CN107087031A - A kind of storage resource load-balancing method and device - Google Patents
A kind of storage resource load-balancing method and device Download PDFInfo
- Publication number
- CN107087031A CN107087031A CN201710323984.0A CN201710323984A CN107087031A CN 107087031 A CN107087031 A CN 107087031A CN 201710323984 A CN201710323984 A CN 201710323984A CN 107087031 A CN107087031 A CN 107087031A
- Authority
- CN
- China
- Prior art keywords
- storage
- task
- physical memory
- memory resources
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- 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; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Abstract
The invention discloses a kind of storage resource load-balancing method and device, applied to the storage managing server in data-storage system, storage managing server obtains the facility information carried in application apparatus writing task, the corresponding task cryptographic Hash of the writing task is calculated according to the facility information, the task cryptographic Hash and the node cryptographic Hash of each dummy node in Storage Virtualization component are compared, according to comparative result, the physical memory resources that selection meets corresponding to the dummy node of setting rule are stored as the physical memory resources of the writing task.The corresponding device of the present invention includes task receiving module and task allocating module.The method and device of the present invention, by the stability and harmony of proof load, improves the performance of the business such as read-write retrieval.
Description
Technical field
The invention belongs to technical field of data storage, more particularly to a kind of storage resource load-balancing method and device.
Background technology
With the development of monitoring business, monitoring scale is increasing, and the pressure-bearing to storage resource is a kind of test.In order to carry
The readwrite performance of high monitoring business is, it is necessary to which the load to storage resource carries out balanced, the pressure of the single resource of reduction.For example as schemed
Cloud storage system shown in 1, the multiple SAN storage resources of carry are, it is necessary to uniformly divide the business on each CDV typically on CDV
Spread out in each storage resource to provide miscellaneous service performance.
Prior art, often does load balancing according to the read-write number of tasks and capacity in single storage resource,
When capacity is enough, choose the minimum storage resource of number of tasks to use.The program is easy to use, and generally duty factor is more equal
Weighing apparatus, safeguards that the number of tasks in a task count, each storage resource can be opened according to read-write task in each storage resource
Closing upgrades in time.But the program has the shortcomings that obvious with the increase of storage service, the complexity of network environment：
In instantaneous intensive business, it is impossible to number of tasks equilibrium is completely secured, load balancing is reached；
The complexity of network environment, such as when closing file failure, may cause number of tasks to remain, at this time according to number of tasks
It is minimum to select resource, it is actual and unbalanced；
Selected the distribution of resource, i.e. file with very big randomness according to number of tasks, catalogue number in resource can be increased, dropped
The performance of low retrieval reading business.
The content of the invention
It is an object of the invention to provide a kind of storage resource load-balancing method and device, it is to avoid is born according to number of tasks
The disadvantages mentioned above having when carrying balanced.
To achieve these goals, technical solution of the present invention is as follows：
A kind of storage resource load-balancing method, applied to the storage managing server in data-storage system, the number
Also include application apparatus and Storage Virtualization component according to storage system, the Storage Virtualization component is deposited including at least two physics
Resource is stored up, each physical memory resources fictionalize at least two dummy nodes, and each dummy node is to that there should be one by storage tube
Manage the virtual node information of server admin, and the node cryptographic Hash calculated according to the virtual node information, the storage
Method for balancing resource load includes：
Storage managing server obtains the facility information carried in application apparatus writing task, is calculated according to the facility information
The corresponding task cryptographic Hash of the writing task；
The task cryptographic Hash and the node cryptographic Hash of each dummy node in Storage Virtualization component are compared, root
According to comparative result, selection meets the physical memory resources corresponding to the dummy node of setting rule as the physics of the writing task
Storage resource is stored.
Preferably, the facility information is the corresponding User Catalog coding of the writing task.
Preferably, the physical memory resources that the selection is met corresponding to the dummy node of setting rule are appointed as described write
The physical memory resources of business are stored, including：
From the dummy node sorted according to node cryptographic Hash, first node cryptographic Hash of selection is than task cryptographic Hash bigger
Dummy node；
Stored the physical memory resources where the dummy node as the physical memory resources of the writing task.
Further, the storage resource load-balancing method also includes：
If selected physical memory resources are unavailable or off-capacity, calculate that all available and capacity is enough
Physical memory resources on writing task number, the minimum physical memory resources of selection writing task number deposit as the physics of the writing task
Storage resource is stored.
Preferably, the dummy node sum that the physical memory resources under the single Storage Virtualization component are fictionalized is not small
In the maximum storage way of single Storage Virtualization component.
The invention also provides a kind of storage resource load balancing apparatus, applied to the storage management in data-storage system
Server, the data-storage system also includes application apparatus and Storage Virtualization component, and the Storage Virtualization component includes
At least two physical memory resources, each physical memory resources fictionalize at least two dummy nodes, each dummy node correspondence
There is a virtual node information managed by storage managing server, and the node calculated according to the virtual node information is breathed out
Uncommon value, the storage resource load balancing apparatus includes：
Task receiving module, for obtaining the facility information carried in application apparatus writing task, according to the facility information
Calculate the corresponding task cryptographic Hash of the writing task；
Task allocating module, for by the node of each dummy node in the task cryptographic Hash and Storage Virtualization component
Cryptographic Hash is compared, according to comparative result, and the physical memory resources that selection meets corresponding to the dummy node of setting rule are made
Physical memory resources for the writing task are stored.
Preferably, the facility information is the corresponding User Catalog coding of the writing task.
Preferably, the task allocating module selection meets the physical memory resources corresponding to the dummy node of setting rule
Stored as the physical memory resources of the writing task, perform following operation：
From the dummy node sorted according to node cryptographic Hash, first node cryptographic Hash of selection is than task cryptographic Hash bigger
Dummy node；
Stored the physical memory resources where the dummy node as the physical memory resources of the writing task.
Further, the task allocating module is additionally operable to：
If selected physical memory resources are unavailable or off-capacity, calculate that all available and capacity is enough
Physical memory resources on writing task number, the minimum physical memory resources of selection writing task number deposit as the physics of the writing task
Storage resource is stored.
Preferably, the dummy node sum that the physical memory resources under the single Storage Virtualization component are fictionalized is not small
In the maximum storage way of single Storage Virtualization component.
A kind of storage resource load-balancing method and device proposed by the present invention, storage managing server is according to writing task pair
The task cryptographic Hash answered, task cryptographic Hash and the node cryptographic Hash of the dummy node in Storage Virtualization component are compared,
According to comparative result, the physical memory resources where selection meets the dummy node of setting rule from dummy node are write as this
The physical memory resources of task are stored.Because the corresponding task cryptographic Hash of writing task is fixed, it is ensured that load balancing
Stability, single application apparatus is tried one's best and be stored on some physical memory resources.By increasing dummy node, deposit physics
Storage resource, which is tried one's best, to scatter, and increases loaded harmony, multiple application apparatus is tried one's best and disperses to be stored in multiple physical stores moneys
On source, in abnormal cases, load balancing adjustment is carried out by calculating the writing task number on physical memory resources.And by ensureing
The stability and harmony of load, improve the performance of the business such as read-write retrieval.
Brief description of the drawings
Fig. 1 is cloud storage system structural representation conventional in video monitoring；
Fig. 2 is the structural representation of Storage Virtualization component；
Fig. 3 is storage resource load-balancing method flow chart of the present invention；
Fig. 4 is experimental data distribution map；
Fig. 5 is storage resource load balancing apparatus structural representation of the present invention.
Embodiment
Technical solution of the present invention is described in further details with reference to the accompanying drawings and examples, following examples are not constituted
Limitation of the invention.
As shown in figure 1, be cloud storage CDS (the Cloud Direct Storage) scheme commonly used in a kind of video monitoring,
In the cloud storage system, application apparatus includes web camera IPC, encoder EC or other data-source devices, what IPC was shot
Video data needs to store into storage resource.Cloud storage management server CDM (Cloud Direct Management) is concentrated
The Storage Virtualization component CDV (Cloud Direct Virtual) of bottom is managed collectively, global unified virtual memory is constituted
Space.The multiple IP SAN/FC SAN/NAS storage devices of carry typically on a CDV, the present embodiment is by an IP SAN/FC
SAN/NAS storage devices are referred to as the multiple physical memory resources of carry on a physical memory resources, i.e., one CDV.
In cloud storage system as shown in Figure 1, when application apparatus will be written and read, it can be visited to CDM requests first
The memory space asked, and will be gone in the discrete storage of video data to the physical memory resources of each bottom standard, when a certain physics is deposited
When storage resource breaks down, CDM can redistribute a new memory space automatically and be conducted interviews to application apparatus, so as to realize
The read-write that business does not stop, supports that there is provided continual video and picture storage service for dynamic capacity-expanding and capacity reducing.It can also carry simultaneously
For live video, video request program playback, alarming and managing, deploy troops on garrison duty withdraw a garrison, the various video business such as map application.
The present embodiment is by taking the cloud storage commonly used in the video monitoring shown in Fig. 1 as an example, to elaborate the storage of the present invention
Method for balancing resource load, applied to the storage managing server in data-storage system.Data storage system in the technical program
System also includes application apparatus and Storage Virtualization component, and Storage Virtualization component includes at least two physical memory resources, each
Physical memory resources fictionalize at least two dummy nodes, and each dummy node is managed by storage managing server that should have one
Virtual node information, and the node cryptographic Hash calculated according to the virtual node information.
For example, as shown in Fig. 2 physical memory resources have three Res1, Res2 and Res3, Res1 is to that should have dummy node
Res1-1, Res1-2 and Res1-3, Res2 is to that should have dummy node Res2-1, Res2-2 and Res2-3, and Res3 is to that should have virtually
Node R es3-1, Res3-2 and Res3-3.Assuming that Res1-1, Res1-2, Res1-3, Res2-1, Res2-2, Res2-3, Res3-
1st, Res3-2, Res3-3 are the corresponding node cryptographic Hash of each dummy node, are arranged in fig. 2 according to the size order of node cryptographic Hash
Row.
In actual monitored system, it is assumed that single CDV maximum storages way is 512 tunnels, the present embodiment deposits each physics
Storage resource fictionalizes 32 dummy nodes, when the sum of all dummy nodes is not more than 512, then increases each physical store money
The dummy node number in source, for example, have 5 physical memory resources, and each physical memory resources fictionalize 32 dummy nodes, virtually
The sum ability 160 of node, at this time needs to increase the dummy node number that each physical memory resources are fictionalized, for example, increase to every
The dummy node number of individual physical memory resources is 105 so that the sum of dummy node is more than 512.It is such to set so that single
The dummy node sum that physical memory resources under individual Storage Virtualization component are fictionalized is not less than single Storage Virtualization component
Maximum storage way, for per all the way storage always there is dummy node to select, be easy to follow-up load balancing.
3 physical memory resources are illustrate only in Fig. 2, dummy node, which is placed on a dotted line, represents that its node cryptographic Hash is suitable
Sequence, is arranged from small to large, is so more prone to selection when subsequently carrying out dummy node selection.It can configure as needed more
Many physical memory resources, are repeated no more here.Because single CDV maximum storages way is limited, single storage in the present embodiment
The dummy node sum that physical memory resources under virtualisation component are fictionalized is not less than the maximum of single Storage Virtualization component
Way is stored, for always having dummy node to select per storage all the way, is easy to follow-up load balancing.
In the present embodiment, storage managing server can manage multiple Storage Virtualization components, each Storage Virtualization group
Part includes at least two physical memory resources, and each physical memory resources fictionalize at least two dummy nodes, dummy node
Sum is not less than the storage way of Storage Virtualization component, and each dummy node is managed by storage managing server that should have one
Virtual node information, the virtual node information is typically the title of the dummy node, or virtual route etc. other have representative
Property and unique information, wherein virtual route refer to what dummy node was carried, for pointing to the road of corresponding physical memory resources
Footpath, storage managing server can find corresponding physical memory resources according to the virtual route of dummy node.Such as basis
The virtual route of dummy node, a corresponding node cryptographic Hash, each dummy node are calculated by Hash (HASH) function
Record has the virtual route and node cryptographic Hash of the dummy node of oneself.
As shown in figure 3, a kind of storage resource load-balancing method of the present embodiment, including：
Storage managing server obtains the facility information carried in application apparatus writing task, is calculated to write according to facility information and appointed
It is engaged in corresponding task cryptographic Hash；
Task cryptographic Hash and the node cryptographic Hash of each dummy node in Storage Virtualization component are compared；
According to comparative result, physical memory resources corresponding to the dummy node of selection satisfaction setting rule are used as the writing task
Physical memory resources stored.
By taking the cloud storage CDS commonly used in the video monitoring in Fig. 1 as an example, the video data of IPC outputs needs to write physics
Storage resource, writing task is sent to CDM, and CDM calculates its corresponding task cryptographic Hash by the hash function of setting.For example adopt
Hash function is MD5, SHA etc., and the present embodiment selection MD5 algorithms, MD5 has very low collision probability and preferable hash
Uniformity.Facility information corresponding to writing task is calculated task cryptographic Hash, facility information can be IPC title,
The information such as sequence number, or User Catalog coding etc. have uniqueness information.Each task for example in video monitoring system
To that should have respective User Catalog to encode, and User Catalog coding is unique, and task is calculated using User Catalog coding
Cryptographic Hash.
Such as one typical User Catalog coding is as follows：
256GB001#1885365065_47640
Because each User Catalog coding is unique in video monitoring system, difference can be obtained by hash functions
Cryptographic Hash, in the business of general scale (each tunnels of CDV maximum storages Shuo Wei 512, physical memory resources maximum number be 128)
In, it ensure that writing task is distributed on each physical memory resources.And User Catalog is to immobilize, stable, pass through
Every time, ensure that in the case of physical memory resources are available can select the task cryptographic Hash that hash functions are obtained every time
The physical memory resources are selected, when physical memory resources recover normal from abnormality, the resource can be chosen again.So
The stability of resource selection is ensured that, i.e., same writing task all writes same physical memory resources.So as to after also allowing for
Continuous progress other business such as plays back and more facilitated when reading, improves the service feature in system.
Assuming that the task cryptographic Hash calculated is hash (UsrDirCode), by task cryptographic Hash and Storage Virtualization component
In the node cryptographic Hash of dummy node be compared, the dummy node of writing task is carried out according to setting rule selection, for example from
In the dummy node sorted according to node cryptographic Hash, virtual bigger than hash (UsrDirCode) of first node cryptographic Hash of selection
Node, dummy node sorts according to node cryptographic Hash size, easy-to-look-up.Assuming that this node is res2-2, corresponding physics is deposited
Storage resource is res2, then write operation is carried out on res2.Rule is set during selection to be adjusted, and be not limited to selection first
First node cryptographic Hash of the individual node cryptographic Hash dummy node bigger than hash (UsrDirCode) or selection compares hash
(UsrDirCode) small dummy node.
Further, if the physical memory resources are unavailable or off-capacity, all available and capacity are calculated
Writing task number on enough physical memory resources, selects the minimum physical memory resources of writing task number as the thing of the writing task
Reason storage resource is stored, it is ensured that be adapted dynamically harmony under abnormal conditions.In addition, if the physical memory resources can not
With or off-capacity, the writing task number on all available and enough physical memory resources of capacity, selection can also be calculated
Writing task number is less than the dummy node on the physical memory resources of given threshold.The capacity of general physical memory resources can be certainly
It is dynamic to detect and be reported to storage managing server, when the capacity of physical memory resources is less than the threshold value of setting, it is believed that the physics
The off-capacity of storage resource, or the storage space volume required for the capacity of physical memory resources is less than the writing task
When, it is believed that the off-capacity of the physical memory resources, otherwise it is considered as capacity enough.The principle that specific selection basis is divided equally,
Select, repeat no more here on the available and enough physical memory resources of capacity.
It should be noted that the technical scheme of the present embodiment is same for the load balancing of storage resource in other storage systems
Sample is applicable, such as in TV programme storage system, and application apparatus is the recording front end in each radio station, and its storage managing server is not
It is the CDM in the present embodiment, but the storage managing server being managed to storage resource.I.e. the technical program is suitable for institute
There is the data-storage system for carrying out data storage management.
As shown in figure 4, in one embodiment, 92 tunnels of configuration storage, physical memory resources number is 20, obtained storage
Distribution results, the longitudinal axis represents the storage writing task number on physical memory resources, and transverse axis represents physical memory resources.Store result symbol
The storage writing task number number deviation (maximum storage number-minimum memory number) in expected all resources is closed within 5, overall balanced point
Cloth.
As shown in figure 5, a kind of storage resource load balancing apparatus of the present embodiment, applied to the storage in data-storage system
Management server.The present embodiment data-storage system also includes application apparatus and Storage Virtualization component, the Storage Virtualization
Component includes at least two physical memory resources, and each physical memory resources fictionalize at least two dummy nodes, each virtual
Node is calculated to that should have a virtual node information managed by storage managing server, and according to the virtual node information
Node cryptographic Hash.The present embodiment storage resource load balancing apparatus includes：
Task receiving module, for obtaining the facility information carried in application apparatus writing task, according to the facility information
Calculate the corresponding task cryptographic Hash of the writing task；
Task allocating module, for by the node of each dummy node in the task cryptographic Hash and Storage Virtualization component
Cryptographic Hash is compared, according to comparative result, and the physical memory resources that selection meets corresponding to the dummy node of setting rule are made
Physical memory resources for the writing task are stored.
Storage resource load-balancing method with above-mentioned Fig. 3 is corresponding, the storage resource load balancing apparatus of the present embodiment：
Preferably, the facility information is the corresponding User Catalog coding of the writing task.
Preferably, task allocating module selection meets the physical memory resources conduct corresponding to the dummy node of setting rule
The physical memory resources of the writing task are stored, and perform following operation：
From the dummy node sorted according to node cryptographic Hash, first node cryptographic Hash of selection is than task cryptographic Hash bigger
Dummy node；
Stored the physical memory resources where the dummy node as the physical memory resources of the writing task.
Preferably, selected physical memory resources are unavailable or during off-capacity, task allocating module is additionally operable to：
If selected physical memory resources are unavailable or off-capacity, calculate that all available and capacity is enough
Physical memory resources on writing task number, the physical store money of the minimum physical memory resources writing task of selection writing task number
Source is stored.
Equally, the dummy node sum that the physical memory resources under the single Storage Virtualization component of the present embodiment are fictionalized is not
Less than the maximum storage way of single Storage Virtualization component, always there is dummy node to be selected hence for per storage all the way
Select, be easy to choose dummy node according to the task cryptographic Hash of writing task, realize load balancing.
The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, without departing substantially from essence of the invention
In the case of refreshing and its essence, those skilled in the art work as can make various corresponding changes and change according to the present invention
Shape, but these corresponding changes and deformation should all belong to the protection domain of appended claims of the invention.
Claims (10)
1. a kind of storage resource load-balancing method, applied to the storage managing server in data-storage system, the data
Storage system also includes application apparatus and Storage Virtualization component, it is characterised in that the Storage Virtualization component is included at least
Two physical memory resources, each physical memory resources fictionalize at least two dummy nodes, and each dummy node is to that should have one
The individual virtual node information managed by storage managing server, and the node Hash calculated according to the virtual node information
Value, the storage resource load-balancing method includes：
Storage managing server obtains the facility information carried in application apparatus writing task, according to being calculated the facility information
The corresponding task cryptographic Hash of writing task；
The task cryptographic Hash and the node cryptographic Hash of each dummy node in Storage Virtualization component are compared, according to than
Relatively result, selection meets the physical memory resources corresponding to the dummy node of setting rule as the physical store of the writing task
Resource is stored.
2. storage resource load-balancing method according to claim 1, it is characterised in that the facility information is write to be described
The corresponding User Catalog coding of task.
3. storage resource load-balancing method according to claim 1, it is characterised in that the selection meets setting rule
Dummy node corresponding to physical memory resources stored as the physical memory resources of the writing task, including：
From the dummy node sorted according to node cryptographic Hash, virtual bigger than task cryptographic Hash of first node cryptographic Hash of selection
Node；
Stored the physical memory resources where the dummy node as the physical memory resources of the writing task.
4. storage resource load-balancing method according to claim 1, it is characterised in that the storage resource load balancing
Method also includes：
If selected physical memory resources are unavailable or off-capacity, all available and enough things of capacity are calculated
The writing task number in storage resource is managed, the minimum physical memory resources of selection writing task number are provided as the physical store of the writing task
Source is stored.
5. storage resource load-balancing method according to claim 1, it is characterised in that the single Storage Virtualization group
The dummy node sum that physical memory resources under part are fictionalized is not less than the maximum storage way of single Storage Virtualization component.
6. a kind of storage resource load balancing apparatus, applied to the storage managing server in data-storage system, the data
Storage system also includes application apparatus and Storage Virtualization component, it is characterised in that the Storage Virtualization component is included at least
Two physical memory resources, each physical memory resources fictionalize at least two dummy nodes, and each dummy node is to that should have one
The individual virtual node information managed by storage managing server, and the node Hash calculated according to the virtual node information
Value, the storage resource load balancing apparatus includes：
Task receiving module, for obtaining the facility information carried in application apparatus writing task, is calculated according to the facility information
The corresponding task cryptographic Hash of the writing task；
Task allocating module, for by the node Hash of each dummy node in the task cryptographic Hash and Storage Virtualization component
Value is compared, and according to comparative result, the physical memory resources that selection meets corresponding to the dummy node of setting rule are used as institute
The physical memory resources for stating writing task are stored.
7. storage resource load balancing apparatus according to claim 6, it is characterised in that the facility information is write to be described
The corresponding User Catalog coding of task.
8. storage resource load balancing apparatus according to claim 6, it is characterised in that the task allocating module selection
The physical memory resources corresponding to the dummy node of setting rule are met to be deposited as the physical memory resources of the writing task
Storage, performs following operation：
From the dummy node sorted according to node cryptographic Hash, virtual bigger than task cryptographic Hash of first node cryptographic Hash of selection
Node；
Stored the physical memory resources where the dummy node as the physical memory resources of the writing task.
9. storage resource load balancing apparatus according to claim 6, it is characterised in that the task allocating module is also used
In：
If selected physical memory resources are unavailable or off-capacity, all available and enough things of capacity are calculated
The writing task number in storage resource is managed, the minimum physical memory resources of selection writing task number are provided as the physical store of the writing task
Source is stored.
10. storage resource load balancing apparatus according to claim 6, it is characterised in that the single Storage Virtualization
The dummy node sum that physical memory resources under component are fictionalized is not less than the maximum storage road of single Storage Virtualization component
Number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710323984.0A CN107087031B (en) | 2017-05-10 | 2017-05-10 | Storage resource load balancing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710323984.0A CN107087031B (en) | 2017-05-10 | 2017-05-10 | Storage resource load balancing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107087031A true CN107087031A (en) | 2017-08-22 |
| CN107087031B CN107087031B (en) | 2020-07-28 |
Family
ID=59612762
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710323984.0A Active CN107087031B (en) | 2017-05-10 | 2017-05-10 | Storage resource load balancing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107087031B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109669775A (en) * | 2018-12-10 | 2019-04-23 | 平安科技（深圳）有限公司 | Distributed task dispatching method, system and storage medium |
| CN109684080A (en) * | 2018-12-10 | 2019-04-26 | 平安科技（深圳）有限公司 | Centralized method for scheduling task, system and storage medium |
| CN110086840A (en) * | 2018-01-26 | 2019-08-02 | 浙江宇视科技有限公司 | Image data recording method, device and computer readable storage medium |
| CN111031567A (en) * | 2020-01-14 | 2020-04-17 | 南通先进通信技术研究院有限公司 | Flow statistical method for core network equipment |
| CN111708606A (en) * | 2020-06-16 | 2020-09-25 | 杨海朝 | Virtual computer storage service system and use method thereof |
| CN112291292A (en) * | 2019-07-26 | 2021-01-29 | 浙江宇视科技有限公司 | Data storage method, device, equipment and medium |
| CN112468590A (en) * | 2020-11-27 | 2021-03-09 | 杭州海康威视系统技术有限公司 | Storage resource mounting method and device |
| CN117194439A (en) * | 2023-11-07 | 2023-12-08 | 杭州优云科技有限公司 | Method for creating resource storage system, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102291452A (en) * | 2011-08-09 | 2011-12-21 | 北京星网锐捷网络技术有限公司 | Virtual machine management method, cloud management server and cloud system based on cloud strategy |
| CN102938790A (en) * | 2012-11-20 | 2013-02-20 | 无锡城市云计算中心有限公司 | Resource allocation method of cloud computing system |
| CN104023058A (en) * | 2014-06-03 | 2014-09-03 | 浙江宇视科技有限公司 | Method and system for managing and allocating storage resources |
| CN105404469A (en) * | 2015-10-22 | 2016-03-16 | 浙江宇视科技有限公司 | Video data storage method and system |
| CN105930545A (en) * | 2016-06-29 | 2016-09-07 | 浙江宇视科技有限公司 | Method and device for migrating files |
-
2017
- 2017-05-10 CN CN201710323984.0A patent/CN107087031B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102291452A (en) * | 2011-08-09 | 2011-12-21 | 北京星网锐捷网络技术有限公司 | Virtual machine management method, cloud management server and cloud system based on cloud strategy |
| CN102938790A (en) * | 2012-11-20 | 2013-02-20 | 无锡城市云计算中心有限公司 | Resource allocation method of cloud computing system |
| CN104023058A (en) * | 2014-06-03 | 2014-09-03 | 浙江宇视科技有限公司 | Method and system for managing and allocating storage resources |
| CN105404469A (en) * | 2015-10-22 | 2016-03-16 | 浙江宇视科技有限公司 | Video data storage method and system |
| CN105930545A (en) * | 2016-06-29 | 2016-09-07 | 浙江宇视科技有限公司 | Method and device for migrating files |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110086840A (en) * | 2018-01-26 | 2019-08-02 | 浙江宇视科技有限公司 | Image data recording method, device and computer readable storage medium |
| CN109669775A (en) * | 2018-12-10 | 2019-04-23 | 平安科技（深圳）有限公司 | Distributed task dispatching method, system and storage medium |
| CN109684080A (en) * | 2018-12-10 | 2019-04-26 | 平安科技（深圳）有限公司 | Centralized method for scheduling task, system and storage medium |
| CN112291292A (en) * | 2019-07-26 | 2021-01-29 | 浙江宇视科技有限公司 | Data storage method, device, equipment and medium |
| CN112291292B (en) * | 2019-07-26 | 2023-04-07 | 浙江宇视科技有限公司 | Data storage method, device, equipment and medium |
| CN111031567A (en) * | 2020-01-14 | 2020-04-17 | 南通先进通信技术研究院有限公司 | Flow statistical method for core network equipment |
| CN111031567B (en) * | 2020-01-14 | 2022-06-24 | 南通先进通信技术研究院有限公司 | Flow statistical method for core network equipment |
| CN111708606A (en) * | 2020-06-16 | 2020-09-25 | 杨海朝 | Virtual computer storage service system and use method thereof |
| CN112468590A (en) * | 2020-11-27 | 2021-03-09 | 杭州海康威视系统技术有限公司 | Storage resource mounting method and device |
| CN112468590B (en) * | 2020-11-27 | 2023-02-07 | 杭州海康威视系统技术有限公司 | Storage resource mounting method and device |
| CN117194439A (en) * | 2023-11-07 | 2023-12-08 | 杭州优云科技有限公司 | Method for creating resource storage system, electronic equipment and storage medium |
| CN117194439B (en) * | 2023-11-07 | 2024-03-22 | 杭州优云科技有限公司 | Method for creating resource storage system, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107087031B (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107087031A (en) | A kind of storage resource load-balancing method and device | |
| US10637916B2 (en) | Method and device for storage resource allocation for video cloud storage | |
| CN102880557B (en) | look-up method of multistage distribution type high-speed cache of heterogeneous data source | |
| CN104937584B (en) | Based on the quality of shared resource to the service quality of virtual machine and application program offer optimization through priority ranking | |
| CN107395659A (en) | A kind of method and device of service handling and common recognition | |
| US9020893B2 (en) | Asynchronous namespace maintenance | |
| JP5160408B2 (en) | Method, system, and computer program for selecting a resource manager to satisfy a service request | |
| US9860317B1 (en) | Throughput throttling for distributed file storage services with varying connection characteristics | |
| CN103167026B (en) | A kind of cloud store environmental data processing method, system and equipment | |
| US9313270B2 (en) | Adaptive asynchronous data replication in a data storage system | |
| US10516732B2 (en) | Disconnected ingest in a distributed storage system | |
| US9805140B2 (en) | Striping of directed graphs and nodes with improved functionality | |
| US10684996B1 (en) | Distributing data of multiple logically independent file systems in distributed storage systems including physically partitioned disks | |
| CN104301671B (en) | Traffic Surveillance Video storage method based on event closeness in HDFS | |
| US10129327B2 (en) | Retrieval of a file from multiple storage nodes | |
| US20170262463A1 (en) | Method and system for managing shrinking inode file space consumption using file trim operations | |
| US20190286629A1 (en) | Method for processing transactions using blockchain network, and transaction management server using the same | |
| WO2022257615A1 (en) | Information processing method and apparatus, and storage medium | |
| CN105224244A (en) | The method and apparatus that a kind of file stores | |
| Gill et al. | Dynamic cost-aware re-replication and rebalancing strategy in cloud system | |
| CN110244901A (en) | Method for allocating tasks and device, distributed memory system | |
| CN108073723A (en) | A kind of file in distributed type assemblies storage is from compressing method and equipment | |
| CN110457307A (en) | Metadata management system, user's cluster creation method, device, equipment and medium | |
| US10545677B2 (en) | Volatile account identification and isolation and resource management in distributed data storage systems | |
| CN109286532B (en) | Management method and device for alarm information in cloud computing system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221111 Address after: 15 / F, Zhongchuang Plaza, 385 Hangtian Middle Road, national civil aerospace industrial base, Xi'an City, Shaanxi Province 710100 Patentee after: Xi'an Yu vision Mdt InfoTech Ltd. Address before: 310051 Zhejiang Jiangling Hangzhou Road, Binjiang District, Jiangling, 88, No. 10 South Block 1-11. Patentee before: ZHEJIANG UNIVIEW TECHNOLOGIES Co.,Ltd. |
|
| TR01 | Transfer of patent right |