CN104023058B - Method and system for managing and allocating storage resources - Google Patents
Method and system for managing and allocating storage resources Download PDFInfo
- Publication number
- CN104023058B CN104023058B CN201410245074.1A CN201410245074A CN104023058B CN 104023058 B CN104023058 B CN 104023058B CN 201410245074 A CN201410245074 A CN 201410245074A CN 104023058 B CN104023058 B CN 104023058B
- Authority
- CN
- China
- Prior art keywords
- slice
- datanode
- storage resource
- encoders
- distribution
- 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.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method and system for managing and allocating storage resources. The method comprises the steps that: a data management node DataNode receives a storage space allocating request of an IP coder; the DataNode selects storage resources for the IP coder according to a first strategy, and allocates storage space slices in the selected storage resources to the IP coder according to a second strategy so that the IP coder mounts the storage resources through an iSCSI according to iSCSI mounting information of the selected storage resources, and performs iSCSI data writing according to initial addresses and space sizes of the allocated storage space slices. Meanwhile, the invention further discloses a system for managing and allocating the storage resources.
Description
Technical field
The present invention relates to the Media Stream iSCSI blocks of internet directly deposit technical field, more particularly to a kind of pipe of storage resource
Reason and distribution method and system.
Background technology
With application of the high definition in monitoring business system, reliability, performance of monitoring video storage etc. are required more next
It is higher.And Internet protocol (IP, Internet Protocol) encoder in front end is to IP storage area networks (IPSAN, IP
Storage Area Network) the block of Media Stream end to end directly the technology of depositing is exactly a kind of preferably video recording storage scheme.Such as
It is that IP encoders directly deposit schematic diagram to the block of Media Stream end to end of IPSAN in correlation technique shown in Fig. 1, on storage mode,
The data management mode that Media Stream block is directly deposited has abandoned media server;On the basis of IP network, it is integrated with the encoder
ISCSI modules so that encoder end to end can be write Video data in IPSAN storage devices based on iSCSI protocol.When
When the image of control point needs to carry out centrally stored, front end IP encoders are set up iSCSI and are connected with IPSAN storage devices, will
The storage resource of IPSAN storage devices is mounted to IP encoders locally by iSCSI protocol, then carries out to storing video flowing
ISCSI protocol is encapsulated, and directly media data is write in IPSAN storage devices by the way of data block.
And on the basis of media stream data block directly deposits technology, a key is become to the allocation manager of memory space and is asked
Topic, video camera video recording memory space needs to meet the demand of the video recording retention phase that user specifies, while flexible dilatation can be supported,
Additionally need raising space availability ratio, it is to avoid waste of storage space.
In correlation technique, the space distribution mechanism of iSCSI blocks storage is monitored, be logic-based volume management (LVM, Logic
Volume Manage) storage virtualization technology, i.e., shielding bottom physical storage entity in the case of, be each video camera
Individually distribute one section of logical memory space, the logical memory space is referred to as storage resource.The hierarchy of block SRM
As shown in Fig. 2 so-called storage resource, is LUN (LUN, the Logic Unit by iSCSI Target (label)
Number) bound with the logical volume of LVM layers, so as to the one section of logical address space provided for iSCSI storage services.And
LVM layers are made up of different entities such as physical volume (PV), volume group (VG) and logical volumes (LV).Physical volume is and physical array or thing
The corresponding entity of reason disk, volume group is the set of the physical volume of all addition this volume groups;And logical volume is divided in volume group
The one piece of logical address space matched somebody with somebody, in iSCSI storage applications, is bound with the LUN of iSCSI Target.
By such scheme, a kind of virtualized storage space management mechanism can be provided for upper strata storage service, be realized
The shielding of the physical disk array actual to bottom.LVM layers only support at present distribution, deletion, the dilatation of logical volume, do not support
The capacity reduction of single logical volume, at the same logical volume itself do not provide small grain size, can allocation manager memory cell, therefore logic
Volume is substantially the technology that a kind of space exclusively enjoys.It is camera allocation storage resource in monitoring iSCSI blocks directly deposit business, for
LVM layers, really the volume group in physical volume composition spatially distribute one piece of logical volume, and by logical volume and are video camera iSCSI
Directly deposit [Target, the LUN] for being created to be bound.So, in the iSCSI Initiator of front end IP encoders, (iSCSI sends out
Rise) hold and set up after ISCSI connections with [Target, the LUN] in IPSAN storage devices, can distribute in IPSAN storage devices
Logical volume be mapped as a Logical Disk of front-end local, Logical Disk is write data to self-defining piece of data form, so as to
Realization by iSCSI protocol is directly stored to media stream data in storage device.
The space distribution mechanism of current iSCSI blocks storage, is that one kind disposably distributes to single storage resource space
Single camera is simultaneously exclusively enjoyed the mechanism for writing the space by the video camera.For alarm linkage class storage or storage code check are moved
The storage service of state change, due to cannot in advance estimate out the size of video camera memory space, therefore is not suitable for using this pre-
First distribute the mechanism of fixed storage space for single camera.Under many circumstances, some old monitoring office points can newly purchase one
A little storage devices or increase disk in original storage device, but the program cannot be supported to provide the storage of existing video camera
, there is more inconvenience in system upgrade transformation in automatic dilatation of the source in system newly-increased storage device, disk.
The content of the invention
To solve existing technical problem, the embodiment of the present invention provides a kind of management of storage resource and distribution method
And system.
Management and the distribution method of a kind of storage resource are embodiments provided, the method includes:
Data management node DataNode receives the distribution memory space request of IP encoders;
The DataNode selects storage resource according to the first strategy for the IP encoders, and is institute according to the second strategy
The memory space piece slice in the selected storage resource of IP encoders distribution is stated, so that the IP encoders are according to selected storage money
The iSCSI carry information in source, by iSCSI carry storage resources, initial address according to the memory space slice for being distributed,
Space size carries out iSCSI data writes.
Wherein, first strategy includes:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently to deposit
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio, then the storage resource for selecting next residual memory space maximum.
Wherein, the method is further included:The DataNode carries out cutting to storage resource in units of slice, and
The slice in storage resource is managed by idle slice queues and occupancy slice queues;
Second strategy includes:During free time slice queues are selected in the storage resource of IP encoders request
Slice distributes to the IP encoders, and after response success, the slice of this sub-distribution is updated to into seizure condition and is dispatched
To in occupancy queue.
Wherein, the method is further included:When the request of the new slice of distribution that the IP encoders send is received, institute
The request that DataNode distributes new slice according to the IP encoders is stated, in judging the storage resource belonging to the former slice
Whether the occupancy of slice reaches the 3rd preset ratio, if not up to the 3rd preset ratio, continues as the IP encoders
The idle slice distributed in identical storage resource, if reaching the 3rd preset ratio, selects described for the IP encoders
Other storage resources of DataNode;
If the slice occupancies of all storage resources of the DataNode all reach the 3rd preset ratio,
Delete the aging data in the DataNode and hang the corresponding slice of deleted data and return to idle slice queues, be described
IP encoders select the maximum storage resource in residue slice spaces, and distribute the idle slice of selected storage resource to the IP
Encoder;
Wherein, the request of the new slice of the distribution, is written in the space of distributed slice by the IP encoders
When ratio reaches the second preset ratio, send to the DataNode.
Wherein, the method is further included:
Before the IP encoders send distribution memory space request to DataNode, the IP encoders are to Master
Management node request distribution DataNode;
The Master management nodes distribute the DataNode according to the 3rd strategy for the IP encoders.
Wherein, the 3rd strategy includes:
Residual memory space, remaining input and output I O process ability according to each DataNode, and based on load balancing, net
The direct-connected factor of network, is the IP encoders distribution DataNode.
Wherein, the Master management nodes include main Master management nodes and auxiliary Master management nodes, the method
Further include:
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the master
During Master management node failures, Master management nodes based on the auxiliary Master management nodes switching.
Wherein, the method is further included:The Master management nodes are in the DataNode failures or suspension
IP encoders corresponding to the failure or the DataNode of suspension distribute new available DataNode.
The embodiment of the present invention additionally provides a kind of management of storage resource and distribution system, and the system includes:IP encoders
With data management node DataNode, wherein,
The IP encoders are used for, and to the DataNode distribution memory space request is sent;
The DataNode is used for, and storage resource is selected for the IP encoders according to the first strategy, and according to the second plan
Memory space piece slice in storage resource selected by slightly described IP encoders distribution;
The IP encoders are additionally operable to, and according to the iSCSI carry information of selected storage resource, are stored by iSCSI carries
Resource, the initial address, space size according to the memory space slice for being distributed carries out iSCSI data writes.
Wherein, the DataNode is further used for, and storage money is selected for the IP encoders according to the following first strategy
Source:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently to deposit
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio, then the storage resource for selecting next residual memory space maximum.
Wherein, the DataNode is further used for, and cutting is carried out in units of slice to storage resource, and by sky
Not busy slice queues and occupancy slice queues are managed to the slice in storage resource;
The DataNode is according to the following second strategy distribution slice:In the storage resource of IP encoders request
The slice in free time slice queues is selected to distribute to the IP encoders, and after response success, by the slice of this sub-distribution
It is updated to seizure condition and is dispatched in occupancy queue.
Wherein, the IP encoders are further used for, the space of distributed slice be written into ratio reach second preset
During ratio, to the DataNode the new slice of distribution is asked;
The DataNode is further used for, and according to the IP encoders request of new slice is distributed, and judges the original
Whether the occupancy of slice reaches the 3rd preset ratio in storage resource belonging to slice, if the not up to the 3rd default ratio
Example, then continue as the idle slice that the IP encoders distribute in identical storage resource, if reaching the 3rd preset ratio,
For other storage resources that the IP encoders select the DataNode;
If the slice occupancies of all storage resources of the DataNode all reach the 3rd preset ratio,
Delete the aging data in the DataNode and hang the corresponding slice of deleted data and return to idle slice queues, be described
IP encoders select the maximum storage resource in residue slice spaces, and distribute the idle slice of selected storage resource to the IP
Encoder.
Wherein, the system also includes:Master management nodes;
The IP encoders are further used for, before distribution memory space request is sent to the DataNode, to institute
State Master management nodes request distribution DataNode;
The Master management nodes are used for, after the request for receiving IP encoders distribution DataNode, according to the
Three strategies distribute the DataNode for the IP encoders.
Wherein, the Master management nodes are further used for, and are the distribution of IP encoders according to the following 3rd strategy
DataNode:
Residual memory space, remaining input and output I O process ability according to each DataNode, and based on load balancing, net
The direct-connected factor of network, is the IP encoders distribution DataNode.
Wherein, the Master management nodes include main Master management nodes and auxiliary Master management nodes;
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the master
During Master management node failures, Master management nodes based on the auxiliary Master management nodes switching.
Wherein, the Master management nodes are further used for, and are the event in the DataNode failures or suspension
IP encoders corresponding to barrier or the DataNode of suspension distribute new available DataNode.
A kind of management of storage resource that the embodiment of the present invention is provided and distribution method and system, are keeping Media Stream
On the premise of the high-performance benefits that iSCSI is directly deposited, unit cutting and the Dynamic Elastic distribution mechanism of memory space is realized, met
Alarm linkage storage, variable code rate storage etc. cannot estimate the scene demand of video camera memory space, reach at utmost to improve and deposit
The effect of storage space availability ratio;Also meet existing video camera memory space automatic after system increases new equipment, new disk
Dilatation demand;Support video camera memory space in any one distribution on multiple devices of storage device cluster;Based on distributed
Storage cluster mechanism, realizes to Master management nodes, the failure tolerant mechanism of any DataNode, ensures to the full extent
The video recording storage of IP encoders is not interrupted, and improves the reliability of video recording storage service.
Description of the drawings
Fig. 1 is that IP encoders directly deposit schematic diagram to the block of Media Stream end to end of IPSAN in correlation technique;
Fig. 2 is the hierarchy schematic diagram of block SRM in correlation technique;
Fig. 3 is a kind of flow chart of the management and distribution method of storage resource of the embodiment of the present invention;
Management and the structural representation of distribution system of the Fig. 4 for a kind of storage resource of the embodiment of the present invention;
Fig. 5 is a kind of hardware networking schematic diagram of the embodiment of the present invention;
Fig. 6 is the flow chart of the management and distribution method of another kind of storage resource of the embodiment of the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings the technical solution of the present invention is further elaborated with specific embodiment.
A kind of management of storage resource provided in an embodiment of the present invention and distribution method, as shown in figure 3, mainly including:
Step 301, data management node DataNode receives the distribution memory space request of IP encoders.
The DataNode of the embodiment of the present invention is deployed in IPSAN storage devices, is set for managing its affiliated IPSAN storage
Standby storage resource, DataNode is cut with piece (slice) to the storage resource of its affiliated IPSAN storage device as unit
Point, and the slice in storage resource is managed by idle slice queues and occupancy slice queues, it is unassigned
Slice is idle condition and is added in idle queues that allocated slice is seizure condition and is dispatched to occupancy queue
In.
IP encoders can include being packaged into the encoder of IP packets with encoding function and by coded data, may be used also
To include with camera function, and by the coded image data after shooting and it is packaged into the web camera IPC of IP packets.
When IP encoders need memory space to carry out the write of iSCSI data, sending distribution memory space to DataNode please
Ask.
Step 302, DataNode selects storage resource according to the first strategy for IP encoders, and is IP according to the second strategy
Memory space slice in the selected storage resource of encoder distribution, so that IP encoders are hung according to the iSCSI of selected storage resource
Information carrying ceases, and by iSCSI carry storage resources, the initial address, space size according to the memory space slice for being distributed is carried out
ISCSI data write.
DataNode, according to the request of IP encoders, is that the IP encoders select storage resource, and in selected storage money
Distribute memory space slice in source;IP encoders are deposited according to the iSCSI carry information of selected storage resource by iSCSI carries
Storage resource, the initial address, space size according to the memory space slice for being distributed carries out iSCSI data writes.
Wherein, first strategy includes:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently to deposit
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio, then the storage resource for selecting next residual memory space maximum.
That is, prioritizing selection currently distributes the slice storage resources at place.Wherein, the first preset ratio can be with root
Set according to being actually needed, for example, it can be set to being 90%.
Second strategy includes:During free time slice queues are selected in the storage resource of IP encoders request
Slice distributes to the IP encoders, and after response success, the slice of this sub-distribution is updated to into seizure condition and is dispatched
To in occupancy queue.
In the embodiment of the present invention, IP encoders are written into ratio and (are distributing to IP codings in the space of distributed slice
The ratio shared by the slice spaces of data is had been written in the slice spaces of device) when reaching the second preset ratio, Ke Yixiang
The new slice of DataNode request distribution, while IP encoders continue data write in former slice;Wherein, described second is pre-
If ratio can be set according to actual needs, for example, it can be set to being 90%;
When the request of the new slice of distribution that the IP encoders send is received, DataNode judges the former slice institutes
(the slice spaces that data are had been written in i.e. described storage resource account for the storage to the occupancy of slice in the storage resource of category
The ratio in the total slice spaces of resource) the 3rd preset ratio whether is reached, if not up to the 3rd preset ratio, continues as institute
The idle slice that IP encoders distribute in identical storage resource is stated, if reaching the 3rd preset ratio, for the IP encoders
Select other storage resources of the DataNode;Wherein, the 3rd preset ratio can be set according to actual needs
(can be identical with the second preset ratio, it is also possible to different), for example, it can be set to being 90%;
If the slice occupancies of all storage resources of the DataNode all reach the 3rd preset ratio,
Delete the aging data (the such as not high data of the most long data of storage time or importance) in the DataNode and incite somebody to action
The corresponding slice of deleted data hangs and returns to idle slice queues, is that the IP encoders select residue slice spaces maximum
Storage resource, and distribute the idle slice of selected storage resource to the IP encoders.
It is described before the IP encoders send distribution memory space request to DataNode in the embodiment of the present invention
IP encoders to Master management nodes request distributes DataNode;
The Master management nodes distribute the DataNode according to the 3rd strategy for the IP encoders;
Wherein, Master management nodes are used to being managed collectively and distributing DataNode.
3rd strategy includes:Residual memory space, remaining input and output (IO) disposal ability according to each DataNode, and
It is described based on load balancing, direct network connection (i.e. as far as possible to the storage resource of IP encoders distribution same subnet address) factor
IP encoders distribute DataNode.Wherein, the residual memory space of each DataNode, remaining I O process ability can be by each
DataNode reports respectively Master management nodes.
By inventive method above embodiment, on the premise of the high-performance benefits for keeping Media Stream iSCSI directly to deposit,
Support video camera memory space in any one distribution on multiple devices of storage device cluster;Realize the unit of memory space
Cutting and Dynamic Elastic distribution mechanism, meeting alarm linkage storage, variable code rate storage etc. cannot estimate video camera memory space
Scene demand, reaches the effect at utmost improving memory space utilization rate.
Preferably, the Master management nodes can include main Master management nodes and auxiliary Master management nodes,
The method of the embodiment of the present invention is further included:
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the master
During Master management node failures, Master management nodes based on the auxiliary Master management nodes switching.
By the redundancy protecting of major-minor Master management nodes, ensured Master management nodes by fault impact so as to
Normal service is provided.
Preferably, the method for the embodiment of the present invention is further included:The Master management nodes are in the DataNode
When failure or suspension, be the failure or suspension DataNode corresponding to IP encoders distribute new available DataNode.
Specifically, the Master management nodes carry out active probe to each DataNode, to judge that each DataNode is
It is no working properly, by detection failure DataNode be set to down status, will detect successfully but with the Communications failure of IP encoders
Counting reaches the DataNode of predetermined threshold value and is set to failure state, and alerts to system manager's triggering, carries out manual intervention, will
Detect the DataNode for successfully and with the Communications failure of IP encoders counting not up to predetermined threshold value and be set to presence;
Accordingly, the Master management nodes are to select described new can use from the DataNode of presence
DataNode。
The embodiment of the present invention additionally provides a kind of management of storage resource and distribution system, as shown in figure 4, mainly including:
DataNode10 and IP encoders 20, wherein,
IP encoders 20 are used for, and to DataNode10 distribution memory space request is sent;
DataNode10 is used for, and storage resource is selected for the IP encoders 20 according to the first strategy, and according to the second plan
The memory space slice that slightly described IP encoders 20 distribute in selected storage resource;
IP encoders 20 are additionally operable to, and according to the iSCSI carry information of selected storage resource, are provided by the storage of iSCSI carries
Source, the initial address, space size according to the memory space slice for being distributed carries out iSCSI data writes.
Preferably, DataNode10 is further used for, and storage is selected for the IP encoders 20 according to the following first strategy
Resource:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently to deposit
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio, then the storage resource for selecting next residual memory space maximum.
Preferably, DataNode10 is further used for, and cutting is carried out in units of slice to storage resource, and by sky
Not busy slice queues and occupancy slice queues are managed to the slice in storage resource;
DataNode10 is according to the following second strategy distribution slice:In the storage resource of the IP encoders 20 request
The slice in free time slice queues is selected to distribute to the IP encoders 20, and after response success, by this sub-distribution
Slice is updated to seizure condition and is dispatched in occupancy queue.
Preferably, IP encoders 20 are further used for, the space of distributed slice be written into ratio reach second preset
During ratio, to the DataNode10 the new slice of distribution is asked;
The DataNode10 is further used for, and according to the IP encoders 20 request of new slice is distributed, and judges described
Whether the occupancy of slice reaches the 3rd preset ratio in storage resource belonging to former slice, if the not up to the 3rd default ratio
Example, then continue as the idle slice that the IP encoders 20 distribute in identical storage resource, if reaching the 3rd preset ratio,
Other storage resources of the DataNode10 are then selected for the IP encoders 20;
If the slice occupancies of all storage resources of the DataNode10 all reach the 3rd preset ratio,
Then delete the aging data in the DataNode10 and hang the corresponding slice of deleted data and return to idle slice queues, be
The IP encoders 20 select the maximum storage resource in remaining slice spaces, and the idle slice of storage resource gives selected by distributing
The IP encoders 20.
Preferably, the system also includes:Master management nodes 30,
IP encoders 20 are further used for, before distribution memory space request is sent to DataNode10, to described
The request distribution DataNode10 of Master management nodes 30;
The Master management nodes 30 are used for, after the request for receiving the distribution of IP encoders 20 DataNode, root
Distribute the DataNode10 for the IP encoders 20 according to the 3rd strategy.
Preferably, Master management nodes 30 are further used for, and are distributed for IP encoders 20 according to the following 3rd strategy
DataNode10:
Residual memory space, remaining I O process ability according to each DataNode10, and based on load balancing, direct network connection
Factor, is the distribution of IP encoders 20 DataNode10.
Preferably, Master management nodes 30 include main Master management nodes and auxiliary Master management nodes;
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the master
During Master management node failures, Master management nodes based on the auxiliary Master management nodes switching.
Preferably, Master management nodes 30 are further used for, and are described in the DataNode10 failures or suspension
IP encoders 20 corresponding to failure or the DataNode10 of suspension distribute new available DataNode.
Specifically, 30 couples of each DataNode10 of the Master management nodes carry out active probe, each to judge
Whether DataNode working properly, by detection failure DataNode be set to down status, will detect successfully but with IP encoders
Communications failure count and reach the DataNode of predetermined threshold value and be set to failure state, and alert to system manager's triggering, carry out
Manual intervention, will detect the DataNode for successfully and with the Communications failure of IP encoders counting not up to predetermined threshold value and be set to
Line states;
Accordingly, the Master management nodes 30 are to select described new can use from the DataNode of presence
DataNode。
Introduce in conjunction with Fig. 4 below a kind of including data management node DataNode10, master (Master) management node 30
Management and distribution system with the storage resource of IP encoders 20, wherein,
DataNode10, for managing the storage resource of its affiliated IPSAN storage device, is disappeared by sending registration, keep-alive
Breath, reports the residual memory space of the affiliated IPSAN storage devices of DataNode10, remaining input defeated to Master management nodes 30
Go out (IO) disposal ability;
Master management nodes 30, for registration, keep-alive message by receiving DataNode10, obtain and manage each
The residual memory space of IPSAN storage devices, remaining I O process ability;It is additionally operable to receiving the request from IP encoders 20
After distribution DataNode messages, the residual memory space, remaining I O process ability according to its each IPSAN storage device for managing,
DataNode is selected for IP encoders 20;
IP encoders 20, for distributing DataNode to the request of Master management nodes 30, and obtain Master management sections
Point 30 is the information of the DataNode of the distribution of IP encoders 20;It is additionally operable to the DataNode according to the distribution of Master management nodes 30
Information, to corresponding DataNode10 request distribution storage resource;
DataNode10 is additionally operable to, and receives after the request of the distribution storage resource of IP encoders 20, is IP encoders
20 selections for carrying out storage resource, and the information of selected storage resource is sent to into IP encoders 20.
Preferably, DataNode10 is further used for, to the storage resource of its affiliated IPSAN storage device with piece
(slice) cutting is carried out for unit, and records affiliated storage resource mark, initial address, space size, the shape of all slice
State, the state includes idle condition and seizure condition;All slice idle queues and occupancy team to each storage resource
Row are managed, and unassigned slice is idle condition and is added in idle queues that allocated slice is occupancy
State is simultaneously dispatched in occupancy queue.
Preferably, DataNode10 is further used for, one or more storages when its affiliated IPSAN storage device configurations
It is the selection that IP encoders 20 carry out storage resource according to following principle during resource:
If currently distributed in the storage resource of slice, still the slice of available free state, then continue to select current
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio (such as 90%), then the storage resource for selecting next remaining space maximum.
Preferably, the information of storage resource selected by DataNode10 includes:The unique mark and iSCSI of selected storage resource
Carry information;
IP encoders 20 are further used for, and after the information for receiving storage resource selected by DataNode10, perform corresponding
The iSCSI carries of storage resource, and after carry success, in the local generation logical DOS drives of IP encoders 20, and to
DataNode10 requests distribute memory space in selected storage resource.
Preferably, DataNode10 is further used for, after the request for receiving the distribution memory space of IP encoders 20, according to
The storage resource mark carried in the request, in the idle slice queues of correspondence storage resource slice is distributed, and by institute
The affiliated resource identifications of distribution slice, initial address, the space size response for distributing slice are to IP encoders 20;Response into
After work(, the slice of this sub-distribution is updated to into seizure condition and is dispatched in occupancy queue.
Preferably, IP encoders 20 are further used for, after the response for receiving DataNode10 distribution slice, according to institute
State the resource identification in response and obtain corresponding local drive, and according to initial address, the space size of distribution slice, convection current
Media block carries out iSCSI and directly deposits the memory space for being written as the distribution of IP encoders 20.
Preferably, the Master management nodes 30 in above-described embodiment can be deployed in data management (DM, Data
Management) in server, DataNode10 can be deployed in IPSAN storage devices, and the multiple stage in the system
IPSAN storage devices constitute storage device cluster.
As shown in figure 5, a kind of hardware networking schematic diagrams of the Fig. 5 for the embodiment of the present invention, in the networking plan, mainly
Including following equipment:
IP encoders, wherein deployment Client modules, support to deposit the iSCSI blocks of each camera chin video recording of IP encoders
Storage.In this programme, Client modules are disposed on IP encoders, for the storage device of this programme, the dynamic of memory space
Distribution mechanism;Client modules to realize above-mentioned system shown in Figure 4 in the function of IP encoders 20 that refers to.
DM servers, wherein deployment Master management nodes, i.e. Master management nodes are deployed in the management servers such as DM
On.On the one hand, by registration, the keep-alive message of reception IPSAN storage devices, all IPSAN storages of management the system are obtained and is set
The information such as standby residual memory space, remaining I O process ability, IP address, presence;On the other hand, in the record of IP encoders
As in iSCSI Stored Procedures, based on factors such as load balancing, direct network connections, the IPSAN for distributing storage video recording to IP encoders sets
It is standby.
In addition, to improve Master management node reliabilities of service, supporting following high-availability mechanism:
In device level, Master management nodes support two-node cluster hot backup mechanism, realize that master-slave redundancy is protected;In primary Master
Management node is delayed after machine, and standby Master management nodes can at short notice switch to primary node, and to outside clothes are provided
Business;
On the network link, network link redundancy scheme is supported;In the case where primary link failure disconnects, can pass through standby
With link, the reliable network connection between Master management nodes and external equipment is realized.
IPSAN storage devices, wherein disposing DataNode.In the scheme of the embodiment of the present invention, the storage of multiple stage IPSAN sets
It is standby to constitute storage device cluster;Meanwhile, in the IPSAN storage devices of the embodiment of the present invention, dispose DataNode.
On the one hand, DataNode by being registered to Master management nodes, keep-alive, report belonging to DataNode
The information such as the residual memory space of IPSAN storage devices, remaining I O process ability, IP address, presence, support Master pipes
Reason node is managed the United Dispatching of storage device cluster;
Meanwhile, DataNode is supported to all storage resources in its affiliated IPSAN storage device, is entered in units of slice
Line storage unit cutting, distribution, management.
Slice way to manages are as follows:
After IPSAN storage devices create arbitrary storage resource, DataNode is automatically according to certain space unit granularity (ratio
Such as 256M bytes), it is several slice by the cutting of storage resource space.The all slice's of DataNode minute book equipment is affiliated
The information such as storage resource mark, initial address, space size, state (idle or occupancy);
All slice to each storage resource, DataNode are come using two queues such as idle queues, occupancy queues
It is managed.The new all slice for creating storage resource, are linked into first in idle queues;It is empty in the storage for receiving IP encoders
Between application request after, DataNode distributes slice from the idle slice queues of specified storage resource and gives IP encoders, is used for
The storage of new video recording;After distribution slice successes, it is seizure condition that DataNode changes allocated slice states, and will
Allocated slice is dispatched to occupancy queue from idle queues.
By a kind of management of storage resource of management and the distribution system realization of the above-mentioned storage resource of the embodiment of the present invention
And distribution method, including:
DataNode manages the storage resource of its affiliated IPSAN storage device, by sending registration, keep-alive message, to
Master management nodes report the residual memory space of the affiliated IPSAN storage devices of DataNode, remaining I O process ability;
Master management nodes are obtained and are managed each IPSAN storages and set by registration, the keep-alive message of reception DataNode
Standby residual memory space, remaining I O process ability;And distribute IPSAN storage devices the request from IP encoders is received
Message after, residual memory space according to its each IPSAN storage device for managing, remaining I O process ability, are IP encoders
Select IPSAN storage devices;
The information of the IPSAN storage devices that IP encoders distribute according to Master management nodes, to corresponding DataNode
Request distribution storage resource;
DataNode is received after the request of the distribution storage resource of IP encoders, is that IP encoders carry out storage resource
Selection, and the information of selected storage resource is sent to into IP encoders.
Preferably, the method is further included:
DataNode carries out cutting to the storage resource of its affiliated IPSAN storage device in units of slice, and records institute
There are affiliated storage resource mark, initial address, space size, the state of slice, the state includes idle condition and takes shape
State;All slice idle queues and occupancy queue to each storage resource are managed, and unassigned slice is sky
Not busy state is simultaneously added in idle queues, and allocated slice is seizure condition and is dispatched in occupancy queue.
Preferably, the method is further included:Multiple storage moneys when the IPSAN storage device configurations belonging to DataNode
It is the selection that IP encoders carry out storage resource according to following principle during source:
If currently distributed in the storage resource of slice, still the slice of available free state, then continue to select current
Storage resource;And,
If it is first pre- that the occupancy slice of the storage resource for currently distributing slice reaches the total slice of storage resource
If it is more than ratio (such as 90%), then the storage resource for selecting next remaining space maximum.
Preferably, the information of storage resource selected by DataNode includes:The unique mark of selected storage resource and iSCSI hang
Information carrying ceases;
The method is further included:IP encoders perform phase after the information for receiving storage resource selected by DataNode
The iSCSI carries of storage resource are answered, and after carry success, in IP encoders logical DOS drives are locally generated, and to DataNode
Request distributes memory space in selected storage resource.
Preferably, the method is further included:
DataNode receive IP encoders distribution memory space request after, according in request carry storage resource mark
Know, in the idle slice queues of correspondence storage resource slice distributed, and by the affiliated resource identifications of distributed slice, divide
IP encoders are given in initial address with slice, space size response;After response success, the slice of this sub-distribution is updated
For seizure condition and be dispatched to occupancy queue in.
Preferably, the method is further included:
After the response for receiving DataNode distribution slice, the resource identification in response obtains correspondence to IP encoders
Local drive, and according to initial address, the space size of distribution slice, streaming media block carries out iSCSI directly to be deposited and is written as IP
The memory space of encoder distribution.
Equipment network figure with reference to shown in Fig. 5, a kind of management of more specifically storage resource and distribution method embodiment
As shown in fig. 6, the method mainly includes:
Step 601, IP encoders were sent out by its client module before storage video recording is started to Master management nodes
Send the message of request distribution DataNode.
Step 602, Master management nodes are received after the request message of IP encoders, empty in storage based on each DataNode
Between, the loading condition of storage input and output (IO, Input Output) ability, select suitable DataNode, and will be selected
The information such as IP address, the PORT COM of DataNode, return to IP encoders.
As one kind preferred embodiment, Master management nodes can also simultaneously consider net when DataNode is selected
Network factor, distributes the IPSAN storage devices of same subnet address to same IP encoders as far as possible.
Step 603, the client modules of IP encoders are received after response, and record is distributed in local file
DataNode information (IP address, PORT COM including DataNode etc.), then sends request to the DataNode for being distributed
The message of distribution storage resource.
Step 604, DataNode is received after the message of request distribution storage resource, if this storage device configurations is more
Individual storage resource, then carry out the selection of storage resource according to following principle:
If currently distributed in the storage resource of slice, still the slice of available free state, then continue to select current
Storage resource;And,
If the occupancy slice of the storage resource for currently distributing slice reaches the of the total slice of the storage resource
It is more than one preset ratio (such as 90%), then the storage resource for selecting next remaining space maximum.
DataNode is completed after storage resource selects, by the unique mark of selected storage resource, storage resource
ISCSI carry information (information such as Target, LUN, initiator with resource write permission including storage resource binding) is logical
Cross response message and return to IP encoders.
Step 605, IP encoders receive distributed storage resource information (unique mark of storage resource, iSCSI carries
Information) after, iSCSI carries are performed to storage resource, and after carry success, logical DOS drives are locally generated in IP encoders,
Such as sda.
Step 606, if current date is write video recording for the first time, IP encoders to DataNode asks date created
Catalogue.
Step 607, DataNode creates corresponding date catalogue, and returns the successful respond that creaties directory to IP encoders.
Step 608, IP encoders to DataNode requests distribute memory space in storage resource, carry in request message
The unique mark of storage resource.
Step 609, DataNode is received after the memory allocation request of IP encoders, by the unique of storage resource
Mark, in the idle slice queues of correspondence storage resource slice is distributed, and by the affiliated resource unique marks of the slice,
The information answers such as initial address, the space size of slice give IP encoders;
After response success, the slice states of this sub-distribution are updated to seizure condition by DataNode, and will be corresponding
Slice is linked into occupancy queue.
Step 610, IP encoders are received after the slice distribution response messages of DataNode, unique by storage resource
Mark obtains the local drive corresponding to the resource, according to initial address, the space size of the slice of distribution, with self-defining
Block data form carry out iSCSI video recording directly deposit write the Logical Disk.Data will finally write IPSAN by iSCSI transmission channels
The storage resource space of storage device.
Step 611, when the space of distributed slice is written to the second preset ratio (such as 90%), IP encoders to
DataNode request distribution slice, carry the unique mark of the affiliated storage resources of the slice in request message;Meanwhile, IP codings
Device continues the storage of new video recording in former slice.
Step 612, DataNode holds first after the slice distribution request messages for carrying storage resource mark are received
The selection flow process of row storage resource:
Slice according to the storage resource (storage resource corresponding to resource identification carried in request message) takes and is
It is no to reach 90% and judged:
If occupation rate is less than 90%, continue the idle slice distributed in the storage resource;Otherwise, this IPSAN is selected
Other storage resources of storage device;
If the slice occupancies of all storage resources of this equipment all reach 90%, start full overlay strategy, that is, delete
Except the content of a part of oldest video recording place slice, not busy slice queues of making the return trip empty are hung, then selected remaining space maximum
Storage resource, and distribute the idle slice of the resource.
After storage resource selection and free time slice distribution is completed, DataNode to IP encoders carry out response:
If in new slice is distributed in same storage resource, then DataNode returns the starting of newly assigned slice
Information such as address, space size, the affiliated storage resource marks of slice etc. give IP encoders;
IP encoders are given if in slice is distributed in new storage resource, then in the same of the above-mentioned slice information of response
When, DataNode (Target, LUN including storage resource binding, with resource writes the iSCSI carry information of storage resource
The information such as the initiator of authority) IP encoders are also returned to together.
If IP encoders receive the iSCSI carry information of new storage resource, local iSCSI is performed to storage resource
Carry, generates local logical DOS drives.
Step 613, IP encoders record initial address, space size, the slice institutes of next slice in local cache
The information such as category storage resource mark, continuation carries out video recording storage in current slice.
Step 614, IP encoders whole spaces of current slice write it is full after, when the video recording of the slice is started
Between, video recording end time, the information such as camera identification notify Master management nodes.
Step 615, Master management nodes are by the video recording time started of slice, video recording end time, camera identification etc.
Information record in database, so as to support VM server through database realize to record a video quick-searching.
Step 616, IP encoders obtain next slice information and its storage resource mark from local cache, and obtaining should
The corresponding local logical DOS drives of storage resource, continue new video recording storage in next slice.
Machine or suspension also, it should be noted that the DataNode in the embodiment of the present invention delays, cause the record of IP encoders
When failing as storage or Signalling exchange, IP encoders can distribute DataNode to the request of Master management nodes immediately, while
Carry the DataNode marks of Communications failure;Master management nodes give IP encoders by new DataNode is distributed, it is ensured that IP
The video recording storage of encoder is not interrupted.
Now, Master management nodes are triggered to the active probe mechanism of DataNode, judge whether DataNode is normal
Work.Meanwhile, add up DataNode and count from the Communications failure of different IP encoders;
When DataNode detection failures, Master management nodes are set to down status, no longer to IP encoders point
With the DataNode.Subsequently only DataNode succeeds in registration to Master management nodes again, the just meeting of Master management nodes
It is online to change corresponding DataNode states, and to IP encoders distributes the DataNode again;
(compare when but DataNode is detected successfully to be counted and reach certain threshold value from the Communications failure of different IP encoders
Such as:Network communication between DataNode and Master management nodes is normal, but disconnects with the network of IP encoders),
Master management nodes are set to failure state, no longer distribute the DataNode to IP encoders, and are touched to system manager
The equipment of sending out is alerted with the Communications failure of IP encoders.In this case, only after keeper's manual intervention, by administration interface
DataNode is updated to into presence, Master management nodes just can redistribute the DataNode to IP encoders.
In addition, increasing IPSAN storage devices or IPSAN storage devices in embodiments of the present invention increases the feelings such as hard disk
Under condition, it is only necessary to which new IPSAN storage devices, new hard drive space are created new storage resource, and add DataNode's
Slice allocation managers.Because new storage resource space is that all video cameras are shared to use, therefore in the feelings without any expense
Under condition, the memory space dilatation to all video cameras is supported automatically.
In sum, the Master that the embodiment of the present invention is proposed and dividing that DataNode two-level managements, spatial elastic are distributed
Cloth iSCSI block directly deposits group scheme, supports that video camera memory space is any one on multiple devices in storage device cluster
Distribution.Meanwhile, the embodiment of the present invention realizes slice unit cutting management, the dynamic elasticity dispenser of iSCSI memory spaces
System, meeting alarm linkage storage, variable code rate storage etc. cannot estimate the scene demand of video camera memory space, reach at utmost
The upper effect for improving memory space utilization rate.
In Master management node failures, system automatically switches to single memory device node memory module, ensures that IP is compiled
The video recording storage of code device does not stop.Meanwhile, by local cache mechanism of the DataNode to slice video informations, ensure slice
Video information is not lost.
In any IPSAN storage device failures, new depositing is distributed to IP encoders immediately by Master management nodes
Storage equipment, the video recording storage for ensureing IP encoders does not stop.Meanwhile, realize at the Fault recovery to DataNode different faults
Reason.
In several embodiments provided by the present invention, it should be understood that disclosed method, device and electronic equipment,
Can realize by another way.Apparatus embodiments described above are only schematic, and for example, the unit is drawn
Point, only a kind of division of logic function can have other dividing mode, such as when actually realizing:Multiple units or component can
To combine, or another system is desirably integrated into, or some features can be ignored, or do not perform.In addition, shown or discussed
The coupling each other of each part or direct-coupling or communication connection can be by some interfaces, equipment or unit
INDIRECT COUPLING or communication connection, can be electrical, machinery or other forms.
The above-mentioned unit as separating component explanation can be or may not be it is physically separate, it is aobvious as unit
The part for showing can be or may not be physical location, you can with positioned at a place, it is also possible to be distributed to multiple network lists
In unit;Part or all of unit therein can according to the actual needs be selected to realize the purpose of this embodiment scheme.
In addition, each functional unit in various embodiments of the present invention can be fully integrated in a processing unit, also may be used
Being each unit individually as a unit, it is also possible to which two or more units are integrated in a unit;It is above-mentioned
Integrated unit both can be realized in the form of hardware, it would however also be possible to employ hardware adds the form of SFU software functional unit to realize.
One of ordinary skill in the art will appreciate that:Realizing all or part of step of said method embodiment can pass through
Completing, aforesaid program can be stored in a computer read/write memory medium the related hardware of programmed instruction, the program
Upon execution, the step of including said method embodiment is performed;And aforesaid storage medium includes:It is movable storage device, read-only
Memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Or, if the above-mentioned integrated unit of the embodiment of the present invention is realized and as independently using in the form of software function module
Production marketing or when using, it is also possible in being stored in a computer read/write memory medium.Based on such understanding, this
The part that the technical scheme of bright embodiment substantially contributes in other words to prior art can in the form of software product body
Reveal and, the computer software product is stored in a storage medium, including some instructions are used so that a computer sets
Standby (can be personal computer, server or network equipment etc.) performs the whole of each embodiment methods described of the invention
Or part.And aforesaid storage medium includes:Movable storage device, ROM, RAM, magnetic disc or CD etc. are various can to store journey
The medium of sequence code.
The above, the only specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, all should contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be defined by the scope of the claims.
Claims (14)
1. a kind of management of storage resource and distribution method, it is characterised in that the method includes:
Data management node DataNode receives the distribution memory space request of IP encoders;
The DataNode selects storage resource according to the first strategy for the IP encoders, and is the IP according to the second strategy
Memory space piece slice in the selected storage resource of encoder distribution, so that the IP encoders are according to selected storage resource
ISCSI carry information, by iSCSI carry storage resources, according to initial address, the space of the memory space slice for being distributed
Size carries out iSCSI data writes;
First strategy includes:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently stored money
Source;And,
If it is first pre- that the occupancy slice of the storage resource for currently distributing slice reaches the total slice of the storage resource
If it is more than ratio, then the storage resource for selecting next residual memory space maximum.
2. the management of storage resource and distribution method according to claim 1, it is characterised in that the method is further included:
The DataNode carries out cutting to storage resource in units of slice, and by idle slice queues and takes slice teams
Row are managed to the slice in storage resource;
Second strategy includes:Slice in selecting free time slice queues in the storage resource of IP encoders request
The IP encoders are distributed to, and after response success, the slice of this sub-distribution is updated to into seizure condition and occupancy is dispatched to
In queue.
3. the management of storage resource and distribution method according to claim 1, it is characterised in that the method is further included:
When the request of the new slice of distribution that the IP encoders send is received, the DataNode distributes according to the IP encoders
Whether the request of new slice, the occupancy for judging slice in the storage resource belonging to original slice reaches the 3rd preset ratio, such as
Fruit is not up to the 3rd preset ratio, then the idle slice that the IP encoders distribute in identical storage resource is continued as, if reached
To the 3rd preset ratio, then other storage resources of the DataNode are selected for the IP encoders;
If the slice occupancies of all storage resources of the DataNode all reach the 3rd preset ratio, delete
The corresponding slice of deleted data is simultaneously hung and returns to idle slice queues by aging data in the DataNode, is that the IP is compiled
Code device selects the maximum storage resource in residue slice spaces, and the idle slice of storage resource selected by distributing to encode to the IP
Device;
Wherein, the request of the new slice of the distribution, is to be written into ratio in the space of distributed slice by the IP encoders
When reaching the second preset ratio, send to the DataNode.
4. the management of storage resource and distribution method according to any one of claims 1 to 3, it is characterised in that the method is entered
One step includes:
Before the IP encoders send distribution memory space request to DataNode, the IP encoders to Master is managed
Node request distribution DataNode;
The Master management nodes distribute the DataNode according to the 3rd strategy for the IP encoders.
5. the management of storage resource and distribution method according to claim 4, it is characterised in that the 3rd strategy includes:
Residual memory space, remaining input and output I O process ability according to each DataNode, and it is straight based on load balancing, network
Even factor, is the IP encoders distribution DataNode.
6. the management of storage resource and distribution method according to claim 4, it is characterised in that the Master management nodes
Including main Master management nodes and auxiliary Master management nodes, the method is further included:
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the main Master
During management node failure, Master management nodes based on the auxiliary Master management nodes switching.
7. the management of storage resource and distribution method according to claim 4, it is characterised in that the method is further included:
The Master management nodes in the DataNode failures or suspension, be the failure or suspension DataNode corresponding to
IP encoders distribute new available DataNode.
8. a kind of management of storage resource and distribution system, it is characterised in that the system includes:IP encoders and data management section
Point DataNode, wherein,
The IP encoders are used for, and to the DataNode distribution memory space request is sent;
The DataNode is used for, and storage resource is selected for the IP encoders according to the first strategy, and is according to the second strategy
Memory space piece slice in storage resource selected by the IP encoders distribution;
The IP encoders are additionally operable to, according to the iSCSI carry information of selected storage resource, by iSCSI carry storage resources,
Initial address, space size according to the memory space slice for being distributed carries out iSCSI data writes;
The DataNode is further used for, and storage resource is selected for the IP encoders according to the following first strategy:
If currently distributing in the storage resource of slice the still slice of available free state, continue to select currently stored money
Source;And,
If it is first pre- that the occupancy slice of the storage resource for currently distributing slice reaches the total slice of the storage resource
If it is more than ratio, then the storage resource for selecting next residual memory space maximum.
9. the management of storage resource and distribution system according to claim 8, it is characterised in that the DataNode is further
For, cutting is carried out in units of slice to storage resource, and by idle slice queues and slice queues are taken to storage
Slice in resource is managed;
The DataNode is according to the following second strategy distribution slice:Select in the storage resource of IP encoders request
Slice in idle slice queues distributes to the IP encoders, and after response success, the slice of this sub-distribution is updated
For seizure condition and be dispatched to occupancy queue in.
10. the management of storage resource and distribution system according to claim 8, it is characterised in that the IP encoders enter one
Walking is used for, and when the space of distributed slice is written into ratio and reaches the second preset ratio, distributes to DataNode requests
New slice;
The DataNode is further used for, and according to the IP encoders request of new slice is distributed, and judges belonging to original slice
Storage resource in the occupancy of slice whether reach the 3rd preset ratio, if not up to the 3rd preset ratio, continues as
The idle slice that the IP encoders distribute in identical storage resource, if reaching the 3rd preset ratio, for IP codings
Device selects other storage resources of the DataNode;
If the slice occupancies of all storage resources of the DataNode all reach the 3rd preset ratio, delete
The corresponding slice of deleted data is simultaneously hung and returns to idle slice queues by aging data in the DataNode, is that the IP is compiled
Code device selects the maximum storage resource in residue slice spaces, and the idle slice of storage resource selected by distributing to encode to the IP
Device.
11. according to any one of claim 8 to 10 management of storage resource and distribution system, it is characterised in that the system
Also include:Master management nodes;
The IP encoders are further used for, before distribution memory space request is sent to the DataNode, to described
Master management nodes request distribution DataNode;
The Master management nodes are used for, after the request for receiving the IP encoders distribution DataNode, according to the 3rd plan
Slightly described IP encoders distribute the DataNode.
12. according to claim 11 management of storage resource and distribution system, it is characterised in that Master management section
Point is further used for, and DataNode is distributed for IP encoders according to the following 3rd strategy:
Residual memory space, remaining input and output I O process ability according to each DataNode, and it is straight based on load balancing, network
Even factor, is the IP encoders distribution DataNode.
13. according to claim 11 management of storage resource and distribution system, it is characterised in that Master management section
Point includes main Master management nodes and auxiliary Master management nodes;
The auxiliary Master management nodes are used to carry out redundancy protecting to the main Master management nodes, in the main Master
During management node failure, Master management nodes based on the auxiliary Master management nodes switching.
14. according to claim 11 management of storage resource and distribution system, it is characterised in that Master management section
Point be further used for, in the DataNode failures or suspension, be the failure or suspension DataNode corresponding to IP
Encoder distributes new available DataNode.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410245074.1A CN104023058B (en) | 2014-06-03 | 2014-06-03 | Method and system for managing and allocating storage resources |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410245074.1A CN104023058B (en) | 2014-06-03 | 2014-06-03 | Method and system for managing and allocating storage resources |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104023058A CN104023058A (en) | 2014-09-03 |
CN104023058B true CN104023058B (en) | 2017-05-17 |
Family
ID=51439627
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410245074.1A Active CN104023058B (en) | 2014-06-03 | 2014-06-03 | Method and system for managing and allocating storage resources |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104023058B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105898197A (en) * | 2014-12-03 | 2016-08-24 | 北京航天长峰科技工业集团有限公司 | Multi-code-stream video data direct storage method |
CN104615508A (en) * | 2015-03-03 | 2015-05-13 | 浪潮电子信息产业股份有限公司 | Method for recovering LVM configuration under Linux system |
CN104850469A (en) * | 2015-05-12 | 2015-08-19 | 浪潮电子信息产业股份有限公司 | Method for realizing data backup recovery and migration in linux system based on LV mirror image |
CN107992491A (en) * | 2016-10-26 | 2018-05-04 | 中国移动通信有限公司研究院 | A kind of method and device of distributed file system, data access and data storage |
CN108762821B (en) * | 2017-04-18 | 2023-04-25 | 海马云(天津)信息技术有限公司 | Device and method for running application of electronic equipment and electronic equipment |
CN107087031B (en) * | 2017-05-10 | 2020-07-28 | 浙江宇视科技有限公司 | Storage resource load balancing method and device |
CN111091851B (en) * | 2019-12-10 | 2021-03-02 | 四川湖山电器股份有限公司 | IPSAN disk management method of recording and broadcasting system |
CN113194158B (en) * | 2021-04-13 | 2023-02-24 | 杭州迪普科技股份有限公司 | Information storage method, device, equipment and computer readable storage medium |
CN113268344A (en) * | 2021-05-18 | 2021-08-17 | 中国联合网络通信集团有限公司 | Resource balancing method and system, first Pod node and resource proxy server |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4633802B2 (en) * | 2005-12-09 | 2011-02-16 | パナソニック株式会社 | Nonvolatile storage device, data read method, and management table creation method |
CN101232521A (en) * | 2008-02-28 | 2008-07-30 | 杭州华三通信技术有限公司 | Video storing method, coding apparatus and storing device |
CN101582903B (en) * | 2009-06-24 | 2012-08-08 | 浙江宇视科技有限公司 | Storing method of streaming media resource and device |
CN101599978B (en) * | 2009-07-17 | 2012-09-26 | 杭州华三通信技术有限公司 | Method and equipment for allocating resources |
CN102255921B (en) * | 2011-08-24 | 2013-11-27 | 杭州华三通信技术有限公司 | Media stream storage method and data management server |
-
2014
- 2014-06-03 CN CN201410245074.1A patent/CN104023058B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN104023058A (en) | 2014-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104023058B (en) | Method and system for managing and allocating storage resources | |
US9697091B2 (en) | System and method for datacenters disaster recovery | |
CN112532668B (en) | Network edge computing method, device and medium | |
CN103797770B (en) | A kind of method and system of shared storage resources | |
CN103763383B (en) | Integrated cloud storage system and its storage method | |
CN106899518A (en) | A kind of method for processing resource and device based on Internet data center | |
US20140366121A1 (en) | Virtual Private Storage Array Service for Cloud Servers | |
CN105933391A (en) | Node capacity expansion method, device and system | |
CN109828868A (en) | Date storage method, device, management equipment and dual-active data-storage system | |
CN109313564A (en) | For supporting the server computer management system of the highly usable virtual desktop of multiple and different tenants | |
CN102594852A (en) | Data access method, node and system | |
US6779082B2 (en) | Network-based disk redundancy storage system and method | |
CN102833352A (en) | Distributed cache management system and method for implementing distributed cache management | |
CN106850710A (en) | A kind of safe data cloud storage system, client terminal, storage server and application process | |
CN107888636A (en) | Data processing method and system based on cloud storage | |
JP2005222539A (en) | Storage system with capability to allocate virtual storage segment among multiple controllers | |
CN108319618B (en) | Data distribution control method, system and device of distributed storage system | |
CN106446159A (en) | Method for storing files, first virtual machine and name node | |
CN112965784B (en) | Service access method, device, equipment and computer readable storage medium | |
JP2000207370A (en) | Distributed file management device and distributed file management system | |
CN107566466A (en) | Load-balancing method and device | |
CN105162869B (en) | A kind of method and apparatus for backup data management | |
CN105187502A (en) | Method and system based on distributed elastic block storage | |
CN109871365A (en) | A kind of distributed file system | |
CN109218086A (en) | A kind of switching network construction method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |