Method and device for storage resource allocation for video cloud storage

• US 10,637,916 B2
• Filed: 11/08/2016
• Issued: 04/28/2020
• Est. Priority Date: 02/19/2016
• Status: Active Grant

• Current Assignee: Hangzhou Hikvision Digital Technology Company Limited
• Original Assignee: Hangzhou Hikvision Digital Technology Company Limited
• Inventors: Ding, Guangkai
• Primary Examiner(s): Nano, Sargon N
• Application Number:
  US15/999,872
• Publication Number:
  US 20190166192A1
• Time in Patent Office: 1,267 Days
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 65/40 Support for services or applications
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1097 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]
• CPC Class Codes: H04N 21/231 Content storage operation, e.g. caching movies for short term storage, replicating data over plural servers, prioritizing data for deletion
• CPC Class Codes: H04N 21/23103 using load balancing strategies, e.g. by placing or distributing content on different disks, different memories or different servers storage management G06F3/0604; allocation of resources considering the load in multiprogramming arrangements G06F9/505; techniques for rebalancing the load in a distributed system G06F9/5083; access to distributed or replicated servers, e.g. load balancing, in data networks H04L67/1001

Abstract

Embodiments of the present disclosure provide a method and apparatus for allocating storage resource of video cloud storage, applicable to a management node in a storage cluster. The method comprising: receiving a video data storage request from a client; obtaining a queue of allocable storage nodes according to an encoder identifier contained in the request; determining if the request is sent for the first time according to the encoder identifier, if so, selecting from the queue of allocable storage nodes a target storage node allocable as a write node according to a stress estimation algorithm; or otherwise, selecting from the queue a target storage node allocable as a write node according to the number of current writes of each of the storage nodes in pre-recorded node synchronization information and the stress estimation algorithm; and feeding back address information of the target storage node to the client. Embodiments of the present disclosure allow storage loads of a cloud storage system to be balanced.

Claims

1. A method for allocating storage resource of video cloud storage, applicable to a management node in a storage cluster, the storage cluster comprising at least one management node and at least one storage node, the method comprising:

receiving a video data storage request sent by a client based on an encoder, the video data storage request containing an identifier of a resource pool into which video data is to be written and an identifier of the encoder;

obtaining, from a plurality of storage nodes corresponding to the identifier of the resource pool, a queue of allocable storage nodes for the video data storage request, wherein, the correspondence between an identifier of a resource pool and a storage node is predefined;

determining whether the identifier_of the encoder is recorded in an encoder allocation tracing list;

if so, selecting from the queue of allocable storage nodes a target storage node allocable as a write node, according to the number of current writes of each of the storage nodes in pre-recorded node synchronization information and a stress estimation algorithm;

or otherwise, selecting, from the queue of allocable storage nodes, a target storage node allocable as the write node, according to the stress estimation algorithm;

wherein the encoder allocation tracing list records the correspondence between an identifier of an encoder that has been allocated a storage node and the allocated storage node, and wherein the stress estimation algorithm is an algorithm for estimating the stress on a storage node according to the node synchronization information; and

if a target storage node allocable as a write node is selected from the queue of allocable storage nodes, feeding back address information of the target storage node to the client based on the encoder to allow the client to write the video data corresponding to the video data storage request into the target storage node.

2. The method according to claim 1, wherein the pre-recorded node synchronization information comprises:

CPU utilization, bandwidth occupied by reading of video data, a number of current reads, bandwidth occupied by writing of video data, disk I/O utilization and memory utilization of a storage node;

the stress estimation algorithm comprises;

estimating writing stress of a storage node according to bandwidth occupied by writing of video data and the number of current writes of the storage node;

calculating the reading stress of a storage node, according to the bandwidth occupied by reading of video data and the number of current reads of the storage node; and

estimating stress of the storage node according to the CPU utilization, the memory utilization, the disk TO utilization, the estimated writing stress and the reading stress of the storage node;

or wherein,

the method further comprises;

if a target storage node allocable as a write node cannot be selected from the queue of allocable storage nodes, selecting randomly a storage node to be the target storage node from the queue of allocable storage nodes, and providing the address information of the target storage node to the client based on the encoder to allow the client to write the video data corresponding to the video data storage request into the target storage node;

or wherein,

the video data storage request further comprises the network address of the client;

obtaining, from a plurality of storage nodes corresponding to the identifier of resource pool, a queue of allocable storage nodes for the video data storage request comprises;

determining, from the plurality of storage nodes corresponding to the identifier of resource pool, storage nodes corresponding to a network address of the client;

wherein the correspondence between ranges of network addresses and storage nodes is predefined; and

obtaining, from the determined storage nodes corresponding to the network address of the client, a queue of allocable storage nodes for the video data storage request;

or wherein,

the method further comprises;

increasing the number of current writes of the target storage node by one, to update the node synchronization information of the target storage node;

or wherein,

selecting from the queue of allocable storage nodes a target storage node allocable as a write node, according to the number of current writes of each of the storage nodes in pre-recorded node synchronization information and a stress estimation algorithm write node comprises;

determining, from the queue of allocable storage nodes, a storage node allocated for the encoder last time, according to the encoder allocation tracing list;

taking a storage node next to the storage node allocated last time in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations as follows;

obtaining the number of current writes of the current storage node, from the pre-recorded node synchronization information for storage nodes;

calculating a recommended number of writes of the current storage node;

determining whether the number of current writes is smaller than the recommended number of writes, and if the number of current writes is smaller than the recommended number of writes, estimating the stress of the storage node according to the stress estimation algorithm;

if the estimated stress is smaller than a predetermined threshold, selecting the current storage node as the target storage node allocable as a write node, and terminating the selection process, or if the estimated stress is not smaller than the predetermined threshold, proceeding back to the step of taking a storage node next to the current storage node in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations until all storage nodes in the queue of allocable storage nodes have been taken as a to-be-determined current storage node;

if the number of current writes is larger than or equal to the recommended number of writes, proceeding back to the step of taking a storage node next to the current storage node in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations until all storage nodes in the queue of allocable storage nodes have been taken as a to-be-determined current storage node.

3. The method according to claim 2, wherein the writing stress of the storage node is estimated as follows:

the estimated writing stress of the storage node=the sum of the numbers of current writes of all the storage nodes in the storage cluster*(the bandwidth occupied by writing of video data of the storage node/the number of current writes of the storage node);

the reading stress of the storage node is calculated as follows;

the reading stress of the storage node=the sum of the numbers of current reads of all the storage nodes in the storage cluster*(the bandwidth occupied by reading of video data of the storage node/the number of current reads of the storage node).

4. The method according to claim 2, wherein the recommended number of writes of the current storage node is calculated as follows:

the recommended number of writes of the current storage node=the total of writes of all the storage nodes in the storage cluster*(the total storage space of the current storage node/the total storage space of all the storage nodes).

5. The method according to claim 2, further comprising:

decreasing the number of current writes of the determined storage node allocated to the encoder last time by one, in the recorded node synchronization information of the storage nodes.

6. The method according to claim 1, wherein selecting, according to a stress estimation algorithm, a target storage node allocable as a write node from the queue of allocable storage nodes comprises:

estimating the stress of each storage node in the queue of allocable storage nodes according to the stress estimation algorithm based on the node synchronization information; and

selecting randomly from the storage nodes a storage node having an estimated stress lower than a predetermined threshold, as the target storage node allocable as the write node.

7. The method according to claim 1, wherein, obtaining, from a plurality of storage nodes corresponding to the identifier of resource pool, a queue of allocable storage nodes for the video data storage request comprises:

determining at least one storage node satisfying an allocation condition from the plurality of storage nodes corresponding to the identifier of the resource pool, according to pre-recorded node synchronization information of the storage nodes; and

establishing the queue of allocable storage nodes for the video data storage request based on the at least one storage node satisfying the allocation condition.

8. The method according to claim 7, wherein, the node synchronization information further comprises:

online status, operation state, and remaining space of a storage node;

a storage node satisfies the allocation condition if the synchronization information indicates that the storage node is online, in a normal operation state, and having remaining storage space;

or,

the node synchronization information comprises;

online status, operation state, remaining space, formatting state, the number of current writes, and a maximum number of current writes of a storage node;

a storage node satisfies the allocation condition if the synchronization information indicates that the storage node is online, in a normal operation state, having remaining storage space, and not being formatted, and the number of current writes is smaller than the maximum number of writes.

9. The method according to claim 1, wherein the video data storage request is sent by the client based on at least one encoder at a preset time interval.

10. An electronic device, comprising:

a housing, a processor, a memory, a circuit board and a power supply circuit, wherein, the circuit board is placed within the space enclosed by the housing;

the processor and memory are disposed on the circuit board;

the power supply circuit is configured for supplying power to circuits and devices of the electronic device;

the memory is configured for storing executable program instructions; and

the processor is configured for executing the program instructions stored in the memory to perform the method for allocating storage resource of video cloud storage according to claim 1.

11. A non-transitory storage medium, configured for storing executable program codes, which, when executed, perform the method for allocating storage resource of video cloud storage according to claim 1.

12. An apparatus for allocating storage resource of video cloud storage, applicable to a management node in a storage cluster, the storage cluster comprising at least one management node and at least one storage node, the apparatus comprising:

a video data storage request receiving module configured for receiving a video data storage request sent by a client based on an encoder, the video data storage request containing the identifier of a resource pool into which a video data is to be written and an identifier of the encoder;

an allocable storage node queue obtaining module configured for obtaining, from a plurality of storage nodes corresponding to the identifier of the resource pool, a queue of allocable storage nodes for the video data storage request, wherein, the correspondence between an identifier of a resource pool and a storage node is predefined;

a determining module configured for determining whether the identifier of the encoder is recorded in an encoder allocation tracing list;

if so, activating a first target storage node determining module, or otherwise, activating a second target storage node determining module, wherein the encoder allocation tracing list records the correspondence between an identifier of an encoder that has been allocated a storage node and the allocated storage node;

the first target storage node determining module is configured for selecting from the queue of allocable storage nodes a target storage node allocable as a write node, according to the number of current writes of each of the storage nodes in pre-recorded node synchronization information and a stress estimation algorithm, wherein the stress estimation algorithm is an algorithm for estimating the stress on a storage node according to the node synchronization information;

the second target storage node determining module is configured for selecting, from the queue of allocable storage node, a target storage node allocable as the write node, according to the stress estimation algorithm; and

a first target storage node address information returning module configured for, if a target storage node allocable as a write node is selected from the queue of allocable storage nodes, feeding back the address information of the target storage node to the client based on the encoder to allow the client to write the video data corresponding to the video data storage request into the target storage node.

13. The apparatus according to claim 12, wherein the pre-recorded node synchronization information comprises:

CPU utilization, bandwidth occupied by reading of video data, a number of current reads, bandwidth occupied by writing of video data, disk I/O utilization and memory utilization of a storage node;

the stress estimation algorithm utilized by the first target storage node determining module and the second target storage node determining module comprises;

estimating writing stress of a storage node according to bandwidth occupied by writing of video data and the number of current writes of the storage node;

calculating the reading stress of a storage node, according to the bandwidth occupied by reading of video data and the number of current reads of the storage node; and

estimating stress of the storage node according to the CPU utilization, the memory utilization, the disk IO utilization, the estimated writing stress and the reading stress of the storage node;

or wherein,

the apparatus further comprises;

a second target storage node address information returning module configured for, if a target storage node allocable as a write node cannot be selected from the queue of allocable storage nodes, selecting randomly a storage node to be the target storage node from the queue of allocable storage nodes, and providing the address information of the target storage node to the client based on the encoder to allow the client to write the video data corresponding to the video data storage request into the target storage node;

or wherein,

the video data storage request further comprises the network address of the client;

the allocable storage node queue obtaining module comprises a first storage node determining submodule and a first allocable storage node queue obtaining submodule;

the first storage node determining submodule is configured for determining, from the plurality of storage nodes corresponding to the identifier_of resource pool, storage nodes corresponding to the network address of the client;

wherein the correspondence between ranges of network addresses and storage nodes is predefined;

the first allocable storage node queue obtaining submodule is configured for obtaining, from the determined storage nodes corresponding to a network address of the client, a queue of allocable storage nodes for the video data storage request;

or wherein,

the apparatus further comprises;

an increasing module configured for increasing the number of current writes of the target storage node by one, to update the node synchronization information of the target storage node;

or wherein,

the first target storage node determining module comprises;

a second storage node determining submodule configured for determining, from the queue of allocable storage nodes, a storage node allocated for the encoder last time, according to the encoder allocation tracing list; and

a storage node processing submodule configured for taking a storage node next to the allocated storage node allocated last time in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations as follows;

obtaining the number of current writes of the current storage node, from the pre-recorded node synchronization information for storage nodes;

calculating a recommended number of writes of the current storage node;

determining whether the number of current writes is smaller than the recommended number of writes, and if the number of current writes is smaller than the recommended number of writes, estimating the stress of the storage node according to the stress estimation algorithm;

if the estimated stress is smaller than a predetermined threshold, selecting the current storage node as the target storage node allocable as a write node, and terminating the selection process, or if the estimated stress is not smaller than the predetermined threshold, proceeding back to the step of taking a storage node next to the current storage node in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations until all storage nodes in the queue of allocable storage nodes have been taken as a to-be-determined current storage node; and

if the number of current writes is larger than or equal to the recommended number of writes, proceeding back to the step of taking a storage node next to the current storage node in the queue of allocable storage nodes as a to-be-determined current storage node and performing the selection operations until all storage nodes in the queue of allocable storage nodes have been taken as to-be-determined current storage node.

14. The apparatus according to claim 13, wherein

the writing stress of the storage node is estimated as follows:

the estimated writing stress of the storage node=the sum of the numbers of current writes of all the storage nodes in the storage cluster*(the bandwidth occupied by writing of video data of the storage node/the number of current writes of the storage node);

the reading stress of the storage node is calculated as follows;

the reading stress of the storage node=the sum of the numbers of current reads of all the storage nodes in the storage cluster*(the bandwidth occupied by reading of video data of the storage node/the number of current reads of the storage node).

15. The apparatus according to claim 13, wherein the storage node processing submodule calculates the recommended number of writes of the current storage node as follows:

the recommended number of writes of the current storage node=the total of writes of all the storage nodes in the storage cluster*(the total storage space of the current storage node/the total storage space of all the storage nodes).

16. The apparatus according to claim 13, further comprising:

a decreasing module configured for decreasing the number of current writes of the determined storage node allocated to the encoder last time by one, in the recorded node synchronization information of the storage nodes.

17. The apparatus according to claim 12, wherein the second target storage node determining module comprises a stress estimation submodule and a target storage node determining submodule;

the stress estimation submodule is configured for estimating the stress of each storage node in the queue of allocable storage nodes according to the stress estimation algorithm based on the node synchronization information;

the target storage node determining submodule is configured for selecting randomly from the storage nodes a storage node having an estimated stress lower than a predetermined threshold, as the target storage node allocable as the write node.

18. The apparatus according to claim 12, wherein

the allocable storage node queue obtaining module comprises a third storage node determining submodule and a second allocable storage node queue determining submodule;

the third storage node determining submodule is configured for determining at least one storage node satisfying an allocation condition from the plurality of storage nodes corresponding to the identifier of the resource pool, according to pre-recorded node synchronization information of the storage nodes;

the second allocable storage node queue determining submodule is configured for establishing the queue of allocable storage nodes for the video data storage request based on the at least one storage node satisfying the allocation condition.

19. The apparatus according to claim 18, wherein, the node synchronization information further comprises:

online status, operation state, and remaining space of a storage node;

a storage node satisfies the allocation condition if the synchronization information indicates that the storage node is online, in a normal operation state, and having remaining storage space;

or,

the node synchronization information comprises;

online status, operation state, remaining space, formatting state, the number of current writes, and a maximum number of current writes of a storage node;

a storage node satisfies the allocation condition if the synchronization information indicates that the storage node is online, in a normal operation state, having remaining storage space, and not being formatted, and the number of current writes is smaller than the maximum number of writes.

20. The apparatus according to claim 12, wherein the video data storage request is sent by the client based on at least one encoder at a preset time interval.