an embodiment of the present disclosure
  • FIG4 shows another application example diagram of cloud disk scheduling provided by an embodiment of the present disclosure
  • FIG5 is a flow chart of a cloud disk management and control method provided by an embodiment of the present disclosure.
  • FIG6 shows a flow chart of a cloud disk management and control method provided by an embodiment of the present disclosure
  • FIG. 7 shows a block diagram of an electronic device provided by an embodiment of the present disclosure.
  • Block Storage System A low-latency, persistent, and highly reliable block-level random storage system that divides data into blocks, each of which operates as an independent hard drive. This means that users can freely create their own file systems on these blocks, thereby better controlling the data environment, which is particularly suitable for processing large amounts of data, such as databases and large-scale virtual machine environments.
  • Cloud Disk Based on the distributed storage architecture, it provides block storage services at the data block level.
  • the data of the cloud disk is persistently stored in the distributed storage system on the backend, and the downtime of any node on the computing end will not affect the cloud disk.
  • Explosion radius represents the size of the fault domain, that is, the impact area when a node fails. The smaller the explosion radius, the smaller the fault domain; the larger the explosion radius, the larger the fault domain.
  • Internet Data Center is a facility for storing, managing and distributing large amounts of data. It provides a range of services including server hosting, network connection, data storage and backup, and security management.
  • Availability Zone represents a physical location that has its own power supply, network connection, and security facilities. There may be one or more availability zones in the same geographic region. Each availability zone is independent of each other and isolated from other availability zones to prevent other availability zones from being affected when a problem occurs in one availability zone. For example, if an availability zone has problems due to power failure, network outage, or natural disaster, other availability zones can continue to operate normally. In a cloud computing environment, an availability zone contains one or more Internet Data Centers (IDCs).
  • IDCs Internet Data Centers
  • Virtual block device A virtualized storage device that simulates the functions of a physical hard drive, allowing a virtual machine to access storage space just like a physical hard drive, thereby reading or writing data in a cloud disk.
  • Garbage Collection An automatic memory management technology.
  • the garbage collector automatically reclaims the memory space occupied by these objects for reuse.
  • the GC task refers to the regular cleaning of useless data blocks or objects to free up storage space and improve resource utilization.
  • the architecture of this centralized management method is relatively simple.
  • the client obtains the location of the segment through the BM, and then can read and write data on the corresponding BS.
  • the centralized BM is responsible for the management, scheduling, and operation and maintenance of the entire storage cluster. It can work more efficiently when there are fewer cloud disks in a single cluster, a limited number of segments, and a small cluster size.
  • a single cluster is expected to support more segments and cloud disks in the future. If it is to support 10 million segments in the future, a single BM will need to provide 360G memory. Obviously, the existing servers cannot support this requirement.
  • FIG2 shows an architecture diagram of a cloud disk management system 200 according to an implementation of the embodiment of the present disclosure.
  • the cloud disk management system 200 includes a cluster management node 201 and one or more clusters 202 .
  • the cluster management node 201 is a service node that manages each cluster 202.
  • the cluster management node 201 may be a resource management node RM.
  • the cluster management node 201 that manages each storage cluster decides one of the clusters 202 as a target cluster.
  • Cluster 202 is a storage cluster.
  • the storage management and control node can be a block storage backend control node (BM), as shown in FIG2 , and the multiple storage management and control nodes are, for example, BM1, BM2, BM3, and BM4.
  • the storage service node can be a block storage backend service node (BS), as shown in FIG2 , and the multiple storage service nodes are, for example, BS1, BS2, BS3, BS4, BS5, BS6, BS7, and BS8.
  • BS block storage backend service node
  • the embodiments of the present disclosure do not specifically limit the number of clusters 202, and the number of storage management and control nodes and storage service nodes deployed in each cluster.
  • Each storage management and control node is used to manage multiple cloud disk sets (partitions).
  • BM1 can manage cloud disk set 1 and cloud disk set 8 (for cloud disk set list 1)
  • BM2 can manage cloud disk set 2 and cloud disk set 3 (for cloud disk set list 2)
  • BM3 can manage cloud disk set 4 and cloud disk set 6 (for cloud disk set list 3)
  • BM4 can manage cloud disk set 5 and cloud disk set 7 (for cloud disk set list 4).
  • multiple storage management and control nodes have a one-to-one mapping relationship with multiple cloud disk set lists (partition lists), that is, cloud disk sets and storage management and control nodes have a mapping relationship.
  • a cloud disk set is a logical concept, which is a collection of a batch of cloud disks.
  • a cloud disk can only belong to one cloud disk set. After the cloud disk set is created, it needs to be loaded by the corresponding storage management node before it can provide corresponding services.
  • cloud disk sets partitions
  • the storage control node controls its corresponding cloud disk collection, including but not limited to: creating cloud disks and generating corresponding cloud disk identification information (Volume ID) for the created cloud disks; mounting cloud disks (opening disks), deleting cloud disks, and scheduling cloud disks based on the cloud disk identification information.
  • creating cloud disks and generating corresponding cloud disk identification information Volume ID
  • mounting cloud disks opening disks
  • deleting cloud disks and scheduling cloud disks based on the cloud disk identification information.
  • the cloud disk identification information includes a cluster ID, a cloud disk set ID, and a cloud disk ID (the location of the cloud disk on the cloud disk set).
  • the location of the cloud disk on the cloud disk set can be represented by a cloud disk number or a cloud disk random number.
  • each cloud disk in the cloud disk set has a unique cloud disk number or a unique cloud disk random number.
  • the cloud disk identification information may also include a reserved field (or preset field), which may identify a reserved function or preset function or other information of the cloud disk.
  • CM central control node
  • its deployment may also include a central control node (Central Master, CM), which is responsible for bypass control work from the global perspective of the cluster 202, including: managing the online and offline of BM in the cluster 202; managing the online and offline of BS in the cluster 202; allocating and creating cloud disk collections, creating and configuring corresponding resources for them, etc.
  • Central Master CM
  • the cloud disk management and control system 200 may also include a client 203, which may be deployed in a computing cluster, and obtain cloud disk identification information by interacting with the cluster management and control node 201, and then access the corresponding cloud disk based on the cloud disk identification information, such as mounting a cloud disk.
  • the client 203 may interact with the storage management and control node through an application program interface (Application Programming Interface, API), obtain information about the storage service node where the Segment is located, and then read and write data on the storage service node.
  • API Application Programming Interface
  • the following describes an application example of a dynamic disk creation process of a cloud disk management and control system 200 based on an embodiment of the present disclosure in conjunction with FIG. 2.
  • the disk creation process mainly includes:
  • Cluster management node 201 selects one of the multiple storage management nodes in cluster 201 as the target storage
  • the selection strategy for the control node e.g. BM1
  • Cluster management node 201 sends a cloud disk creation request to BM1;
  • BM1 may be configured with a proxy or management component (proxy), and the proxy of BM1 selects one of multiple cloud disk sets as a target cloud disk set (for example, cloud disk set 8);
  • proxy proxy or management component
  • BM1 completes the creation of the target cloud disk, generates the cloud disk identification information (Volume ID) of the target cloud disk, and returns the disk creation success information and the cloud disk identification information of the target cloud disk to the cluster management and control node 201 through the original path.
  • the following describes an application example of a dynamic disk opening process of a cloud disk management and control system 200 based on an embodiment of the present disclosure in conjunction with FIG. 2.
  • the disk opening process mainly includes:
  • the client 203 queries the cluster management and control node 201 for information about the target cluster, cloud disk identification information of the target cloud disk, and information about the target storage management and control node through the device identification information (DEVID) of the virtual block device;
  • DEVID device identification information
  • the cluster management and control node 201 parses the corresponding cloud disk identification information (Volume ID) based on the device identification information provided by the client 203, and obtains the cluster identification (Cluster ID), cloud disk collection identification (Partition ID) and cloud disk identification (the location of the cloud disk on the cloud disk collection) by parsing the Volume ID. Then, based on the mapping relationship between the cloud disk collection and the storage management and control node, the storage management and control node (for example, BM1) corresponding to the cloud disk collection identification (Partition ID) is determined as the target storage management and control node;
  • BM1 cloud disk collection identification
  • the cluster management node 201 returns the cluster identifier (e.g., cluster name) of the target cluster, the Internet Protocol (IP) address of BM1, and the Volume ID to the client 203;
  • the cluster identifier e.g., cluster name
  • IP Internet Protocol
  • Client 203 sends a cloud disk mount request for the target cloud disk to BM1 based on BM1's IP address;
  • BM1 responds to the cloud disk mount request and mounts the target cloud disk.
  • the cloud disk scheduling process mainly includes:
  • Each BM in the cluster obtains the traffic information of each BS in the cluster.
  • the acquisition method can be that each BS in the cluster sends a heartbeat to the central control node (CM) to report its own traffic information; the acquisition method can also be that each BM in the cluster sends a slow heartbeat to each BS in the cluster, so that each BM can obtain the traffic information of each BS.
  • the slow heartbeat is sent once every 10 seconds, for example;
  • Each BM performs cloud disk scheduling in turn according to a preset order.
  • BM1, BM2, BM3 and BM4 perform cloud disk scheduling in turn, including traffic balancing scheduling (dynamic or static), so as to schedule cloud disk segments of the cloud disk to multiple storage service nodes.
  • CM is used to control each BM to perform cloud disk scheduling in turn according to a preset order and perform certain interference from the global perspective of the cluster according to the global coordination mechanism. If CM fails, each BM will continue to perform cloud disk scheduling in turn.
  • the cloud disk scheduling process mainly includes:
  • Each BS in the cluster is divided into multiple groups, forming storage service node groups.
  • Each BM corresponds to each storage service node group one by one.
  • the BM has jurisdiction over the corresponding storage service node group, including initiating downtime operations on the BS of the storage service node group.
  • BS1 and BS2 form group 1
  • BM1 corresponds to group 1 and has jurisdiction over BS1 and BS2
  • BS3 and BS4 form Group 2
  • BM2 corresponds to Group 2 and has jurisdiction over BS3 and BS4
  • BS5 and BS6 form Group 3
  • BM3 corresponds to Group 3 and has jurisdiction over BS5 and BS6
  • BS7 and BS8 form Group 4
  • BM4 corresponds to Group 4 and has jurisdiction over BS7 and BS8;
  • the BM sends a fast heartbeat to the corresponding BS, such as once every 200 milliseconds, to obtain the operating status information of the corresponding BS for fault diagnosis (downtime diagnosis) of the BS;
  • the BM determines that there is a faulty BS in the group, that is, there is a downtime BS, it will perform a downtime operation on the BS. For example, if BM1 determines that BS2 in group 1 is downtime, BM1 will perform a downtime operation on the BS to implement downtime scheduling (also called fault scheduling or abnormal scheduling).
  • downtime scheduling also called fault scheduling or abnormal scheduling
  • each BM reports a heartbeat to the RM and reports its own operating status information, that is, it is determined that the BM is not faulty; the cluster granularity information setting is reported by a BM in the cluster (for example, BM1 by default).
  • the central management and control node reports a heartbeat to each BM, obtains the operating status information of each BM, such as whether it is faulty, and obtains the load status of each cloud disk set on the BM.
  • the cloud disk management and control system of the embodiment of the present disclosure provides a cloud disk management and control solution based on distributed storage management and control nodes, which horizontally expands the storage management and control nodes in a storage cluster, that is, a storage cluster includes multiple storage management and control nodes, and each storage management and control node is responsible for the management and control of several cloud disk sets.
  • each storage management and control node works at the same time, avoiding the waste of resources caused by the one-master-multiple-standby architecture and improving resource utilization; on the other hand, when a storage management and control node fails, other storage management and control nodes take over the cloud disk set it is responsible for.
  • the explosion radius is much smaller than the explosion radius of the centralized BM architecture; on the other hand, it can support large-scale clusters.
  • the solution of the embodiment of the present disclosure can support larger clusters of 200 or even 1,000 units; on the other hand, the embodiment of the present disclosure also provides an effective cluster resource scheduling solution for the cloud disk management and control architecture of the distributed storage management and control nodes, so that the cloud disk scheduling is not affected by the architecture update.
  • the above application scenarios provided in the embodiments of the present disclosure are for ease of understanding, and the embodiments of the present disclosure do not specifically limit the application scenarios of the technical solution.
  • the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present disclosure are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to select or edit authorization or rejection.
  • FIG5 shows a flow chart of a cloud disk management method according to an embodiment of the present disclosure.
  • the cloud disk management method can be applied to a cluster management node.
  • the cloud disk management method includes:
  • Step S501 Obtain the cloud disk identification information of the target cloud disk; the cloud disk identification information is used to identify the target cloud disk and the target storage management and control node and target cluster corresponding to the target cloud disk; the target cluster includes multiple storage management and control nodes, multiple storage management nodes, and multiple storage management nodes.
  • the control nodes include a target storage control node; the target storage control node is used to create a target cloud disk on at least one storage service node of the target cluster and generate cloud disk identification information.
  • each storage management and control node is used to manage multiple cloud disk sets (partitions), and the cloud disk identification information includes cluster identification, partition identification and cloud disk identification; among them, the cluster identification of the target cloud disk is used to determine the target cluster; the partition identification of the target cloud disk is used to identify the target partition to which the target cloud disk belongs, so as to determine the storage management and control node corresponding to the target partition as the target storage management and control node based on the mapping relationship between the partition and the storage management and control node; the cloud disk identification of the target cloud disk is used to determine the position of the target cloud disk on the target partition.
  • the storage management node in a storage cluster is horizontally expanded, that is, a storage cluster includes multiple storage management nodes, and each storage management node is responsible for the management of several groups of cloud disks (for example, a collection of multiple cloud disks).
  • each storage management node works at the same time, avoiding the waste of resources caused by the one-master-multiple-standby architecture, improving resource utilization, and supporting large-scale clusters; on the other hand, in the event of a failure of a storage management node, only a part of the cloud disks in the storage cluster are affected, so the explosion radius is much smaller than the explosion radius of the centralized BM (one-master-multiple-standby) architecture.
  • multiple storage management and control nodes take turns to execute cloud disk scheduling in a preset order.
  • the method executed by the target storage management and control node in the embodiment of the present disclosure may also include: obtaining traffic information of each storage service node in the target cluster; in response to it being the turn of the target storage management and control node to execute cloud disk scheduling, scheduling at least one cloud disk corresponding to the target storage management and control node based on the traffic information of each storage service node.
  • each storage control node in the target cluster obtains the traffic information of each storage service node in the target cluster.
  • the acquisition method may be that each storage service node in the target cluster sends a heartbeat to the central control node (CM) to report its own traffic information; the acquisition method may also be that each storage control node in the target cluster sends a slow heartbeat to each storage service node in the target cluster, so that the target storage control node can obtain the traffic information of each storage service node in the target cluster.
  • each storage control node performs cloud disk scheduling in turn according to a preset order. When it is the turn of the target storage control node to perform cloud disk scheduling, the target storage control node can perform cloud disk scheduling on each part it ion under its control, including dynamic traffic balancing scheduling or static traffic balancing scheduling.
  • each storage service node of the target cluster is divided into multiple storage service node groups, and multiple storage management and control nodes correspond one-to-one to multiple storage service node groups.
  • the method performed by the target storage management and control node in the disclosed embodiment may also include: based on the operation status information of each storage service node in the storage service node group corresponding to the target storage management and control node, judging whether there is a faulty storage service node in the storage service node group corresponding to the target storage management and control node; in response to the existence of a faulty storage service node, performing a downtime operation on the faulty storage service node.
  • the method executed by the target storage control node in the embodiment of the present disclosure may further include: in response to a failed storage control node in the target cluster, controlling each cloud disk corresponding to the failed storage control node. That is, when a storage control node fails, other storage control nodes take over the cloud disks for which they are responsible.
  • an effective cluster resource scheduling solution can be provided for the cloud disk management and control architecture of the distributed storage management and control node, so that the cloud disk scheduling is not affected by the architecture update.
  • the embodiment of the present disclosure provides a cloud disk management and control device, which is applied to a cluster management and control node and may include: a cloud disk identification information acquisition module, used to obtain cloud disk identification information of a target cloud disk; the cloud disk identification information is used to identify the target cloud disk and the target storage management and control node and target cluster corresponding to the target cloud disk; the target cluster includes multiple storage management and control nodes, and the multiple storage management and control nodes include the target storage management and control node; the target storage management and control node is used to create the target cloud disk on at least one storage service node of the target cluster and generate the cloud disk identification information; a target storage management and control node determination module, Used to determine the target cluster, the target storage management and control node and the target cloud disk according to the cloud disk identification information.
  • a cloud disk identification information acquisition module used to obtain cloud disk identification information of a target cloud disk
  • the cloud disk identification information is used to identify the target cloud disk and the target storage management and control node and target cluster corresponding to the target cloud disk
  • each storage management and control node is used to manage multiple cloud disk sets
  • the cloud disk identification information includes a cloud disk identification, a cloud disk set identification and a cluster identification.
  • the target storage management and control node determination module is specifically used to: use the cluster identification, cloud disk set identification and cloud disk identification of the target cloud disk to respectively determine the target cluster, the target cloud disk set to which the target cloud disk belongs, and the position of the target cloud disk on the target cloud disk set; based on the mapping relationship between the cloud disk set and the storage management and control node, determine the storage management and control node corresponding to the target cloud disk set as the target storage management and control node.
  • the device may also include: a target storage management and control node selection module, used to select the target storage management and control node from the multiple storage management and control nodes; a cloud disk creation request sending module, used to send a cloud disk creation request to the target storage management and control node, so that the target storage management and control node creates the target cloud disk on the at least one storage service node and generates the cloud disk identification information; a cloud disk identification information receiving module, used to receive the cloud disk identification information returned by the target storage management and control node.
  • a target storage management and control node selection module used to select the target storage management and control node from the multiple storage management and control nodes
  • a cloud disk creation request sending module used to send a cloud disk creation request to the target storage management and control node, so that the target storage management and control node creates the target cloud disk on the at least one storage service node and generates the cloud disk identification information
  • a cloud disk identification information receiving module used to receive the cloud disk identification information returned by the target storage management and control node.
  • the embodiment of the present disclosure provides a cloud disk management and control device, which is applied to a target storage management and control node and may include: a target cloud disk creation module, used to create a target cloud disk on at least one storage service node of a target cluster in response to a cloud disk creation request; the target cluster includes multiple storage management and control nodes, and the multiple storage management and control nodes include the target storage management and control node; a cloud disk identification information generation module, used to generate cloud disk identification information of the target cloud disk; the cloud disk identification information is used to identify the target cloud disk, the target storage management and control node and the target cluster.
  • a target cloud disk creation module used to create a target cloud disk on at least one storage service node of a target cluster in response to a cloud disk creation request
  • the target cluster includes multiple storage management and control nodes, and the multiple storage management and control nodes include the target storage management and control node
  • a cloud disk identification information generation module used to generate cloud disk identification information of the target cloud disk
  • the cloud disk identification information is used
  • each storage management and control node is used to manage multiple cloud disk collections
  • the cloud disk identification information includes a cluster identification, a cloud disk collection identification and a cloud disk identification; wherein the cluster identification of the target cloud disk is used to determine the target cluster; the cloud disk collection identification of the target cloud disk is used to identify the target cloud disk collection to which the target cloud disk belongs, so as to determine the storage management and control node corresponding to the target cloud disk collection as the target storage management and control node based on the mapping relationship between the cloud disk collection and the storage management and control node; the cloud disk identification of the target cloud disk is used to determine the position of the target cloud disk on the target cloud disk collection.
  • the target cloud disk creation module is specifically used to: determine the target cloud disk set from multiple cloud disk sets corresponding to the target storage management and control node based on a preset load strategy; select the target cloud disk from the target cloud disk set, and create the target cloud disk on the at least one storage service node.
  • the device may further include a mounting module, configured to mount the target cloud disk in response to a cloud disk mounting request for the target cloud disk.
  • the multiple storage management and control nodes perform cloud disk scheduling in turn according to a preset order.
  • the device may also include: a traffic information acquisition module, used to obtain the traffic information of each storage service node in the target cluster; a scheduling module, used to schedule at least one cloud disk corresponding to the target storage management and control node based on the traffic information of each storage service node in response to the target storage management and control node's turn to perform cloud disk scheduling.
  • each storage service node of the target cluster is divided into a plurality of storage service node groups, and the plurality of storage management and control nodes correspond one-to-one to the plurality of storage service node groups.
  • the device may further include: a judgment module, Used to determine whether there is a faulty storage service node in the storage service node group corresponding to the target storage management and control node based on the operating status information of each storage service node in the storage service node group corresponding to the target storage management and control node; a downtime operation module is used to perform a downtime operation on the faulty storage service node in response to the existence of the faulty storage service node.
  • the device further includes a fault management and control module for managing each cloud disk corresponding to the faulty storage management and control node in response to the presence of a faulty storage management and control node in the target cluster.
  • each module in each device of the embodiment of the present disclosure can be referred to the corresponding description in the above method, and have corresponding beneficial effects, which will not be repeated here.
  • FIG7 is a block diagram of an electronic device for implementing an embodiment of the present disclosure.
  • the electronic device includes: a memory 701 and a processor 702, wherein the memory 701 stores a computer program that can be run on the processor 702. When the processor 702 executes the computer program, the method in the above embodiment is implemented.
  • the number of the memory 701 and the processor 702 can be one or more.
  • the electronic device also includes: a communication interface 703, which is used to communicate with external devices and perform data exchange transmission.
  • the bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus.
  • ISA Industry Standard Architecture
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in FIG. 7, but it does not mean that there is only one bus or one type of bus.
  • the memory 701, the processor 702 and the communication interface 703 are integrated on a chip, the memory 701, the processor 702 and the communication interface 703 can communicate with each other through an internal interface.
  • An embodiment of the present disclosure provides a computer-readable storage medium storing a computer program, which implements the method provided in any embodiment of the present disclosure when the program is executed by a processor.
  • An embodiment of the present disclosure provides a computer program product, including a computer program.
  • the computer program is executed by a processor, the method provided in any embodiment of the present disclosure is implemented.
  • An embodiment of the present disclosure also provides a chip, which includes a processor for calling and executing instructions stored in the memory from the memory, so that a communication device equipped with the chip executes a method provided by any embodiment of the present disclosure.
  • An embodiment of the present disclosure also provides a chip, including: an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected via an internal connection path, and the processor is used to execute the code in the memory.
  • the processor is used to execute the method provided in any embodiment of the application.
  • the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that Specifically, the processor may be a processor supporting Advanced RISC Machines (ARM) architecture.
  • ARM Advanced RISC Machines
  • the above-mentioned memory may include a read-only memory and a random access memory, and may also include a non-volatile random access memory.
  • the memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories.
  • the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
  • the volatile memory may include a random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting description, many forms of RAM are available.
  • SRAM Static RAM
  • DRAM Dynamic Random Access Memory
  • SDRAM Synchronous DRAM
  • DDR SDRAM Double Data Rate SDRAM
  • ESDRAM Enhanced SDRAM
  • SLDRAM Sync Link DRAM
  • DRRAM Direct Rambus RAM
  • the computer program product includes one or more computer instructions.
  • the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.
  • the user information including but not limited to user device information, user personal information, user operation information, etc.
  • data including but not limited to data used for processing, data for analysis, data for storage, data for display, etc.
  • user information and data are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant information and data must comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.
  • first and second are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated.
  • a feature defined as “first” or “second” may explicitly or implicitly include at least one of the features.
  • the meaning of “plurality” is two or more, unless otherwise clearly and specifically defined.
  • Any process or method description described in the flowchart or otherwise herein can be understood to represent a module, segment or portion of code including one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiments of the present disclosure includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.
  • the logic and/or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or used in combination with these instruction execution systems, devices or apparatuses.
  • each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module.
  • the above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
  • the storage medium may be a read-only memory, a disk or an optical disk, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

一种云盘管控方法、系统、电子设备及存储介质，涉及分布式存储技术领域。该方法包括：获取目标云盘的云盘标识信息；云盘标识信息用于标识目标云盘以及目标云盘对应的目标存储管控节点和目标集群；目标集群包括多个存储管控节点，多个存储管控节点中包括目标存储管控节点；目标存储管控节点用于在目标集群的至少一个存储服务节点上创建目标云盘，并生成云盘标识信息；根据云盘标识信息确定目标集群、目标存储管控节点和目标云盘。该方法可以提高资源利用率，支持大规模集群。

Description

云盘管控方法、系统、电子设备及存储介质
本公开要求于2023年12月12日提交中国专利局、申请号为202311713410.6、申请名称为“云盘管控方法、系统、电子设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本公开中。
技术领域
本公开涉及分布式存储技术领域，尤其涉及一种云盘管控方法、系统、电子设备及存储介质。
背景技术
分布式存储是一种数据存储技术，它将数据分散存储在多个物理位置的网络中的设备上。块存储系统是一种低时延、持久性和高可靠的块级随机分布式存储系统。在块存储系统中，数据被分割成一块(block)一块的，每个块都作为独立的硬盘进行操作。这种方式提供了高性能和灵活性，特别适合处理大量数据，如数据库和大规模的虚拟机环境。在分布式块存储的应用场景下，云盘提供给用户数据块级别的块存储服务，它将物理硬盘的存储空间虚拟化，并通过网络提供给用户，用户可以像使用本地硬盘一样使用云盘，进行数据的读写操作。因此，在块存储系统中，如何实现对云盘的高效管控，减小块存储系统的故障域，显得十分重要。
发明内容
本公开实施例提供一种云盘管控方法、系统、电子设备及存储介质，以缓解或解决现有技术中存在的一项或多项技术问题。
第一方面，本公开实施例提供了一种云盘管控方法，包括：获取目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘以及所述目标云盘对应的目标存储管控节点和目标集群；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；所述目标存储管控节点用于在所述目标集群的至少一个存储服务节点上创建所述目标云盘，并生成所述云盘标识信息；根据所述云盘标识信息确定所述目标集群、所述目标存储管控节点和所述目标云盘。
第二方面，本公开实施例提供了一种云盘管控方法，包括：响应于云盘创建请求，在目标集群的至少一个存储服务节点上创建目标云盘；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；生成所述目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘、所述目标存储管控节点和所述目标集群。
第三方面，本公开实施例提供了一种云盘管控系统，包括：集群管控节点，用于执行本公开第一方面实施例提供的方法；至少一个集群，所述集群包括多个存储管控节点和多个存储服务节点，所述存储管控节点用于执行本公开第二方面实施例提供的方法。
第四方面，本公开实施例提供一种电子设备，包括存储器、处理器及存储在存储器上的计算机程序，处理器在执行计算机程序时实现本公开任一实施例提供的方法。
第五方面，本公开实施例提供一种计算机可读存储介质，计算机可读存储介质内存储有计算机程序，计算机程序被处理器执行时实现本公开任一实施例提供的方法。
第六方面，本公开实施例提供一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现本公开任一实施例提供的方法。
根据本公开实施例的云盘管控方法，通过对一个存储集群内的存储管控节点进行水平扩展，即一个存储集群包括多个存储管控节点，每个存储管控节点负责若干组云盘(例如为多个云盘集合)的管控。一方面，各存储管控节点同时工作，避免了一主多备架构造成的资源浪费，提高了资源利用率，并可以支持大规模集群；另一方面，在某一个存储管控节点故障，存储集群中受影响的云盘仅仅是一部分，因此，爆炸半径远小于集中式BM架构的爆炸半径。进一步地，可以针对分布式存储管控节点的云盘管控架构，提供行之有效的集群资源调度方案，使云盘调度不受架构更新的影响。
上述概述仅仅是为了说明书的目的，并不意图以任何方式进行限制。除上述描述的示意性的方面、实施方式和特征之外，通过参考附图和以下的详细描述，本公开进一步的方面、实施方式和特征将会是容易明白的。
附图说明
在附图中，除非另外规定，否则贯穿多个附图相同的附图标记表示相同或相似的部件或元素。这些附图不一定是按照比例绘制的。应该理解，这些附图仅描绘了根据本公开公开的一些实施方式，而不应将其视为是对本公开范围的限制。
图1示出相关技术中提供的云盘管控系统的架构图；
图2示出本公开实施例提供的云盘管控系统的架构图；
图3示出本公开实施例提供的云盘调度的一个应用示例图；
图4示出本公开实施例提供的云盘调度的另一个应用示例图；
图5示出本公开实施例提供的云盘管控方法的流程图；
图6示出本公开实施例提供的云盘管控方法的流程图；
图7示出本公开实施例提供的电子设备的框图。
具体实施方式
在下文中，仅简单地描述了某些示例性实施例。正如本领域技术人员可认识到的那样，在不脱离本公开的精神或范围的情况下，可通过各种不同方式修改所描述的实施例。因此， 附图和描述被认为本质上是示例性的而非限制性的。
为便于理解本公开实施例的技术方案，以下对本公开实施例的相关技术进行说明，以下相关技术作为可选方案与本公开实施例的技术方案可以进行任意结合，其均属于本公开实施例的保护范围。
下文将使用以下术语解释：
块存储(Block Storage)系统：一种低时延、持久性和高可靠的块级随机存储系统，将数据分为块(block)，每个块都作为独立的硬盘驱动器进行操作。这意味着用户可以自由地在这些块上创建自己的文件系统，从而更好地控制数据环境，特别适合处理大量数据，如数据库和大规模的虚拟机环境。
云盘(Cloud Disk)：基于分布式存储架构，提供数据块级别的块存储服务。云盘的数据被持久化地存储在后端的分布式存储系统中，计算端任何节点的宕机不会对云盘产生影响。
爆炸半径：表征故障域的大小，即某一节点出故障时的影响面。爆炸半径越小，故障域越小；爆炸半径越大，故障域越大。
互联网数据中心(Internet Data Center，IDC)：是存放、管理和分发大量数据的设施。它提供了包括服务器托管、网络连接、数据存储和备份、安全管理等在内的一系列服务。
可用区(Availability Zone)：代表一个物理位置，这个物理位置拥有自己的电力供应、网络连接以及安全设施等。在同一地理区域(Region)内，可能存在一个或多个可用区。每个可用区都是相互独立的，与其他可用区隔离，以防止在一个可用区出现问题时影响其他可用区。比如，如果一个可用区因为电力故障、网络中断或者自然灾害等原因出现问题，其他可用区可以继续正常运行。在云计算环境中，一个可用区包含一个或多个互联网数据中心(Internet Data Center，IDC)。
集群(Cluster)：是一组相互独立的、通过高速计算机网络互联的计算机，它们构成了一个组，并以单一系统的模式加以管理。一个集群中的计算机或服务器通常通过高速网络连接在一起，共享存储、软件或其他资源，如计算集群、存储集群等。
集群管控节点：用于管控各个存储集群的服务节点，如资源管控节点(Resource Master，RM)。
存储管控节点：本公开实施例中，每个存储集群有若干个存储管控节点，如块存储后端控制节点(Block Master，BM)，多个存储管控节点共同管控一个存储集群。
存储服务节点：每个存储集群有多个存储服务节点，每个存储服务节点用于根据存储管控节点的调度，进行数据写入或读出，如块存储后端服务节点(Block Server，BS)。
虚拟块设备(Device)：虚拟化的存储设备，它模拟了物理硬盘驱动器的功能，使得虚拟机可以像访问物理硬盘一样访问存储空间，从而读取或写入云盘中的数据。
创建云盘(创盘)：系统根据用户指定的云盘大小和类型以及当前集群的资源使用情 况，进行资源调度，包括：在选定的存储服务节点上创建一个新的云盘，并设置相关的元数据，如存储卷的大小、类型、状态等；将新创建的云盘注册到集群的元数据服务中，以便用户可以找到和使用它。
挂载云盘(开盘)：将云盘挂载到服务器实例，如弹性计算服务(Elastic Compute Service，ECS)实例，使该云盘作为该服务器实例的数据盘。
垃圾回收(Garbage Collection，GC)：一种自动内存管理技术。在计算机系统中，当对象不再被程序使用时，垃圾回收器会自动回收这些对象所占用的内存空间，以便重新利用。在集群环境中，GC任务是指定期清理无用的数据块或对象，以释放存储空间并提高资源利用率。
相关技术中，存储集群采用集中式BM架构。图1示出相关技术中云盘管控系统的架构图。如图1所示，云盘管控系统通常包括客户端(Client)、集群管控节点RM和至少一个存储集群(图1以存储集群1和存储集群2为示例)，存储集群包括存储管控节点BM和多个存储服务节点BS(图1以BS1、BS2、BS3……为示例)。
一方面，在集中式BM架构中，不管存储集群的规模大小，一个存储集群部署一组BM，单组BM管理整个存储集群相关的客户端、BS和云盘的元数据，负责调度云盘分片(即Segment，图1以SEG0、SEG1、SEG2……为示例)以及分配垃圾回收等后台任务，完成集群级别的所有管控工作。单组BM采用一主两备的服务方式，即在正常工作状态时，由一个主BM来完成存储集群的所有管控工作，当主BM宕机时，由备BM接管存储集群的管控工作。因此，对于单个集群来说，只有一台BM工作，其他BM处于休眠状态，造成了资源的浪费。而且，在某一个存储集群的一组BM异常的时候，另外一个存储集群即使有充足的机器，也没办法支持这个异常的存储集群。
另一方面，这种集中管理的方式架构相对简单，客户端通过BM获取Segment的位置，然后就可以去对应的BS上读写数据。集中式BM负责整个存储集群的管理、调度和运维相关工作，在单集群云盘数目较少、Segment数目有限和集群规模较小的情况下可以比较高效的工作。但是，随着规模的增大，未来单集群预计需要支持更多的Segment和云盘，以未来要支持到1000万Segment来计算，那就需要单个BM提供360G内存，很明显现有的服务器是不能支持这一需求的，而且，集中式BM一旦发生宕机，整个集群粒度的管控操作都会失败，直到一主两备三个BM中的一个可以正常启动提供服务为止。因此，集中式BM架构的存储集群，其爆炸半径是整个存储集群的所有云盘。
本公开实施例旨在提供一种基于分布式BM的云盘管控方案。图2示出本公开实施例一种实施方式的云盘管控系统200的架构图。如图2所示，云盘管控系统200包括集群管控节点201和一个或多个集群202。
其中，集群管控节点201为管控各个集群202的服务节点。示例性地，集群管控节点201可以为资源管控节点RM。示例性地，当计算节点创建对应的存储资源时，由管理各个存储集群的集群管控节点201从各集群202中决策出其中一个作为目标集群。
集群202为存储集群。对于一个集群202，其部署包括多个存储管控节点和多个存储服务节点。示例性地，存储管控节点可以为块存储后端控制节点(BM)，如图2所示，多个存储管控节点例如为BM1、BM2、BM3和BM4。存储服务节点可以为块存储后端服务节点(BS)，如图2所示，多个存储服务节点例如为BS1、BS2、BS3、BS4、BS5、BS6、BS7和BS8。需要说明的是，本公开实施例对集群202的个数，以及每个集群所部署的存储管控节点和存储服务节点的个数均不作具体限定。
每个存储管控节点用于管控多个云盘集合(partition)。例如，BM1可以管控云盘集合1和云盘集合8(对于云盘集合列表1)，BM2可以管控云盘集合2和云盘集合3(对于云盘集合列表2)，BM3可以管控云盘集合4和云盘集合6(对于云盘集合列表3)，BM4可以管控云盘集合5和云盘集合7(对于云盘集合列表4)。对于一个集群202来说，多个存储管控节点与多个云盘集合列表(partition list)具有一一对应的映射关系，也即云盘集合与存储管控节点具有映射关系。
其中，云盘集合是一个逻辑概念，是一批云盘的集合，某一块云盘只能属于一个云盘集合。云盘集合被创建后，需要由对应的存储管控节点加载(load)后才能提供相应的服务。示例性地，为了逻辑简单，云盘集合(partition)支持动态创建，但是不能修改和删除。
存储管控节点对其相应云盘集合的管控，包括但不限于：创建云盘，并为创建的云盘生成相应的云盘标识信息(Volume ID)；基于云盘标识信息挂载云盘(开盘)、删除云盘、调度云盘等。
其中，云盘标识信息包括集群标识(Cluster ID)、云盘集合标识(Partition ID)和云盘标识(云盘在云盘集合上的位置)。云盘在云盘集合上的位置可以用云盘编号或云盘随机数表示，对于一个云盘集合来说，该云盘集合中的每个云盘具有唯一的云盘编号或唯一云盘随机数。示例性地，云盘标识信息还可以包括预留字段(或叫做预设字段)，预留字段可以标识云盘的预留功能或预设功能或其他信息等。
对于一个集群202，其部署还可以包括中心管控节点(Central Master，CM)，用于从集群202的全局角度负责旁路化的管控工作，包括：管理该集群202中BM的上下线；管理该集群202中BS的上下线；分配创建云盘集合，为其创建和配置相应的资源等。
示例性地，云盘管控系统200还可以包括客户端203，可以部署在计算集群，通过与集群管控节点201交互，获取云盘标识信息，进而基于云盘标识信息访问相应的云盘，如挂载云盘。客户端203可以通过应用程序接口(Application Programming Interface，API)与存储管控节点进行交互，得到Segment所在的存储服务节点的信息，进而去该存储服务节点上读写数据。
下面结合图2描述基于本公开实施例的云盘管控系统200进行动态创盘流程的应用示例。创盘流程主要包括：
(1)集群管控节点201从集群201的多个存储管控节点中，选择一个作为目标存储 管控节点(例如为BM1)，选择策略可以是随机选择，也可以是基于哈希(hash)规则；
(2)集群管控节点201向BM1发送云盘创建请求；
(3)BM1可以配置有代理或管控组件(proxy)，BM1的proxy从多个云盘集合中选择一个作为目标云盘集合(例如为云盘集合8)；
(4)BM1完成目标云盘的创建，生成目标云盘的云盘标识信息(Volume ID)，并通过原路径向集群管控节点201返回创盘成功的信息以及该目标云盘的云盘标识信息。
下面结合图2描述基于本公开实施例的云盘管控系统200进行动态开盘流程的应用示例。开盘流程主要包括：
(1)客户端203通过虚拟块设备的设备标识信息(DEVID)，向集群管控节点201查询目标集群的信息、目标云盘的云盘标识信息和目标存储管控节点的信息；
(2)集群管控节点201基于客户端203给出的设备标识信息，解析出对应的云盘标识信息(Volume ID)，通过解析Volume ID，得到集群标识(Cluster ID)、云盘集合标识(Partition ID)和云盘标识(云盘在云盘集合上的位置)，然后基于云盘集合与存储管控节点的映射关系，确定与云盘集合标识(Partition ID)对应的存储管控节点(例如为BM1)作为目标存储管控节点；
(3)集群管控节点201向客户端203返回目标集群的集群标识(例如为集群名称)、BM1的互联网协议(Internet Protocol，IP)地址、Volume ID；
(4)客户端203基于BM1的IP地址，向BM1发送针对目标云盘的云盘挂载请求；
(5)BM1响应云盘挂载请求，挂载目标云盘。
下面结合图3描述基于本公开实施例的云盘管控系统进行云盘调度的一个应用示例。该应用示例中，云盘调度流程主要包括：
(1)集群中的每一个BM获取该集群各BS的流量信息。获取方式可以是集群中各BS向中心管控节点(CM)发送心跳，上报各自的流量信息；获取方式也可以是集群中各BM分别向集群中各BS发送慢速心跳，从而每个BM都能得到各BS的流量信息，慢速心跳例如为每10秒发送一次；
(2)各BM按照预设顺序轮流执行云盘调度，例如，BM1、BM2、BM3和BM4轮流执行云盘调度，包括流量均衡调度(动态或静态)、从而将云盘的云盘分片(Segment)调度到多个存储服务节点上。示例性地，CM用于从集群的全局角度按照全局协调机制，控制各BM按照预设顺序轮流执行云盘调度，并进行一定的干涉，如果CM宕机，则各BM一直轮流执行云盘调度。
下面结合图4描述基于本公开实施例的云盘管控系统进行云盘调度的另一个应用示例。该应用示例中，云盘调度流程主要包括：
(1)集群中各BS划分为多组，形成个存储服务节点组，各BM与各存储服务节点组一一对应，BM拥有与其对应的存储服务节点组的管辖权，包括对该存储服务节点组的BS发起宕机操作。例如：BS1和BS2形成组1，BM1与组1对应，拥有对BS1和BS2的管辖权； BS3和BS4形成组2，BM2与组2对应，拥有对BS3和BS4的管辖权；BS5和BS6形成组3，BM3与组3对应，拥有对BS5和BS6的管辖权；BS7和BS8形成组4，BM4与组4对应，拥有对BS7和BS8的管辖权；
(2)BM向与其对应的BS发送快速心跳，如每200毫秒一次，从而得到相应BS的运行状态信息，用于对BS进行故障判断(宕机判断)；
(3)如果BM判断出组内BS中存在故障BS，即存在宕机的BS，则对该BS进行宕机操作，例如，BM1判断出组1中的BS2宕机，则BM1会对BS进行宕机操作，实现宕机调度(也叫做故障调度或异常调度)。
下面结合图2描述本公开实施例的云盘管控系统的心跳部署的应用示例。如图2所示，针对一个集群202内，各BM向RM报心跳，上报各自的运行状态信息，即确定该BM未故障；集群粒度的信息设置由该集群中的一个BM(例如默认为BM1)上报。中心管控节点向各BM报心跳，获取各BM的运行状态信息，如是否故障，以及获取BM上各云盘集合的负载状况。
本公开实施例的云盘管控系统提供了一种基于分布式存储管控节点的云盘管控方案，对一个存储集群内的存储管控节点进行水平扩展，即一个存储集群包括多个存储管控节点，每个存储管控节点负责若干个云盘集合的管控。一方面，各存储管控节点同时工作，避免了一主多备架构造成的资源浪费，提高了资源利用率；另一方面，在某一个存储管控节点故障，由其他存储管控节点接管其负责的云盘集合，在此过程中，存储集群中受影响的云盘仅仅是一部分，因此，爆炸半径远小于集中式BM架构的爆炸半径；又一方面，可以支持大规模集群，经仿真本公开实施例的方案可以支持200台甚至1000台这样更大规模的集群；再一方面，本公开实施例还针对分布式存储管控节点的云盘管控架构，提供了行之有效的集群资源调度方案，使云盘调度不受架构更新的影响。
需要说明的是，本公开实施例中提供的上述应用场景是为了便于理解，本公开实施例对技术方案的应用场景不作具体限定。此外，本公开所涉及的用户信息(包括但不限于用户设备信息、用户个人信息等)和数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)，均为经用户授权或者经过各方充分授权的信息和数据，并且相关数据的收集、使用和处理需要遵守相关国家和地区的相关法律法规和标准，并提供有相应的操作入口，供用户选择或编辑授权或者拒绝。
下面以具体的实施例对本公开的技术方案以及本公开的技术方案如何解决前述技术问题进行详细说明。所列举的若干具体的实施例可以相互结合，对于相同或相似的概念或过程可能在某些实施例中不再赘述。
图5示出根据本公开实施例的云盘管控方法的流程图。该云盘管控方法可以应用于集群管控节点。如图5所示，该云盘管控方法包括：
步骤S501：获取目标云盘的云盘标识信息；云盘标识信息用于标识目标云盘以及目标云盘对应的目标存储管控节点和目标集群；目标集群包括多个存储管控节点，多个存储管 控节点中包括目标存储管控节点；目标存储管控节点用于在目标集群的至少一个存储服务节点上创建目标云盘，并生成云盘标识信息。
本公开实施例中，目标集群为存储集群，例如为图2所示中的集群202。目标集群包括多个存储管控节点和多个存储服务节点。例如，如图2所示，存储管控节点可以为块存储后端控制节点，如图2所示，多个存储管控节点例如为BM1、BM2、BM3和BM4。存储服务节点可以为块存储后端服务节点，如图2所示，多个存储服务节点例如为BS1、BS2、BS3、BS4、BS5、BS6、BS7和BS8。多个存储管控节点中包括目标存储管控节点，目标存储管控节点为创建目标云盘的存储管控节点，且生成目标云盘的云盘标识信息(Volume ID)。
步骤S502：根据云盘标识信息确定目标集群、目标存储管控节点和目标云盘。
云盘标识信息包括：用于确定目标集群的信息，如集群标识(Cluster ID)；用于确定目标存储管控节点的信息；用于确定目标云盘的信息。基于目标云盘的云盘标识信息即可确定该目标云盘对应的目标集群、目标存储管控节点以及目标云盘，进而使目标存储管控节点对目标云盘进行相应的管控，包括但不限于挂载、删除、调度等管控。
在一种实施方式中，每个存储管控节点用于管控多个云盘集合(partition)，云盘标识信息包括云盘标识、云盘集合标识和集群标识，在步骤S501中，根据云盘标识信息确定目标集群、目标存储管控节点和目标云盘，可以包括：利用目标云盘的集群标识、云盘集合标识和云盘标识，分别确定目标集群、目标云盘所属的目标云盘集合以及目标云盘在目标云盘集合上的位置；基于云盘集合与存储管控节点的映射关系，确定与目标云盘集合对应的存储管控节点作为目标存储管控节点。
其中，partition是一个逻辑概念，是一批云盘的集合，某一块云盘只能属于一个partition。partition被创建后，需要由对应的存储管控节点加载(load)后才能提供相应的服务。示例性地，为了逻辑简单，partition支持动态创建，但是不能修改和删除。
每个存储管控节点用于管控多个partition，对于一个集群来说，partition与存储管控节点具有映射关系。因此，利用目标云盘的partition标识即可确定目标云盘所属的partition，进而基于partition与存储管控节点的映射关系，得到相应的目标存储管控节点的信息。示例性地，一个partition与BM的IP地址具有映射关系，基于目标云盘的partition标识，可以确定目标BM的IP地址。
在一种实施方式中，本公开实施例的方法还可以包括：从多个存储管控节点中，选择目标存储管控节点；向目标存储管控节点发送云盘创建请求，以使目标存储管控节点在至少一个存储服务节点上创建目标云盘，并生成云盘标识信息；接收目标存储管控节点返回的云盘标识信息。
示例性地，当计算节点创建对应的存储资源时，由集群管控节点从各集群中决策出其中一个作为目标集群；集群管控节点可以从目标集群的多个存储管控节点中，选择一个作为目标存储管控节点，选择策略可以是随机选择，也可以是基于哈希规则，本公开实施例对此不作具体限定；进一步地，集群管控节点向目标存储管控节点发送云盘创建请求，由 目标存储管控节点完成目标云盘的创建，生成目标云盘的云盘标识信息(Volume ID)，并向集群管控节点返回该目标云盘的Volume ID。
在一种实施方式中，客户端可以通过虚拟块设备的设备标识信息(DEVID)，向集群管控节点查询目标集群的信息、目标云盘的云盘标识信息和目标存储管控节点的信息。
示例性地，集群管控节点可以基于设备标识信息，解析出对应的Volume ID，通过解析Volume ID，得到集群标识(Cluster ID)、云盘集合标识(Partition ID)和云盘标识(云盘在云盘集合上的位置)，然后基于云盘集合与存储管控节点的映射关系，确定与云盘集合标识(Partition ID)对应的目标存储管控节点的IP地址，并向客户端返回目标集群的集群标识(例如为集群名称)、目标存储管控节点的IP地址以及Volume ID；客户端基于这些信息，可以向目标存储管控节点发送针对目标云盘的管控请求，如云盘挂载请求、云盘删除请求等。
图6示出根据本公开实施例的云盘管控方法的流程图。该云盘管控方法可以应用于存储管控节点，本公开实施例以目标存储管控节点为示例，即创建目标云盘的存储管控节点。如图6所示，该云盘管控方法包括：
步骤S601：响应于云盘创建请求，在目标集群的至少一个存储服务节点上创建目标云盘；目标集群包括多个存储管控节点，多个存储管控节点中包括目标存储管控节点；
步骤S602：生成目标云盘的云盘标识信息；云盘标识信息用于标识目标云盘、目标存储管控节点和目标集群。
其中，每个存储管控节点用于管控多个云盘集合(partition)，云盘标识信息包括集群标识、partition标识和云盘标识；其中，目标云盘的集群标识用于确定目标集群；目标云盘的partition标识用于标识目标云盘所属的目标partition，以基于partition与存储管控节点的映射关系，确定与目标partition对应的存储管控节点作为目标存储管控节点；目标云盘的云盘标识用于确定目标云盘在目标partition上的位置。
在步骤S601中，在目标集群的至少一个存储服务节点上创建目标云盘，包括：基于预设的负载策略，从目标存储管控节点对应的多个partition中确定目标partition；从目标partition中选择目标云盘，并在至少一个存储服务节点上创建目标云盘。
示例性地，存储管控节点响应于云盘创建请求，从多个partition选择一个作为目标partition。选择策略可以是负载均衡策略，即根据各partition已经创建云盘的数量，选择负载少的partition作为目标partition。选择策略也可以是随机选择策略，选择策略也可以是基于距离优先的策略，即选择与计算节点网络拓扑距离较近的partition作为目标partition，本公开实施例对此不作限定。
其中，存储管控节点可以配置有代理或管控组件，用于从多个partition选择一个作为目标partition，并将相应的操作请求发给对应的存储管控节点。具体可参见上述创盘流程的应用示例，在此不再赘述。
在一种实施方式中，本公开实施例的方法还可以包括：响应于针对目标云盘的云盘挂 载请求，挂载目标云盘。具体可参见上述动态开盘流程的应用示例，在此不再赘述。
根据本公开实施例的云盘管控方法，通过对一个存储集群内的存储管控节点进行水平扩展，即一个存储集群包括多个存储管控节点，每个存储管控节点负责若干组云盘(例如为多个云盘集合)的管控。一方面，各存储管控节点同时工作，避免了一主多备架构造成的资源浪费，提高了资源利用率，并可以支持大规模集群；另一方面，在某一个存储管控节点故障的情况下，存储集群中受影响的云盘仅仅是一部分，因此，爆炸半径远小于集中式BM(一主多备)架构的爆炸半径。
在一种实施方式中，多个存储管控节点按照预设顺序轮流执行云盘调度，本公开实施例由目标存储管控节点执行的方法还可以包括：获取目标集群中各存储服务节点的流量信息；响应于轮到目标存储管控节点执行云盘调度，基于各存储服务节点的流量信息，对目标存储管控节点对应的至少一个云盘进行调度。
示例性地，目标集群中的每一个存储管控节点获取目标集群各存储服务节点的流量信息。获取方式可以是目标集群中各存储服务节点向中心管控节点(CM)发送心跳，上报各自的流量信息；获取方式也可以是目标集群中各存储管控节点分别向目标集群中各存储服务节点发送慢速心跳，从而目标存储管控节点可以得到目标集群中每个存储服务节点的流量信息。进一步地，各存储管控节点按照预设顺序轮流执行云盘调度，当轮到目标存储管控节点进行云盘调度时，目标存储管控节点可以对其管控的各part it ion上的云盘进行云盘调度，包括动态流量均衡调度或静态流量均衡调度。
在一种实施方式中，目标集群的各存储服务节点划分为多个存储服务节点组，多个存储管控节点与多个存储服务节点组一一对应，本公开实施例由目标存储管控节点执行的方法还可以包括：基于目标存储管控节点对应的存储服务节点组中各存储服务节点的运行状态信息，判断目标存储管控节点对应的存储服务节点组中是否存在故障存储服务节点；响应于存在故障存储服务节点，对故障存储服务节点执行宕机操作。具体可参见结合图4的云盘调度应用示例，在此不再赘述。
在一种实施方式中，本公开实施例由目标存储管控节点执行的方法还可以包括：响应于目标集群中存在故障存储管控节点，管控故障存储管控节点对应的各云盘。也就是说，在某一个存储管控节点故障，由其他存储管控节点接管其负责的云盘。
根据本公开实施例的云盘管控方法，可以针对分布式存储管控节点的云盘管控架构，提供行之有效的集群资源调度方案，使云盘调度不受架构更新的影响。
与本公开实施例提供的方法相对应地，本公开实施例提供一种云盘管控装置，该云盘管控装置应用于集群管控节点，可以包括：云盘标识信息获取模块，用于获取目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘以及所述目标云盘对应的目标存储管控节点和目标集群；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；所述目标存储管控节点用于在所述目标集群的至少一个存储服务节点上创建所述目标云盘，并生成所述云盘标识信息；目标存储管控节点确定模块， 用于根据所述云盘标识信息确定所述目标集群、所述目标存储管控节点和所述目标云盘。
在一种实施方式中，每个存储管控节点用于管控多个云盘集合，所述云盘标识信息包括云盘标识、云盘集合标识和集群标识，目标存储管控节点确定模块具体用于：利用所述目标云盘的集群标识、云盘集合标识和云盘标识，分别确定所述目标集群、所述目标云盘所属的目标云盘集合以及所述目标云盘在所述目标云盘集合上的位置；基于云盘集合与存储管控节点的映射关系，确定与所述目标云盘集合对应的存储管控节点作为所述目标存储管控节点。
在一种实施方式中，该装置还可以包括：目标存储管控节点选择模块，用于从所述多个存储管控节点中，选择所述目标存储管控节点；云盘创建请求发送模块，用于向所述目标存储管控节点发送云盘创建请求，以使所述目标存储管控节点在所述至少一个存储服务节点上创建所述目标云盘，并生成所述云盘标识信息；云盘标识信息接收模块，用于接收所述目标存储管控节点返回的所述云盘标识信息。
与本公开实施例提供的方法相对应地，本公开实施例提供一种云盘管控装置，该云盘管控装置应用于目标存储管控节点，可以包括：目标云盘创建模块，用于响应于云盘创建请求，在目标集群的至少一个存储服务节点上创建目标云盘；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；云盘标识信息生成模块，用于生成所述目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘、所述目标存储管控节点和所述目标集群。
在一种实施方式中，每个存储管控节点用于管控多个云盘集合，所述云盘标识信息包括集群标识、云盘集合标识和云盘标识；其中，所述目标云盘的集群标识用于确定所述目标集群；所述目标云盘的云盘集合标识用于标识所述目标云盘所属的目标云盘集合，以基于云盘集合与存储管控节点的映射关系，确定与所述目标云盘集合对应的存储管控节点作为所述目标存储管控节点；所述目标云盘的云盘标识用于确定所述目标云盘在所述目标云盘集合上的位置。
在一种实施方式中，目标云盘创建模块具体用于：基于预设的负载策略，从所述目标存储管控节点对应的多个云盘集合中确定所述目标云盘集合；从所述目标云盘集合中选择所述目标云盘，并在所述至少一个存储服务节点上创建所述目标云盘。
在一种实施方式中，该装置还可以包括挂载模块，用于响应于针对所述目标云盘的云盘挂载请求，挂载所述目标云盘。
在一种实施方式中，所述多个存储管控节点按照预设顺序轮流执行云盘调度，该装置还可以包括：流量信息获取模块，用于获取所述目标集群中各存储服务节点的流量信息；调度模块，用于响应于轮到所述目标存储管控节点执行云盘调度，基于各存储服务节点的流量信息，对所述目标存储管控节点对应的至少一个云盘进行调度。
在一种实施方式中，所述目标集群的各存储服务节点划分为多个存储服务节点组，所述多个存储管控节点与所述多个存储服务节点组一一对应，该装置还可以包括：判断模块， 用于基于所述目标存储管控节点对应的存储服务节点组中各存储服务节点的运行状态信息，判断所述目标存储管控节点对应的存储服务节点组中是否存在故障存储服务节点；宕机操作模块，用于响应于存在所述故障存储服务节点，对所述故障存储服务节点执行宕机操作。
在一种实施方式中，该装置还包括故障管控模块，用于响应于所述目标集群中存在故障存储管控节点，管控所述故障存储管控节点对应的各云盘。
本公开实施例各装置中的各模块的功能可以参见上述方法中的对应描述，并具备相应的有益效果，在此不再赘述。
图7为用来实现本公开实施例的电子设备的框图。如图7所示，该电子设备包括：存储器701和处理器702，存储器701内存储有可在处理器702上运行的计算机程序。处理器702执行该计算机程序时实现上述实施例中的方法。存储器701和处理器702的数量可以为一个或多个。
该电子设备还包括：通信接口703，用于与外界设备进行通信，进行数据交互传输。
如果存储器701、处理器702和通信接口703独立实现，则存储器701、处理器702和通信接口703可以通过总线相互连接并完成相互间的通信。该总线可以是工业标准体系结构(Industry Standard Architecture，ISA)总线、外部设备互连(Peripheral Component Interconnect，PCI)总线或扩展工业标准体系结构(Extended Industry Standard Architecture，EISA)总线等。该总线可以分为地址总线、数据总线、控制总线等。为便于表示，图7中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
可选地，在具体实现上，如果存储器701、处理器702及通信接口703集成在一块芯片上，则存储器701、处理器702及通信接口703可以通过内部接口完成相互间的通信。
本公开实施例提供了一种计算机可读存储介质，其存储有计算机程序，该程序被处理器执行时实现本公开任一实施例中提供的方法。
本公开实施例提供了一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现本公开任一实施例中提供的方法。
本公开实施例还提供了一种芯片，该芯片包括，包括处理器，用于从存储器中调用并运行存储器中存储的指令，使得安装有芯片的通信设备执行本公开任一实施例提供的方法。
本公开实施例还提供了一种芯片，包括：输入接口、输出接口、处理器和存储器，输入接口、输出接口、处理器以及存储器之间通过内部连接通路相连，处理器用于执行存储器中的代码，当代码被执行时，处理器用于执行申请任一实施例提供的方法。
应理解的是，上述处理器可以是中央处理器(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processing，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field Programmable gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者是任何常规的处理器等。值得说明 的是，处理器可以是支持进阶精简指令集机器(Advanced RISC Machines，ARM)架构的处理器。
可选地，上述存储器可以包括只读存储器和随机存取存储器，还可以包括非易失性随机存取存储器。该存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以包括只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以包括随机存取存储器(Random Access Memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用。例如：静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic Random Access Memory，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Sync Link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行计算机程序指令时，全部或部分地产生按照本公开的流程或功能。计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输。
需要说明的是，本公开所涉及的用户信息(包括但不限于用户设备信息、用户个人信息、用户操作信息等)和数据(包括但不限于用于处理的数据、分析的数据、存储的数据、展示的数据等)等，均为经用户授权或者经过各方充分授权的信息和数据，并且相关信息和数据的收集、使用和处理需要遵守相关国家和地区的相关法律法规和标准，并提供有相应的操作入口，供用户选择授权或者拒绝。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包括于本公开的至少一个实施例或示例中。而且，描述的具体特征、结构、材料或者特点可以在任一个或多个实施例或示例中以合适的方式结合。此外，在不相互矛盾的情况下，本领域的技术人员可以将本说明书中描述的不同实施例或示例以及不同实施例或示例的特征进行结合和组合。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或隐含地包括至少一个该特征。在本公开的描述中，“多个”的含义是两个或两个以上，除非另有明确具体的限定。
流程图中或在此以其他方式描述的任何过程或方法描述可以被理解为，表示包括一个或更多个用于实现特定逻辑功能或过程的步骤的可执行指令的代码的模块、片段或部分。并且本公开的优选实施方式的范围包括另外的实现，其中可以不按所示出或讨论的顺序，包括根据所涉及的功能按基本同时的方式或按相反的顺序，来执行功能。
在流程图中表示或在此以其他方式描述的逻辑和/或步骤，例如：可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行系统、装置或设备(如基于计算机的系统、包括处理器的系统或其他可以从指令执行系统、装置或设备取指令并执行指令的系统)使用，或结合这些指令执行系统、装置或设备而使用。
应理解的是，本公开的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行系统执行的软件或固件来实现。上述实施例方法的全部或部分步骤是可以通过程序来指令相关的硬件完成，该程序可以存储于一种计算机可读存储介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。
此外，在本公开各个实施例中的各功能单元可以集成在一个处理模块中，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模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。上述集成的模块如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读存储介质中。该存储介质可以是只读存储器，磁盘或光盘等。
以上所述，仅为本公开的具体实施方式，但本公开的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，可轻易想到其各种变化或替换，这些都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应以权利要求的保护范围为准。

Claims (16)

  1. 一种云盘管控方法，应用于集群管控节点，所述方法包括：
    获取目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘以及所述目标云盘对应的目标存储管控节点和目标集群；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；所述目标存储管控节点用于在所述目标集群的至少一个存储服务节点上创建所述目标云盘，并生成所述云盘标识信息；
    根据所述云盘标识信息确定所述目标集群、所述目标存储管控节点和所述目标云盘。
  2. 根据权利要求1所述的方法，其中，每个存储管控节点用于管控多个云盘集合，所述云盘标识信息包括云盘标识、云盘集合标识和集群标识，根据所述云盘标识信息确定所述目标集群、所述目标存储管控节点和所述目标云盘，包括：
    利用所述目标云盘的集群标识、云盘集合标识和云盘标识，分别确定所述目标集群、所述目标云盘所属的目标云盘集合以及所述目标云盘在所述目标云盘集合上的位置；
    基于云盘集合与存储管控节点的映射关系，确定与所述目标云盘集合对应的存储管控节点作为所述目标存储管控节点。
  3. 根据权利要求1所述的方法，还包括：
    从所述多个存储管控节点中，选择所述目标存储管控节点；
    向所述目标存储管控节点发送云盘创建请求，以使所述目标存储管控节点在所述至少一个存储服务节点上创建所述目标云盘，并生成所述云盘标识信息；
    接收所述目标存储管控节点返回的所述云盘标识信息。
  4. 一种云盘管控方法，应用于目标存储管控节点，所述方法包括：
    响应于云盘创建请求，在目标集群的至少一个存储服务节点上创建目标云盘；所述目标集群包括多个存储管控节点，所述多个存储管控节点中包括所述目标存储管控节点；
    生成所述目标云盘的云盘标识信息；所述云盘标识信息用于标识所述目标云盘、所述目标存储管控节点和所述目标集群。
  5. 根据权利要求4所述的方法，其中，每个存储管控节点用于管控多个云盘集合，所述云盘标识信息包括集群标识、云盘集合标识和云盘标识；其中，所述目标云盘的集群标识用于确定所述目标集群；所述目标云盘的云盘集合标识用于标识所述目标云盘所属的目标云盘集合，以基于云盘集合与存储管控节点的映射关系，确定与所述目标云盘集合对应的存储管控节点作为所述目标存储管控节点；所述目标云盘的云盘标识用于确定所述目标云盘在所述目标云盘集合上的位置。
  6. 根据权利要求5所述的方法，其中，在目标集群的至少一个存储服务节点上 创建目标云盘，包括：
    基于预设的负载策略，从所述目标存储管控节点对应的多个云盘集合中确定所述目标云盘集合；
    从所述目标云盘集合中选择所述目标云盘，并在所述至少一个存储服务节点上创建所述目标云盘。
  7. 根据权利要求4所述的方法，其中，还包括：
    响应于针对所述目标云盘的云盘挂载请求，挂载所述目标云盘。
  8. 根据权利要求4至7任一项所述的方法，其中，所述多个存储管控节点按照预设顺序轮流执行云盘调度，所述方法还包括：
    获取所述目标集群中各存储服务节点的流量信息；
    响应于轮到所述目标存储管控节点执行云盘调度，基于各存储服务节点的流量信息，对所述目标存储管控节点对应的至少一个云盘进行调度。
  9. 根据权利要求4至7任一项所述的方法，其中，所述目标集群的各存储服务节点划分为多个存储服务节点组，所述多个存储管控节点与所述多个存储服务节点组一一对应，所述方法还包括：
    基于所述目标存储管控节点对应的存储服务节点组中各存储服务节点的运行状态信息，判断所述目标存储管控节点对应的存储服务节点组中是否存在故障存储服务节点。
  10. 根据权利要求9所述的方法，还包括：
    响应于存在所述故障存储服务节点，对所述故障存储服务节点执行宕机操作。
  11. 根据权利要求4至7任一项所述的方法，还包括：
    响应于所述目标集群中存在故障存储管控节点，管控所述故障存储管控节点对应的各云盘。
  12. 一种云盘管控系统，包括：
    集群管控节点，用于执行权利要求1至3任一项所述的方法；
    至少一个集群，所述集群包括多个存储管控节点和多个存储服务节点，所述存储管控节点用于执行权利要求4至11任一项所述的方法。
  13. 根据权利要求12所述的云盘管控系统，其中，每个集群还包括中心管控节点，用于将集群中各存储服务节点的流量信息发送给各存储管控节点，并控制集群中各存储管控节点按照预设顺序轮流执行云盘调度；和/或，所述中心管控节点用于为每个存储管控节点分配创建多个云盘集合。
  14. 一种电子设备，包括存储器、处理器及存储在存储器上的计算机程序，所述处理器在执行所述计算机程序时实现权利要求1至11中任一项所述的方法。
  15. 一种计算机可读存储介质，计算机可读存储介质内存储有计算机程序，计算机程序被处理器执行时实现权利要求1至11中任一项所述的方法。
  16. 一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现如权利要求1至11中任一项所述的方法。
PCT/CN2024/114916 2023-12-12 2024-08-27 云盘管控方法、系统、电子设备及存储介质 Pending WO2025123783A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202311713410.6 2023-12-12
CN202311713410.6A CN117714476A (zh) 2023-12-12 2023-12-12 云盘管控方法、系统、电子设备及存储介质

Publications (1)

Publication Number Publication Date
WO2025123783A1 true WO2025123783A1 (zh) 2025-06-19

Family

ID=90152757

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/114916 Pending WO2025123783A1 (zh) 2023-12-12 2024-08-27 云盘管控方法、系统、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN117714476A (zh)
WO (1) WO2025123783A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117714476A (zh) * 2023-12-12 2024-03-15 阿里云计算有限公司 云盘管控方法、系统、电子设备及存储介质
CN118796468B (zh) * 2024-06-28 2025-06-17 北京火山引擎科技有限公司 磁盘簇资源调度方法、装置、系统、设备及存储介质

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108241469A (zh) * 2016-12-27 2018-07-03 北京金山云网络技术有限公司 一种数据存储方法、装置及系统
CN110708345A (zh) * 2018-07-09 2020-01-17 阿里巴巴集团控股有限公司 云盘升级方法、调度方法及云主机、调度装置和系统
WO2020088340A1 (zh) * 2018-11-01 2020-05-07 阿里巴巴集团控股有限公司 一种提供云计算服务的方法及系统
CN113282564A (zh) * 2021-07-24 2021-08-20 阿里云计算有限公司 数据存储方法、系统、节点和存储介质
CN114697331A (zh) * 2022-03-29 2022-07-01 浪潮云信息技术股份公司 一种云盘挂载数量调度方法、系统、装置及存储介质
CN115168287A (zh) * 2022-06-17 2022-10-11 阿里巴巴（中国）有限公司 数据处理方法以及装置
CN115827148A (zh) * 2022-08-25 2023-03-21 杭州海康威视系统技术有限公司 一种资源管理方法、装置、电子设备及存储介质
CN115883662A (zh) * 2022-11-29 2023-03-31 阿里巴巴（中国）有限公司 存储系统及其调度方法、装置、计算机设备和存储介质
CN116436934A (zh) * 2023-04-10 2023-07-14 阿里巴巴（中国）有限公司 云盘调度方法、系统、电子设备及存储介质
CN117714476A (zh) * 2023-12-12 2024-03-15 阿里云计算有限公司 云盘管控方法、系统、电子设备及存储介质

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108241469A (zh) * 2016-12-27 2018-07-03 北京金山云网络技术有限公司 一种数据存储方法、装置及系统
CN110708345A (zh) * 2018-07-09 2020-01-17 阿里巴巴集团控股有限公司 云盘升级方法、调度方法及云主机、调度装置和系统
WO2020088340A1 (zh) * 2018-11-01 2020-05-07 阿里巴巴集团控股有限公司 一种提供云计算服务的方法及系统
CN113282564A (zh) * 2021-07-24 2021-08-20 阿里云计算有限公司 数据存储方法、系统、节点和存储介质
CN114697331A (zh) * 2022-03-29 2022-07-01 浪潮云信息技术股份公司 一种云盘挂载数量调度方法、系统、装置及存储介质
CN115168287A (zh) * 2022-06-17 2022-10-11 阿里巴巴（中国）有限公司 数据处理方法以及装置
CN115827148A (zh) * 2022-08-25 2023-03-21 杭州海康威视系统技术有限公司 一种资源管理方法、装置、电子设备及存储介质
CN115883662A (zh) * 2022-11-29 2023-03-31 阿里巴巴（中国）有限公司 存储系统及其调度方法、装置、计算机设备和存储介质
CN116436934A (zh) * 2023-04-10 2023-07-14 阿里巴巴（中国）有限公司 云盘调度方法、系统、电子设备及存储介质
CN117714476A (zh) * 2023-12-12 2024-03-15 阿里云计算有限公司 云盘管控方法、系统、电子设备及存储介质

Also Published As

Publication number Publication date
CN117714476A (zh) 2024-03-15

Similar Documents

Publication Publication Date Title
US11226847B2 (en) Implementing an application manifest in a node-specific manner using an intent-based orchestrator
US8386540B1 (en) Scalable relational database service
US6249883B1 (en) System and method for monitoring domain controllers
US20190220361A1 (en) Monitoring Containers In A Distributed Computing System
EP3811597A1 (en) Zone redundant computing services using multiple local services in distributed computing systems
US12038879B2 (en) Read and write access to data replicas stored in multiple data centers
US20180026867A1 (en) Monitoring of replicated data instances
JP2019101703A (ja) 記憶システム及び制御ソフトウェア配置方法
CN104750757B (zh) 一种基于HBase的数据存储方法和设备
CN111090495B (zh) 节点管理方法、装置、设备、存储介质和系统
CN111158949A (zh) 容灾架构的配置方法、切换方法及装置、设备和存储介质
WO2018054079A1 (zh) 一种存储文件的方法、第一虚拟机及名称节点
CN115113800A (zh) 多集群管理方法、装置、计算设备及存储介质
JP2019191951A (ja) 情報処理システム及びボリューム割当て方法
WO2023029485A1 (zh) 数据处理方法、装置、计算机设备及计算机可读存储介质
CN109582459A (zh) 应用的托管进程进行迁移的方法及装置
CN112363820A (zh) 基于异构硬件的统一资源池化容器调度引擎及其调度方法
CN117714476A (zh) 云盘管控方法、系统、电子设备及存储介质
US8031637B2 (en) Ineligible group member status
CN111752892A (zh) 分布式文件系统及其实现方法、管理系统、设备及介质
CN103500126B (zh) 一种云计算平台的自动化容错配置方法
WO2021212677A1 (zh) 基于区块链开发实现的服务管理方法、装置及设备
WO2026040817A1 (zh) 一种支持多租户的多集群应用故障迁移方法及系统
CN107943615B (zh) 基于分布式集群的数据处理方法与系统
CN103067418A (zh) 一种元数据重建方法、设备及系统

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: 24902209

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE