CN104023058A - Method and system for managing and allocating storage resources - Google Patents
Method and system for managing and allocating storage resources Download PDFInfo
- Publication number
- CN104023058A CN104023058A CN201410245074.1A CN201410245074A CN104023058A CN 104023058 A CN104023058 A CN 104023058A CN 201410245074 A CN201410245074 A CN 201410245074A CN 104023058 A CN104023058 A CN 104023058A
- Authority
- CN
- China
- Prior art keywords
- slice
- storage resources
- encoder
- datanode
- management 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
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 Media Stream iSCSI piece that the present invention relates to the Internet is directly deposited technical field, relates in particular to a kind of management and distribution method and system of storage resources.
Background technology
Along with the application of high definition in monitoring business system, require more and more higher to reliability, the performance etc. of monitoring video storage.And front end Internet protocol (IP, Internet Protocol) encoder to the piece of Media Stream end to end of IP storage area network (IPSAN, IP Storage Area Network) directly the technology of depositing be exactly a kind of storage scheme of video recording preferably.As shown in Figure 1, directly deposit schematic diagram for IP encoder in correlation technique to the piece of Media Stream end to end of IPSAN, on storage mode, the data management mode that Media Stream piece is directly deposited has been abandoned media server; On the basis of IP network, in encoder, integrated iSCSI module, can write video recording data in IPSAN memory device end to end based on iSCSI agreement encoder.In the time that the image of control point need to carry out centralized stores, front end IP encoder is set up iSCSI with IPSAN memory device and is connected, the storage resources of IPSAN memory device is mounted to IP encoder this locality by iSCSI agreement, then store video stream is carried out to iSCSI protocol encapsulation, directly adopt the mode of data block that media data is write in IPSAN memory device.
And directly deposit at media stream data piece on the basis of technology, the allocation manager of memory space is become to a key issue, video camera video recording memory space need to meet the video recording of user's appointment and retain the demand of phase, can support flexible dilatation simultaneously, need in addition to improve space availability ratio, avoid waste of storage space.
In correlation technique, the allocation of space mechanism of monitoring iSCSI piece storage, logic-based volume management (LVM, Logic Volume Manage) storage virtualization technology, the in the situation that of shielding bottom physical store entity, for each video camera distributes separately one section of logical memory space, this logical memory space is called storage resources.The hierarchy of piece SRM as shown in Figure 2, so-called storage resources, by the LUN (LUN of iSCSI Target (label), Logic Unit Number) bind with the logical volume of LVM layer, thus the one section of logical address space providing for iSCSI storage service.And LVM layer is made up of different entities such as physical volume (PV), volume group (VG) and logical volumes (LV).Physical volume is the entity corresponding with physical array or physical disk, and volume group is the set of all physical volumes that add this volume group; Logical volume is the logical address space distributing in volume group, in iSCSI storage application, binds with the LUN of iSCSI Target.
By such scheme, can provide a kind of virtualized storage space management mechanism for upper strata storage service, realize the shielding of the physical disk array to bottom reality.LVM layer is only supported distribution, deletion, the dilatation of logical volume at present, does not support the capacity reduction of single logical volume, simultaneously logical volume itself do not provide small grain size, can allocation manager memory cell, therefore logical volume is the technology that a kind of space exclusively enjoys in essence.In monitoring iSCSI piece is directly deposited business, be camera allocation storage resources, for LVM layer, actual is to distribute a logical volume on the volume group space of physical volume composition, and by logical volume with bind for video camera iSCSI directly deposits created [Target, LUN].Like this, [Target on the iSCSI Initiator of front end IP encoder (iSCSI initiation) end and IPSAN memory device, LUN] set up after ISCSI connection, the logical volume distributing on IPSAN memory device can be mapped as to a Logical Disk of front end this locality, write data to Logical Disk with self-defining blocks of data form, thereby realize, media stream data is directly deposited in memory device by iSCSI agreement.
The allocation of space mechanism of current iSCSI piece storage is a kind ofly to distribute to single camera and exclusively enjoyed the mechanism that writes this space by this video camera disposable single storage resources space.For the storage service of alarm interlock class storage or the dynamic change of storage code check, owing to cannot estimating out the size of video camera memory space in advance, be therefore not suitable for adopting this mechanism of distributing in advance fixed storage space for single camera.Under many circumstances, some old monitoring office points can newly be purchased some memory devices or increase disk on original memory device, but this scheme cannot be supported the automatic dilatation in the time that system increases memory device, disk newly to existing video camera storage resources, in system upgrade transformation, has more inconvenience.
Summary of the invention
For solving the technical problem of existing existence, the embodiment of the present invention provides a kind of management and distribution method and system of storage resources.
The embodiment of the present invention provides a kind of management and distribution method of storage resources, and the method comprises:
Data management node DataNode receives the memory allocated space request of IP encoder;
Described DataNode is that described IP encoder is selected storage resources according to the first strategy, and distribute the memory space sheet slice in selected storage resources according to the second strategy for described IP encoder, for described IP encoder according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
Wherein, described the first strategy comprises:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
Wherein, the method further comprises: described DataNode carries out cutting taking slice as unit to storage resources, and by idle slice queue with take slice queue the slice in storage resources is managed;
Described the second strategy comprises: in the storage resources of described IP encoder request, select the slice in idle slice queue to distribute to described IP encoder, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
Wherein, the method further comprises: in the time receiving the request of the new slice of distribution that described IP encoder sends, the request that described DataNode distributes new slice according to described IP encoder, whether the occupancy that judges slice in the affiliated storage resources of described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode for described IP encoder;
If the slice occupancy of all storage resources of described DataNode all reaches described the 3rd preset ratio, delete the aging data in described DataNode and slice corresponding deleted data is hung and gets back to idle slice queue, for described IP encoder is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder;
Wherein, the request of the new slice of described distribution, is when being written into ratio by described IP encoder in the space of distributed slice and reaching the second preset ratio, sends to described DataNode.
Wherein, the method further comprises:
Before described IP encoder sends the request of memory allocated space to DataNode, described IP encoder distributes DataNode to the request of Master management node;
Described Master management node distributes described DataNode according to the 3rd strategy for described IP encoder.
Wherein, described the 3rd strategy comprises:
According to the residual memory space of each DataNode, residue input and output IO disposal ability, and based on load balancing, direct network connection factor, for described IP encoder distributes DataNode.
Wherein, described Master management node comprises main Master management node and auxiliary Master management node, and the method further comprises:
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
Wherein, the method further comprises: described Master management node is in the time of described DataNode fault or suspension, for the corresponding IP encoder of DataNode of described fault or suspension distributes new available DataNode.
The embodiment of the present invention also provides a kind of management and distribution system of storage resources, and this system comprises: IP encoder and data management node DataNode, wherein,
Described IP encoder is used for, and sends the request of memory allocated space to described DataNode;
Described DataNode is used for, and is that described IP encoder is selected storage resources, and distributes the memory space sheet slice in selected storage resources according to the second strategy for described IP encoder according to the first strategy;
Described IP encoder also for, according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
Wherein, described DataNode is further used for, and is that described IP encoder is selected storage resources according to following the first strategy:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
Wherein, described DataNode is further used for, and storage resources is carried out to cutting taking slice as unit, and by idle slice queue with take slice queue the slice in storage resources is managed;
Described DataNode distributes slice according to following the second strategy: in the storage resources of described IP encoder request, select the slice in idle slice queue to distribute to described IP encoder, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
Wherein, described IP encoder is further used for, and when being written into ratio and reaching the second preset ratio, distributes new slice to described DataNode request in the space of distributed slice;
Described DataNode is further used for, the request that distributes new slice according to described IP encoder, whether the occupancy that judges slice in the affiliated storage resources of described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode for described IP encoder;
If the slice occupancy of all storage resources of described DataNode all reaches described the 3rd preset ratio, delete the aging data in described DataNode and slice corresponding deleted data is hung and gets back to idle slice queue, for described IP encoder is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder.
Wherein, this system also comprises: Master management node;
Described IP encoder is further used for, and before sending the request of memory allocated space to described DataNode, distributes DataNode to described Master management node request;
Described Master management node is used for, and is receiving that described IP encoder distributes after the request of DataNode, distributes described DataNode according to the 3rd strategy for described IP encoder.
Wherein, described Master management node is further used for, and is IP encoder distribution DataNode according to following the 3rd strategy:
According to the residual memory space of each DataNode, residue input and output IO disposal ability, and based on load balancing, direct network connection factor, for described IP encoder distributes DataNode.
Wherein, described Master management node comprises main Master management node and auxiliary Master management node;
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
Wherein, described Master management node is further used for, in the time of described DataNode fault or suspension, for the corresponding IP encoder of DataNode of described fault or suspension distributes new available DataNode.
The management of a kind of storage resources that the embodiment of the present invention provides and distribution method and system, under the prerequisite of the high-performance advantage that keeps Media Stream iSCSI directly to deposit, unit cutting and the Dynamic Elastic distribution mechanism of memory space are realized, meet alarm interlock storage, variable code rate storage etc. and cannot estimate the scene demand of video camera memory space, reach the effect that at utmost improves memory space utilance; Also meet existing video camera memory space and increased the automatic dilatation demand after new equipment, new disk in system; Support video camera memory space at memory device cluster any to the distribution on multiple devices; Based on distributed storage Clustering mechanism, realize the failure tolerant mechanism to Master management node, any DataNode, ensure that to the full extent the video recording storage of IP encoder is not interrupted, improve the reliability of video recording storage service.
Brief description of the drawings
Fig. 1 is that in correlation technique, IP encoder is directly deposited schematic diagram to the piece of Media Stream end to end of IPSAN;
Fig. 2 is the hierarchy schematic diagram of piece SRM in correlation technique;
Fig. 3 is a kind of management of storage resources and the flow chart of distribution method of the embodiment of the present invention;
Fig. 4 is a kind of management of storage resources and the structural representation of distribution system 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 management of another kind of storage resources and the flow chart of distribution method of the embodiment of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, the technical solution of the present invention is further elaborated.
The management of a kind of storage resources that the embodiment of the present invention provides and distribution method, as shown in Figure 3, mainly comprise:
Step 301, data management node DataNode receives the memory allocated space request of IP encoder.
The DataNode of the embodiment of the present invention is deployed in IPSAN memory device, for managing the storage resources of its affiliated IPSAN memory device, DataNode carries out cutting to the storage resources of IPSAN memory device under it taking sheet (slice) as unit, and by idle slice queue with take slice queue the slice in storage resources is managed, the slice not being assigned with is idle condition and joins in idle queues, and the slice being assigned with is that seizure condition being dispatched to takies in queue.
IP encoder can comprise having encoding function and data encapsulation after coding is become to the encoder of IP packet, can also comprise with camera function, and by the coded image data after taking and be packaged into the web camera IPC of IP packet.
IP encoder need memory space carry out iSCSI data write fashionable, to DataNode send the request of memory allocated space.
Step 302, DataNode is IP encoder selection storage resources according to the first strategy, and distribute the memory space slice in selected storage resources according to the second strategy for IP encoder, for IP encoder according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
DataNode is according to the request of IP encoder, is that described IP encoder is selected storage resources, and in selected storage resources memory allocated space slice; IP encoder, according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carries out iSCSI data according to the initial address of distributed memory space slice, space size and writes.
Wherein, described the first strategy comprises:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
That is to say, preferentially select the storage resources at current distribution slice place.Wherein, the first preset ratio can be set according to actual needs, for example, can be set as 90%.
Described the second strategy comprises: in the storage resources of described IP encoder request, select the slice in idle slice queue to distribute to described IP encoder, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
In the embodiment of the present invention, IP encoder is written into ratio (in distributing to the slice space of IP encoder the shared ratio in slice space of data writing) and reaches the second preset ratio during in the space of distributed slice, can distribute new slice to DataNode request, while IP encoder continues data and writes in former slice; Wherein, described the second preset ratio can be set according to actual needs, for example, can be set as 90%;
In the time receiving the request of the new slice of distribution that described IP encoder sends, DataNode judges whether the occupancy (be in described storage resources the slice space of data writing account for the ratio in the total slice of described storage resources space) of slice in the storage resources under described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode for described IP encoder; Wherein, described the 3rd preset ratio can be set (can be identical with the second preset ratio, also can be different) according to actual needs, for example, can be set as 90%;
If the slice occupancy of all storage resources of described DataNode all reaches described the 3rd preset ratio, delete the aging data (data that data as the longest in memory time or importance are not high etc.) in described DataNode and slice corresponding deleting data is hung and gets back to idle slice queue, for described IP encoder is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder.
In the embodiment of the present invention, before described IP encoder sends the request of memory allocated space to DataNode, described IP encoder distributes DataNode to the request of Master management node;
Described Master management node distributes described DataNode according to the 3rd strategy for described IP encoder;
Wherein, Master management node is for unified management and distribution DataNode.
The 3rd strategy comprises: according to the residual memory space of each DataNode, residue input and output (IO) disposal ability, and based on load balancing, direct network connection (distributing the storage resources of same subnet address to IP encoder) factor as far as possible, for described IP encoder distributes DataNode.Wherein, the residual memory space of each DataNode, residue IO disposal ability can report respectively Master management node by each DataNode.
By above embodiment of the method for the present invention, under the prerequisite of the high-performance advantage that keeps Media Stream iSCSI directly to deposit, support video camera memory space at memory device cluster any to the distribution on multiple devices; Realize unit cutting and the Dynamic Elastic distribution mechanism of memory space, met alarm interlock storage, variable code rate storage etc. and cannot estimate the scene demand of video camera memory space, reached the effect that at utmost improves memory space utilance.
Preferably, described Master management node can comprise main Master management node and auxiliary Master management node, and the method for the embodiment of the present invention further comprises:
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
By the redundancy protecting of major-minor Master management node, thereby ensure that Master management node is not subject to fault effects that normal service is provided.
Preferably, the method for the embodiment of the present invention further comprises: described Master management node is in the time of described DataNode fault or suspension, for the corresponding IP encoder of DataNode of described fault or suspension distributes new available DataNode.
Concrete, described Master management node carries out active probe to each DataNode, whether working properly to judge each DataNode, be made as down status by surveying failed DataNode, to survey successfully but be made as failure state with the DataNode that the Communications failure counting of IP encoder reaches predetermined threshold value, and trigger alarm to system manager, and carry out manual intervention, will survey successfully and be made as presence with the DataNode that the Communications failure counting of IP encoder does not reach predetermined threshold value;
Accordingly, described Master management node is to select described new available DataNode from the DataNode of presence.
The embodiment of the present invention also provides a kind of management and distribution system of storage resources, as shown in Figure 4, mainly comprises: DataNode10 and IP encoder 20, wherein,
IP encoder 20 for, to DataNode10 send the request of memory allocated space;
DataNode10 is used for, and selects storage resources according to the first strategy for described IP encoder 20, and distributes the memory space slice in selected storage resources according to the second strategy for described IP encoder 20;
IP encoder 20 also for, according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
Preferably, DataNode10 is further used for, and selects storage resources according to following the first strategy for described IP encoder 20:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
Preferably, DataNode10 is further used for, and storage resources is carried out to cutting taking slice as unit, and by idle slice queue with take slice queue the slice in storage resources is managed;
DataNode10 distributes slice according to following the second strategy: in the storage resources of asking at described IP encoder 20, select the slice in idle slice queue to distribute to described IP encoder 20, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
Preferably, IP encoder 20 is further used for, and when being written into ratio and reaching the second preset ratio, distributes new slice to described DataNode10 request in the space of distributed slice;
Described DataNode10 is further used for, the request that distributes new slice according to described IP encoder 20, whether the occupancy that judges slice in the affiliated storage resources of described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder 20 and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode10 for described IP encoder 20;
If the slice occupancy of all storage resources of described DataNode10 all reaches described the 3rd preset ratio, delete the aging data in described DataNode10 and slice corresponding deleted data is hung and gets back to idle slice queue, for described IP encoder 20 is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder 20.
Preferably, this system also comprises: Master management node 30,
IP encoder 20 is further used for, and before sending the request of memory allocated space to DataNode10, distributes DataNode10 to described Master management node 30 requests;
Described Master management node 30 for, receiving that described IP encoder 20 distributes after the request of DataNode, according to the 3rd strategy distribute described DataNode10 for described IP encoder 20.
Preferably, Master management node 30 is further used for, and distributes DataNode10 according to following the 3rd strategy for IP encoder 20:
According to the residual memory space of each DataNode10, residue IO disposal ability, and based on load balancing, direct network connection factor, for described IP encoder 20 distributes DataNode10.
Preferably, Master management node 30 comprises main Master management node and auxiliary Master management node;
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
Preferably, Master management node 30 is further used for, in the time of described DataNode10 fault or suspension, for the corresponding IP encoder 20 of DataNode10 of described fault or suspension distributes new available DataNode.
Concrete, described Master management node 30 carries out active probe to each DataNode10, whether working properly to judge each DataNode, be made as down status by surveying failed DataNode, to survey successfully but be made as failure state with the DataNode that the Communications failure counting of IP encoder reaches predetermined threshold value, and trigger alarm to system manager, carry out manual intervention, will survey successfully and be made as presence with the DataNode that the Communications failure counting of IP encoder does not reach predetermined threshold value;
Accordingly, described Master management node 30 is to select described new available DataNode from the DataNode of presence.
Introduce in conjunction with Fig. 4 management and the distribution system that one comprises the storage resources of data management node DataNode10, master (Master) management node 30 and IP encoder 20 more below, wherein,
DataNode10, for managing the storage resources of its affiliated IPSAN memory device, by transmission register, keep-alive message, report residual memory space, residue input and output (IO) disposal ability of IPSAN memory device under DataNode10 to Master management node 30;
Master management node 30, for by the registration, the keep-alive message that receive DataNode10, obtains and manages residual memory space, the residue IO disposal ability of each IPSAN memory device; Also for after the request distribution DataNode message receiving from IP encoder 20, according to the residual memory space of each IPSAN memory device of its management, residue IO disposal ability, for IP encoder 20 is selected DataNode;
IP encoder 20, for distributing DataNode to Master management node 30 request, and obtains the information of the DataNode that Master management node 30 distributes for IP encoder 20; Also for the information of the DataNode that distributes according to Master management node 30, to corresponding DataNode10 request memory allocated resource;
DataNode10 also for, receive after the request from the memory allocated resource of IP encoder 20, for IP encoder 20 carries out the selection of storage resources, and the information of selected storage resources is sent to IP encoder 20.
Preferably, DataNode10 is further used for, the storage resources of IPSAN memory device under it is carried out to cutting taking sheet (slice) as unit, and recording affiliated storage resources mark, initial address, space size, the state of all slice, described state comprises idle condition and seizure condition; To all slice of each storage resources, by idle queues with take queue and manage, the slice not being assigned with is idle condition and joins in idle queues, and the slice being assigned with is that seizure condition being dispatched to takies in queue.
Preferably, DataNode10 is further used for, when IPSAN storage device configurations under it when one or more storage resources, be that IP encoder 20 carries out the selection of storage resources according to following principle:
If on the current storage resources that is distributing slice, still have the slice of idle condition, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio (as 90%) that slice reaches the total slice of described storage resources more than, select the storage resources of next remaining space maximum.
Preferably, the information of the selected storage resources of DataNode10 comprises: the unique identification of selected storage resources and iSCSI carry information;
IP encoder 20 is further used for, after receiving the information of the selected storage resources of DataNode10, carry out the iSCSI carry of respective stored resource, and after carry success, at the local formation logic drive of IP encoder 20, and to DataNode10 request memory allocated space in selected storage resources.
Preferably, DataNode10 is further used for, receiving after the request in IP encoder 20 memory allocated spaces, according to the storage resources mark of carrying in described request, in the idle slice queue of corresponding stored resource, distribute slice, and by resource identification under distributed slice, distribute initial address, the space size of slice to reply to IP encoder 20; After replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
Preferably, IP encoder 20 is further used for, receiving that DataNode10 distributes after replying of slice, obtain corresponding local drive according to the resource identification in described replying, and according to distributing initial address, the space size of slice, convection current media block carries out iSCSI and directly deposits and be written as the memory space that IP encoder 20 distributes.
Preferably, Master management node 30 in above-described embodiment can be deployed in data management (DM, Data Management) in server, DataNode10 can be deployed in IPSAN memory device, and many IPSAN memory devices composition memory device set groups in described system.
As shown in Figure 5, a kind of hardware networking schematic diagram that Fig. 5 is the embodiment of the present invention, in described networking plan, mainly comprises following equipment:
IP encoder, wherein disposes Client module, supports the iSCSI piece storage to the each camera chin video recording of IP encoder.In this programme, in IP encoder deploy Client module, for the memory device of this programme, the dynamic assignment mechanism of memory space; Client module is in order to realize the function of the IP encoder 20 of mentioning in above-mentioned system shown in Figure 4.
DM server, wherein disposes Master management node, and Master management node is deployed on the management servers such as DM.On the one hand, by receiving registration, the keep-alive message of IPSAN memory device, obtain the information such as residual memory space, residue IO disposal ability, IP address, presence of all IPSAN memory devices of management native system; On the other hand, in the video recording iSCSI of IP encoder Stored Procedure, based on the factor such as load balancing, direct network connection, give the IPSAN equipment of IP encoder memory allocated video recording.
In addition, for improving Master management node reliability of service, support following high-availability mechanism:
In device level, Master management node is supported two-node cluster hot backup mechanism, realizes master-slave redundancy protection; After primary Master management node is delayed machine, Master management node for subsequent use can switch to primary node at short notice, provides service to outside;
On network link, network enabled link redundancy mechanism; In the situation that primary link failure disconnects, can pass through reserve link, the reliable network realizing between Master management node and external equipment is connected.
IPSAN memory device, wherein disposes DataNode.In the scheme of the embodiment of the present invention, many IPSAN memory devices have formed memory device cluster; Meanwhile, on the IPSAN of embodiment of the present invention memory device, dispose DataNode.
On the one hand, DataNode by registering to Master management node, keep-alive, report the information such as residual memory space, residue IO disposal ability, IP address, presence of IPSAN memory device under DataNode, support the United Dispatching management of Master management node to memory device cluster;
Meanwhile, DataNode supports all storage resources on IPSAN memory device under it, carries out memory cell cutting, distribution, management taking slice as unit.
Slice way to manage is as follows:
Create after arbitrary storage resources at IPSAN memory device, DataNode according to certain space unit granularity (such as 256M byte), is several slice by the cutting of storage resources space automatically.The information such as the affiliated storage resources mark of all slice of DataNode minute book equipment, initial address, space size, state (idle or take);
To all slice of each storage resources, idle queues for DataNode, take two queues such as queue and manage.The new all slice that create storage resources, first hang in idle queues; After the memory space application request of receiving IP encoder, DataNode distributes slice to IP encoder from the idle slice queue of designated store resource, for the storage of new video recording; After distributing slice success, it is seizure condition that DataNode revises the slice state having distributed, and the slice having distributed is dispatched to and takies queue from idle queues.
By the management of above-mentioned storage resources and management and the distribution method of a kind of storage resources that distribution system is realized of the embodiment of the present invention, comprising:
DataNode manages the storage resources of IPSAN memory device under it, by transmissions register, keep-alive message, to Master management node report IPSAN memory device under DataNode residual memory space, remain IO disposal ability;
Master management node, by receiving registration, the keep-alive message of DataNode, obtains and manages residual memory space, the residue IO disposal ability of each IPSAN memory device; And after the message receiving from the request distributing IP SAN memory device of IP encoder, according to the residual memory space of each IPSAN memory device of its management, residue IO disposal ability, be that IP encoder is selected IPSAN memory device;
The information of the IPSAN memory device that IP encoder distributes according to Master management node, to corresponding DataNode request memory allocated resource;
DataNode receives after the request from the memory allocated resource of IP encoder, for IP encoder carries out the selection of storage resources, and the information of selected storage resources is sent to IP encoder.
Preferably, the method further comprises:
DataNode carries out cutting to the storage resources of IPSAN memory device under it taking slice as unit, and records affiliated storage resources mark, initial address, space size, the state of all slice, and described state comprises idle condition and seizure condition; To all slice of each storage resources, by idle queues with take queue and manage, the slice not being assigned with is idle condition and joins in idle queues, and the slice being assigned with is that seizure condition being dispatched to takies in queue.
Preferably, the method further comprises: when the IPSAN storage device configurations under DataNode when multiple storage resources, be the selection that IP encoder carries out storage resources according to following principle:
If on the current storage resources that is distributing slice, still have the slice of idle condition, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio (as 90%) that slice reaches the total slice of storage resources more than, select the storage resources of next remaining space maximum.
Preferably, the information of the selected storage resources of DataNode comprises: the unique identification of selected storage resources and iSCSI carry information;
The method further comprises: IP encoder is after receiving the information of the selected storage resources of DataNode, carry out the iSCSI carry of respective stored resource, and after carry success, at the local formation logic drive of IP encoder, and to DataNode request memory allocated space in selected storage resources.
Preferably, the method further comprises:
DataNode is receiving after the request in IP encoder memory allocated space, according to the storage resources mark of carrying in request, in the idle slice queue of corresponding stored resource, distribute slice, and by resource identification under distributed slice, distribute initial address, the space size of slice to reply the encoder to IP; After replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
Preferably, the method further comprises:
IP encoder is receiving that DataNode distributes after replying of slice, obtain corresponding local drive according to the resource identification in replying, and according to the initial address, the space size that distribute slice, convection current media block carries out iSCSI and directly deposits the memory space that is written as the distribution of IP encoder.
In conjunction with the equipment network figure shown in Fig. 5, as shown in Figure 6, the method mainly comprises for a kind of management of more specifically storage resources and distribution method embodiment:
Step 601, IP encoder, before starting storage video recording, sends request the message that distributes DataNode to Master management node by its client module.
Step 602, Master management node is received after the request message of IP encoder, based on each DataNode at memory space, storage input and output (IO, Input Output) loading condition of ability, select suitable DataNode, and by the information such as IP address, PORT COM of selected DataNode, return to IP encoder.
As one preferred embodiment, Master management node, in the time selecting DataNode, also can be considered network factors simultaneously, distributes the IPSAN memory device of same subnet address to same IP encoder as far as possible.
Step 603, after the client module of IP encoder is received and is replied, in local file, record distributed DataNode information (comprising IP address, the PORT COM etc. of DataNode), then send request the message of memory allocated resource to distributed DataNode.
Step 604, after DataNode receives the message of request memory allocated resource, if this storage device configurations multiple storage resources, carry out the selection of storage resources according to following principle:
If on the current storage resources that is distributing slice, still have the slice of idle condition, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio (as 90%) that slice reaches the total slice of described storage resources more than, select the storage resources of next remaining space maximum.
After DataNode completes storage resources selection, the iSCSI carry information of the unique identification of selected storage resources, storage resources (comprising the information such as Target, LUN, the initiator with resource write permission that storage resources is bound) is returned to IP encoder by response message.
Step 605, IP encoder is received distributed storage resources information (unique identification of storage resources, iSCSI carry information) afterwards, and storage resources is carried out to iSCSI carry, and after carry success, at the local formation logic drive of IP encoder, as sda.
Step 606, if current date is to write for the first time video recording, IP encoder is to DataNode request date created catalogue.
Step 607, DataNode creates corresponding date catalogue, and returns to IP encoder the successful respond that creaties directory.
Step 608, IP encoder, to DataNode request memory allocated space in storage resources, carries the unique identification of storage resources in request message.
Step 609, after DataNode receives the memory allocation request of IP encoder, by the unique identification of storage resources, in the idle slice queue of corresponding stored resource, distribute slice, and by the information answer such as initial address, space size of resource unique identification, slice under this slice to IP encoder;
After replying successfully, the slice state of this sub-distribution is updated to seizure condition by DataNode, and corresponding slice is hung into taking queue.
Step 610, after IP encoder receives that the slice of DataNode distributes response message, obtain the corresponding local drive of this resource by storage resources unique identification, according to initial address, the space size of slice distributing, carry out iSCSI video recording with self-defining blocks of data form and directly deposit and write this Logical Disk.Data will be by iSCSI transmission channel, finally writes the storage resources space of IPSAN memory device.
Step 611, be written to the second preset ratio (as 90%) during in the space of distributed slice, IP encoder distributes slice to DataNode request, carries the unique identification of the affiliated storage resources of this slice in request message; Meanwhile, IP encoder continues the storage of new video recording in former slice.
Step 612, DataNode, after receiving the slice distribution request message that carries storage resources mark, first carries out the selection flow process of storage resources:
Take and whether reach 90% and judge according to the slice of this storage resources (the corresponding storage resources of resource identification carrying in request message):
If occupation rate is less than 90%, continue to distribute the idle slice in this storage resources; Otherwise, select other storage resources of this IPSAN memory device;
If the slice occupancy of all storage resources of this equipment all reaches 90%, start full overlay strategy, delete the most always the record a video content of place slice of a part, by the not busy slice queue of making the return trip empty of its extension, then select the storage resources of remaining space maximum, and distribute the idle slice of this resource.
After completing storage resources selection and idle slice distribution, DataNode replys to IP encoder:
If distribute new slice in same storage resources, DataNode returns under the initial address, space size, slice of newly assigned slice information such as storage resources mark etc. to IP encoder;
If distribute slice to IP encoder in new storage resources,, in replying above-mentioned slice information, the iSCSI carry information of storage resources (comprising the information such as Target, LUN, the initiator with resource write permission that storage resources is bound) is also returned to together IP encoder by DataNode.
If IP encoder is received the iSCSI carry information of new storage resources, storage resources is carried out to local iSCSI carry, generate local logical DOS drives.
Step 613, IP encoder records under the initial address, space size, slice of next slice the information such as storage resources mark in local cache, continues to record a video storage in current slice.
Step 614, IP encoder, after whole spaces of current slice are write completely, is notified Master management node by information such as the video recording time started of this slice, video recording end time, video camera marks.
Step 615, Master management node by information recordings such as the video recording time started of slice, video recording end time, video camera marks in database, thereby support VM server through database to realize the quick-searching to video recording.
Step 616, IP encoder is obtained next slice information and storage resources mark thereof from local cache, obtains the local logical DOS drives that this storage resources is corresponding, continues new video recording storage in next slice.
Also it should be noted that, DataNode in the embodiment of the present invention delay machine or suspension, while causing the video recording storage of IP encoder or Signalling exchange failure, IP encoder can distribute DataNode to the request of Master management node immediately, carry the DataNode mark of Communications failure simultaneously; Master management node will distribute new DataNode to IP encoder, ensures that the video recording storage of IP encoder is not interrupted.
Now, Master management node is triggered to the active probe mechanism of DataNode, judges whether DataNode normally works.Meanwhile, the Communications failure of accumulative total DataNode and different IP encoder counting;
In the time that DataNode surveys unsuccessfully, Master management node is set to down status, no longer distributes this DataNode to IP encoder.The follow-up DataNode of only having succeeds in registration to Master management node again, and Master management node just can be revised corresponding DataNode state for online, and again distributes this DataNode to IP encoder;
When DataNode survey successfully, but and the Communications failure counting of different IP encoder while reaching certain threshold value (such as: the network communication between DataNode and Master management node is normal, but disconnect with the network of IP encoder), Master management node is set to failure state, no longer distribute this DataNode to IP encoder, and to the Communications failure alarm of system manager's trigger equipment and IP encoder.In this case, after only having keeper's manual intervention, by administration interface, DataNode is updated to presence, Master management node just can be redistributed this DataNode to IP encoder.
In addition, increasing in embodiments of the present invention IPSAN memory device or IPSAN memory device increases in the situations such as hard disk, only need to create new storage resources to new IPSAN memory device, new hard drive space, and add the slice allocation manager of DataNode.Because new storage resources space is that all video cameras are shared and used, therefore, without any expense in the situation that, automatically support the memory space dilatation to all video cameras.
In sum, Master and the DataNode two-stage management that the embodiment of the present invention proposes, the distributed iSCSI piece that spatial elastic is distributed are directly deposited group scheme, support video camera memory space in memory device cluster any one to the distribution on multiple devices.Simultaneously, the embodiment of the present invention has realized the cutting management of slice unit, the dynamic elasticity distribution mechanism of iSCSI memory space, meet alarm interlock storage, variable code rate storage etc. and cannot estimate the scene demand of video camera memory space, reach the effect that improves to the full extent memory space utilance.
In the time of Master management node fault, system automatically switches to single memory device node memory module, ensures that the video recording storage of IP encoder does not stop.Meanwhile, the local cache mechanism by DataNode to slice video information, guarantee slice video information is not lost.
In the time of any IPSAN storage device failure, immediately distribute new memory device to IP encoder by Master management node, ensure that the video recording storage of IP encoder does not stop.Meanwhile, realize the wrong Recovery processing to DataNode different faults.
In several embodiment provided by the present invention, should be understood that disclosed method, device and electronic equipment can be realized by another way.Apparatus embodiments described above is only schematic, for example, the division of described unit, be only that a kind of logic function is divided, when actual realization, can there is other dividing mode, as: multiple unit or assembly can be in conjunction with, maybe can be integrated into another system, or some features can ignore, or do not carry out.In addition, the coupling each other of shown or discussed each part or direct-coupling or communication connection can be by some interfaces, and the indirect coupling of equipment or unit or communication connection can be electrical, machinery or other form.
The above-mentioned unit as separating component explanation can or can not be also physically to separate, and the parts that show as unit can be or can not be also physical locations, can be positioned at a place, also can be distributed in multiple network element; Can select according to the actual needs part or all of unit wherein to realize the object of the present embodiment scheme.
In addition, the each functional unit in various embodiments of the present invention can all be integrated in a processing unit, can be also that each unit is distinguished separately as a unit, also can be integrated in a unit two or more unit; Above-mentioned integrated unit both can adopt the form of hardware to realize, and the form that also can adopt hardware to add SFU software functional unit realizes.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can complete by the relevant hardware of program command, aforesaid program can be stored in a computer read/write memory medium, this program, in the time carrying out, is carried out the step that comprises said method embodiment; And aforesaid storage medium comprises: movable storage device, read-only memory (ROM, Read-Only Memory), the various media that can be program code stored such as random access memory (RAM, Random Access Memory), magnetic disc or CD.
Or, if the above-mentioned integrated unit of the embodiment of the present invention is realized and during as production marketing independently or use, also can be stored in a computer read/write memory medium using the form of software function module.Based on such understanding, the part that the technical scheme of the embodiment of the present invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product is stored in a storage medium, comprises that some instructions are in order to make a computer equipment (can be personal computer, server or the network equipment etc.) carry out all or part of of method described in the present invention each embodiment.And aforesaid storage medium comprises: various media that can be program code stored such as movable storage device, ROM, RAM, magnetic disc or CDs.
The above; be only the specific embodiment of the present invention, but protection scope of the present invention is not limited to this, any be familiar with those skilled in the art the present invention disclose technical scope in; can expect easily changing or replacing, within all should being encompassed in protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of described claim.
Claims (16)
1. the management of storage resources and a distribution method, is characterized in that, the method comprises:
Data management node DataNode receives the memory allocated space request of IP encoder;
Described DataNode is that described IP encoder is selected storage resources according to the first strategy, and distribute the memory space sheet slice in selected storage resources according to the second strategy for described IP encoder, for described IP encoder according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
2. the management of storage resources and distribution method according to claim 1, is characterized in that, described the first strategy comprises:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
3. the management of storage resources and distribution method according to claim 1, it is characterized in that, the method further comprises: described DataNode carries out cutting taking slice as unit to storage resources, and by idle slice queue with take slice queue the slice in storage resources is managed;
Described the second strategy comprises: in the storage resources of described IP encoder request, select the slice in idle slice queue to distribute to described IP encoder, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
4. the management of storage resources and distribution method according to claim 1, it is characterized in that, the method further comprises: in the time receiving the request of the new slice of distribution that described IP encoder sends, the request that described DataNode distributes new slice according to described IP encoder, whether the occupancy that judges slice in the affiliated storage resources of described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode for described IP encoder,
If the slice occupancy of all storage resources of described DataNode all reaches described the 3rd preset ratio, delete the aging data in described DataNode and slice corresponding deleted data is hung and gets back to idle slice queue, for described IP encoder is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder;
Wherein, the request of the new slice of described distribution, is when being written into ratio by described IP encoder in the space of distributed slice and reaching the second preset ratio, sends to described DataNode.
5. according to management and the distribution method of storage resources described in claim 1 to 4 any one, it is characterized in that, the method further comprises:
Before described IP encoder sends the request of memory allocated space to DataNode, described IP encoder distributes DataNode to the request of Master management node;
Described Master management node distributes described DataNode according to the 3rd strategy for described IP encoder.
6. the management of storage resources and distribution method according to claim 5, is characterized in that, described the 3rd strategy comprises:
According to the residual memory space of each DataNode, residue input and output IO disposal ability, and based on load balancing, direct network connection factor, for described IP encoder distributes DataNode.
7. the management of storage resources and distribution method according to claim 5, is characterized in that, described Master management node comprises main Master management node and auxiliary Master management node, and the method further comprises:
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
8. the management of storage resources and distribution method according to claim 5, it is characterized in that, the method further comprises: described Master management node is in the time of described DataNode fault or suspension, for the corresponding IP encoder of DataNode of described fault or suspension distributes new available DataNode.
9. the management of storage resources and a distribution system, is characterized in that, this system comprises: IP encoder and data management node DataNode, wherein,
Described IP encoder is used for, and sends the request of memory allocated space to described DataNode;
Described DataNode is used for, and is that described IP encoder is selected storage resources, and distributes the memory space sheet slice in selected storage resources according to the second strategy for described IP encoder according to the first strategy;
Described IP encoder also for, according to the iSCSI carry information of selected storage resources, by iSCSI carry storage resources, carry out iSCSI data according to the initial address of distributed memory space slice, space size and write.
10. the management of storage resources and distribution system according to claim 9, is characterized in that, described DataNode is further used for, and is that described IP encoder is selected storage resources according to following the first strategy:
If still have the slice of idle condition on the current storage resources that is distributing slice, continue to select current storage resources; And,
Distributing if current slice storage resources take the first preset ratio that slice reaches the total slice of described storage resources more than, select the storage resources of next residual memory space maximum.
11. management and the distribution systems of storage resources according to claim 9, it is characterized in that, described DataNode is further used for, and storage resources is carried out to cutting taking slice as unit, and by idle slice queue with take slice queue the slice in storage resources is managed;
Described DataNode distributes slice according to following the second strategy: in the storage resources of described IP encoder request, select the slice in idle slice queue to distribute to described IP encoder, and after replying successfully, the slice of this sub-distribution is updated to seizure condition and is dispatched to take in queue.
12. management and the distribution systems of storage resources according to claim 9, it is characterized in that, described IP encoder is further used for, and when being written into ratio and reaching the second preset ratio, distributes new slice to described DataNode request in the space of distributed slice;
Described DataNode is further used for, the request that distributes new slice according to described IP encoder, whether the occupancy that judges slice in the affiliated storage resources of described former slice reaches the 3rd preset ratio, if do not reach the 3rd preset ratio, continue as described IP encoder and distribute the idle slice in identical storage resources, if reach the 3rd preset ratio, select other storage resources of described DataNode for described IP encoder;
If the slice occupancy of all storage resources of described DataNode all reaches described the 3rd preset ratio, delete the aging data in described DataNode and slice corresponding deleted data is hung and gets back to idle slice queue, for described IP encoder is selected the storage resources of residue slice space maximum, and distribute the idle slice of selected storage resources to described IP encoder.
13. according to management and the distribution system of storage resources described in claim 9 to 12 any one, it is characterized in that, this system also comprises: Master management node;
Described IP encoder is further used for, and before sending the request of memory allocated space to described DataNode, distributes DataNode to described Master management node request;
Described Master management node is used for, and is receiving that described IP encoder distributes after the request of DataNode, distributes described DataNode according to the 3rd strategy for described IP encoder.
14. according to the management of storage resources described in claim 13 and distribution system, it is characterized in that, described Master management node is further used for, and is IP encoder distribution DataNode according to following the 3rd strategy:
According to the residual memory space of each DataNode, residue input and output IO disposal ability, and based on load balancing, direct network connection factor, for described IP encoder distributes DataNode.
15. according to the management of storage resources described in claim 13 and distribution system, it is characterized in that, described Master management node comprises main Master management node and auxiliary Master management node;
Described auxiliary Master management node is for carrying out redundancy protecting to described main Master management node, and in the time of described main Master management node fault, described auxiliary Master management node switches to main Master management node.
16. according to the management of storage resources described in claim 13 and distribution system, it is characterized in that, described Master management node is further used for, in the time of described DataNode fault or suspension, for the corresponding IP encoder of DataNode of described fault or suspension 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 true CN104023058A (en) | 2014-09-03 |
| CN104023058B 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) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN107087031A (en) * | 2017-05-10 | 2017-08-22 | 浙江宇视科技有限公司 | A kind of storage resource load-balancing method and device |
| CN107992491A (en) * | 2016-10-26 | 2018-05-04 | ***通信有限公司研究院 | A kind of method and device of distributed file system, data access and data storage |
| CN108762821A (en) * | 2017-04-18 | 2018-11-06 | 海马云（天津）信息技术有限公司 | Device and method, the electronic equipment of electronic equipment operation application |
| CN111091851A (en) * | 2019-12-10 | 2020-05-01 | 四川湖山电器股份有限公司 | IPSAN disk management method of recording and broadcasting system |
| CN113194158A (en) * | 2021-04-13 | 2021-07-30 | 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 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| US20090049229A1 (en) * | 2005-12-09 | 2009-02-19 | Matsushita Electric Industrial Co., Ltd. | Nonvolatile memory device, method of writing data,and method of reading out data |
| CN101582903A (en) * | 2009-06-24 | 2009-11-18 | 杭州华三通信技术有限公司 | Storing method of streaming media resource and device |
| CN101599978A (en) * | 2009-07-17 | 2009-12-09 | 杭州华三通信技术有限公司 | Resource allocation methods and equipment |
| CN102255921A (en) * | 2011-08-24 | 2011-11-23 | 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
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090049229A1 (en) * | 2005-12-09 | 2009-02-19 | Matsushita Electric Industrial Co., Ltd. | Nonvolatile memory device, method of writing data,and method of reading out data |
| 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 |
| CN101582903A (en) * | 2009-06-24 | 2009-11-18 | 杭州华三通信技术有限公司 | Storing method of streaming media resource and device |
| CN101599978A (en) * | 2009-07-17 | 2009-12-09 | 杭州华三通信技术有限公司 | Resource allocation methods and equipment |
| CN102255921A (en) * | 2011-08-24 | 2011-11-23 | 杭州华三通信技术有限公司 | Media stream storage method and data management server |
Cited By (10)
| 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 | ***通信有限公司研究院 | A kind of method and device of distributed file system, data access and data storage |
| CN108762821A (en) * | 2017-04-18 | 2018-11-06 | 海马云（天津）信息技术有限公司 | Device and method, the electronic equipment of electronic equipment operation application |
| 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 |
| CN107087031A (en) * | 2017-05-10 | 2017-08-22 | 浙江宇视科技有限公司 | A kind of storage resource load-balancing method and device |
| CN111091851A (en) * | 2019-12-10 | 2020-05-01 | 四川湖山电器股份有限公司 | IPSAN disk management method of recording and broadcasting system |
| CN113194158A (en) * | 2021-04-13 | 2021-07-30 | 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104023058B (en) | 2017-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104023058A (en) | Method and system for managing and allocating storage resources | |
| CN103797770B (en) | A kind of method and system of shared storage resources | |
| US10521135B2 (en) | Data system with data flush mechanism | |
| CN102594852B (en) | Data access method, node and system | |
| US11010064B2 (en) | Data system with flush views | |
| JP6492226B2 (en) | Dynamic resource allocation based on network flow control | |
| CN107092442B (en) | Storage system resource allocation method and device | |
| CN102681789B (en) | Method for generating a virtual volume (VV) in a storage system architecture | |
| CN102782670B (en) | Memory cache data center | |
| US20170177224A1 (en) | Dynamic storage transitions employing tiered range volumes | |
| US11853587B2 (en) | Data storage system with configurable durability | |
| CN106201357A (en) | The construction method of a kind of storage pool and system | |
| EP2864885A2 (en) | System and method for datacenters disaster recovery | |
| CN103765372A (en) | Configuring object storage system for input/output operations | |
| CN103748545A (en) | Data storage system and data storage control method | |
| CN103765371A (en) | Data storage system exporting logical volumes as storage objects | |
| US11262918B1 (en) | Data storage system with uneven drive wear reduction | |
| CN102868727B (en) | Method for realizing high availability of logical volume | |
| CN108512693A (en) | A kind of trans-regional disaster recovery method and device | |
| US11681443B1 (en) | Durable data storage with snapshot storage space optimization | |
| US10776173B1 (en) | Local placement of resource instances in a distributed system | |
| CN105187502A (en) | Method and system based on distributed elastic block storage | |
| CN113268472A (en) | Distributed data storage system and method | |
| CN105162869B (en) | A kind of method and apparatus for backup data management | |
| CN104793981A (en) | Online snapshot managing method and device for virtual machine cluster |
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 |