CN1540524A - Method for accessing distributed and virtualized storage in local network - Google Patents
Method for accessing distributed and virtualized storage in local network Download PDFInfo
- Publication number
- CN1540524A CN1540524A CNA2003101030449A CN200310103044A CN1540524A CN 1540524 A CN1540524 A CN 1540524A CN A2003101030449 A CNA2003101030449 A CN A2003101030449A CN 200310103044 A CN200310103044 A CN 200310103044A CN 1540524 A CN1540524 A CN 1540524A
- Authority
- CN
- China
- Prior art keywords
- module
- metadata
- kernel
- distributed
- management
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
Based on virtualized networked storage system of storing LAN, system kernel module maps operation address requested from I/O of file system to proper networked-disk so as to realize function of virtualized storage. Meanwhile, request of synchronous operation carried out for kernel metadata through Ethernet is submitted to other nodes in computer cluster system. Then, under integrating control of central control and management module in supervisor node machine, corresponding operations are carried out for metadata, operated result is added into metadata table in kernel module in order to make metadata of virtual storage in all nodes of cluster system keep consistent. Thus, the virtualized system provides a view of networked storage resources in consolidation form to realize distributed virtualized storage system.
Description
Technical field
The implementation method of distributed virtualization storage belongs to field of storage area networks in the storage area network, relates in particular to virtual technical field of memory wherein.
Background technology
Storage area network (Storage Area Networks:SAN) technology is special technology at the Computer Storage field, has high reliability (High Reliability), extendability (Extendibility) and high-performance characteristics such as (High Performance).Virtual storage system is on actual physical memory device basis, fictionalizing logic magnetic disc equipment uses for file system, it is considered to the technology of the storage environment of can streamlining management large-scale, complicated, isomery, it has storage resources control flexibly, can dynamically change functions such as storage resources and data image.
But present virtual storage system has only the unit version, can not be adapted to distributed cluster environment system.Sistina company has issued the part of LVM (Logical Volume Manager) as the (SuSE) Linux OS kernel, does not also support the virtual storage under the distributed environment at present.This realizes that the function of virtual storage has proposed difficulty for the general Network of Workstation that is of wide application, and the present invention intends addressing this problem.
The present invention has realized the virtual storage under the distributed environment, is that (Network of Workstation cluster) realizes virtual storage at the distributed system under the SAN environment.
Summary of the invention
The object of the present invention is to provide a kind of method that under the SAN environment, realizes the distributed virtualization storage system.Adopt this method can on distributed system, realize the dynamic management of unified storage resources, more effectively utilize network storage resource.The core of this method is on unit virtualization system basis, and design is the method for the virtualization system metadata between each node machine synchronously, keeps the consistance of kernel metadata and disk metadata; Simultaneously by setting management node, the virtualization operations and the behavior of unified management distributed virtual storage system offer the virtual storage of the virtual storage view realization of the unified unanimity of file system on each node machine based on unified management module, meta data block and the control and management platform of Network of Workstation.
The invention is characterized in: it is a kind of passing through by the management node machine, the distributed virtualization storage system that a distributed group of planes and optical fiber switch are formed, regulation according to network transmission protocol TCP and fiber optic network host-host protocol FC, provide the file system on each node machine a unified consistent virtual storage view by net-raid through optical networking, realize virtual storage based on unified management module, meta data block and the control and management platform of Network of Workstation.
Native system has carried out the validity test of function on the TH-MSNS network store system of computing machine high-performance research institute of Tsing-Hua University.Distributed virtual China storage system has been installed on the group of planes of 8 nodes of front end, by the network storage resource virtualization under this system's realization cluster environment, the storage and uniform of a plurality of main frames is provided, realized the virtual store pond, the function of dynamically arbitrarily distributing and revise the logic magnetic disc size of disk space promptly is provided, logic magnetic disc data image and snapshot functions are provided simultaneously, have fully verified validity of the present invention.
Description of drawings:
Fig. 1: distributed virtualization storage system principle schematic;
Fig. 2: distributed virtualization memory system architecture synoptic diagram;
Fig. 3: distributed virtualization storage system software structural drawing;
Fig. 4: the pie graph of logic magnetic disc volume set VG;
Fig. 5: distributed virtualization storage system software module frame chart;
Fig. 6: the program flow diagram of central authorities' control and administration module;
Fig. 7: the program flow chart of the method for the invention;
Fig. 8: the state transition graph of management configuration module;
Fig. 9: the inner core figure of kernel module LVM;
Figure 10: the network storage resource view that this method provides to each distributed node machine.
The hardware platform of concrete application of the present invention is the distributed type assemblies network storage environment, and the server of a plurality of group of planes is by storage networking shared network disk space.A node group of planes and management node machine all are connected on the Ethernet, carry out data communication by IP network.Concrete hardware JA(junction ambient) as shown in Figure 1.
Wherein, on a distributed group of planes and the management node machine optical channel card has been installed all, has been linked to each other with optical fiber switch, formed the storing fiber optic network by optical channel card.Net-raid provides the service of storage resources for a distributed group of planes by optical networking.Network disk realizes that by I/O node machine carry scsi disk array concrete hardware block diagram as shown in Figure 2.
The operating system of Network of Workstation is Linux, and the kernel version is more than the 2.4.18.The I/O logical place at distributed virtualization storage system place as shown in Figure 3.
As shown in Figure 4, PV (Physical Volume) is the unit of the bottom among the Linux LVM, and it is single equipment or subregion.Each PV comprises VGDA (Volume Group Descriptor Area), and it comprises some special information that are used for the LVM configuration.PV can both add among the VG (Volume Group) that has existed at any time.A VG can regard a very big storage pool as, and it comprises one or more LV (Logical Volume), and each LV uses for file system as actual logical device.Figure has represented PV, the relation between VG and the LV.The core concept of LVM system is based on the physical disk resource and fictionalizes logic magnetic disc, uses for file system.The operation address of the I/O request command that LVM sends according to the suitable revised file of configuring condition system is carried out this operation with it after logical address space is mapped to physical address space.
Below in conjunction with Fig. 5, method of the present invention is done detailed description.
The user can send the configuration operation of virtual storage to the distributed virtualization storage system by operation interface, User login is controlled the operation of virtual storage system by corresponding operational order behind the management node machine.
The operation that the user can carry out comprises:
● LV Object Operations (creating and deletion increase capacity, minimizing capacity);
● PV Object Operations (information material of inquiry PV);
● the operation of VG (details of inquiry).
Central authorities' control and administration module produce the operational order field according to the user's data input, and its organization definition is as follows:
typedef struct lvm_msg_tcp { int cmd_type；//COMMAND type int cmd_index； //COMMAND index int send_host_id； int receive_host_id； unsigned int length； int ack； void *data； //buffer pointer }_attribute_((packed))lvm_msg_tcp_t；
Wherein, cmd_type is used for representing operation types, and cmd_index is used for representing the instruction operated, and other domain of instruction is some additional order parameter information.
Central authorities' control and administration module are being controlled all storage resources on the network.It carries out unified numbering to the node machine in a network storage resource and the group of planes, san1 for example, san2, host1, host2 or the like, all these is the elementary cell that this module need be managed, central authorities control and administration module when starting by reading the configuration file mode and having obtained these information by inquiring about (transmission INQUIRY orders to network disk).The clustered node configuration information is stored in/etc/clvm/node_table, control of each central authorities and administration module read in node configuration information when starting, and by udp broadcast to collect all child node status informations, finally generate informational table of nodes, shown in it is defined as follows:
typedef struct node_table_entry //construct cluster_geometry{ int status; // node state struct in_addr addr; // node i net address } _ attribute_ ((packed)) node_table_entry_t; Typedef struct{ int node_nr; // interstitial content node_table_entry_t table[MAX_NODE_NR]; } _ attribute_ ((packed)) node_table_t; Extern node_table_t node_table;
For central authorities control and administration module, the flow process that obtains a typical bookkeeping after the operational order field as described in Figure 6:
(1) inquiry of information
The operation requests that a kind of user sends is the inquiry of some information, LV for example, and the inquiry of VG information etc., this part operation central authorities control and administration module be the information table of inquiry oneself the maintenance object information that just can obtain operating directly.The information resources table that comprises in central management and control module has:
The cluster nodes information table, host information table just, a simple example is as follows;
| Node ID | Node IP | Nodename |
| 1 | ?192.168.0.1 | Service node 1 |
| 2 | ?192.168.0.2 | |
| 3 | ?192.168.0.3 | Service node 3 |
The LV information list, logical volume information table just, a simple example is as follows:
| The LV title | Affiliated VG | Affiliated node | The network disk that comprises |
| ????LV1 | ??VG1 | ????2 | Disk 1, disk 4, disk 5 |
| ????LV2 | ??VG2 | ????3 | Disk 6, disk 7 |
| ????LV3 | ??VG3 | ????2 | |
The VG information list, a simple example is as follows:
| The VG title | The network disk that comprises | |
| ?VG1 | Disk 1, disk 4, disk 5 | |
| ?VG2 | Disk 6, disk 7 | |
| ? | Disk | 2, disk 3 |
If user's operation is information inquiry, return results information was given the user after central authorities' control and administration module were directly inquired about above-mentioned data so.
(2) functional configuration, for example the capacity of logic magnetic disc (LV) increases and reduces
In the distributed virtualization storage system, the LVM module in the kernel is finished the virtual functions of physical volume; Synchronization module synchronizing metadata between each node machine makes each node organizational security hold consistent view; Control module is being controlled whole system operation, and they are three most crucial parts of total system.The function of each module among the figure is respectively in table 1:
Table 1 distributed virtualization storage system software module list
| Title | Function |
| The cLVM kernel module | Mapping conversion I/O address function, the virtual store function of realization core |
| Management and configuration module | Kernel module and synchronization module are carried out some management and configuration operation |
| Synchronization module | Storing metadata between each node of a group of planes keeps consistent with network disk synchronously |
| The central authorities' control and the administration module of management node | The work of whole distributed virtualization storage system is coordinated in control, and the user is provided interface operable |
As shown in Figure 9, core functions is being safeguarded a VGDA (virtual metadata) tabulation vg, creates VG at every turn and all will put into vg to the pairing VGDA of VG, then corresponding VGDA is deleted from table during deletion VG.
Be a structure vg_t in kernel source code in the VGDA structure, wherein two territory pv and lv type are respectively the array of pointers of pv_t and lv_t.This two numbers group is deposited the pointer of direction structure pv_t and lv_t respectively, indicates the PV and the LV that constitute this VG.
Lv_cur among the Vg_t and pv_cur then illustrate current LV number and PV number.
The lv_stripes territory of Lv_t illustrates the method for this LV mapping, if lv_stripes＜2 explanations are Linear mappings, if lv_stripes＞=2 then explanation LV are the striped mappings and are mapped on lv_stripes the PV.Lv_stripesize then illustrates the size of each stripe, and unit is sector.Lv_current_pe among the Lv_t then is the array of a pe_t pointer, and the pe that current this LV was mapped to has been described.
The pe territory of Pv_t is the array of pe_t pointer, indicates the pe of this PV.Pv_on_disk, vg_on_disk and lv_on_disk should be the metadata that leaves on this PV.
The core functions module is intercepted and captured all magnetic disc i/os operations of (SuSE) Linux OS, and the logical space address by the I/O operation is to the realization virtualization that is converted of actual physical address.
The interface that kernel module offers central authorities control and administration module is the IOCTL power function, different operating parameter correspondences different operational orders, shown in the face that the is described as follows table 2 of these interface functions:
| User's operation | The IOCTL order | Describe |
| Vgcreate | VG_CREATE | Create a new Volume Group. |
| Vgremove | VG_REMOVE | Delete an already present Volume Group. |
| Vgrename | VG_RENAME | To the already present Volume Group operation of renaming. |
| Vgextend | VG_EXTEND | An already present Volume Group is expanded. |
| Vgreduce | ?VG_REDUCE | Reduce?a?volume?group。 |
| Vgmerge | VG_MERGE | Merge two Volume Group. |
| Vgsplit | ?VG_SPLIT | Split?a?volume?group。 |
| Lvcreate | LV_CREATE | Create a Logical Volume. |
| Lvremove | LV_REMOVE | Delete an already present Logical Volume. |
| Lvrename | LV_RENAME | A LV is renamed |
| Lvresize | LV_RESIZE | It is little to change a LV |
| Lvextend | LV_EXTEND | Increase the size of a LV |
| Lvreduce | LV_REDUCE | Reduce the size of a LV |
Whether successful value is given central authorities' control and administration module to return an expression after the IOCTL interface function is called.Central authorities' control and administration module are determined the flow process of operation according to this value.
If operation failure then directly returns to the user with failure result, if operate successfully, then the invoke synchronous module is carried out the metadata synchronization operation of distributed cluster nodes, and the result of this operation is returned to the user.For the embodiment of the calling interface of synchronization module and concrete metadata, the back will be described.
When the node machine in the group of planes carries out virtualization operations (for example carrying out dilatation operation at some logical volumes that has existed) to network storage resource, need carry out the kernel data of other node machines synchronously to keep them consistent with content on the network disk.The major function of synchronization module is carried out the kernel metadata exactly synchronously.
Kernel metadata synchronization operation between the node machine comprises VG metadata synchronization and LV metadata synchronization, and the work of synchronous operation becomes two parts:
(1) modification of user's space data comprises metadata, the change metadata backup file that changes memory disk and changes corresponding equipment file etc.;
(2) object information in the distributed virtualization storage system core functions module is promptly upgraded in the renewal of kernel spacing information.
And specifically need synchronous metadata to be the VGDA information module, synchronous function is exactly the information of reading VGDA from the internal memory of management node, send to then in other the internal memory of distributed cluster nodes machine, make the internal storage data of other node machine and the internal storage data of management node keep consistency.
Distributed virtualization system synchronization module externally provides main interface and power function to have:
void?sync_init_request(int?argc，char
**argv，Cluster_Message
*msg)；
void?sync_do_request(Cluster_Message
*msg)；
Wherein the Cluster_Message structure is the data format protocol of Network Transmission.When a node of Network of Workstation has carried out behind the virtualization operations, the management of system and configuration module are with the interface function sync_init_request (char of invoke synchronous module
*Argv, Cluster_Message
*Msg) fill msg structure and be broadcast to each server node and carry out synchronous operation.The primary structure of ClusterMessage is defined as follows:
typedef?struct?MSG_cLVM{
int?cmd_type；
int?cmd_index；
int?ack；
int?send_host_id；
int?receive_host_id；
unsigned?int?length；
void
*data；
}Cluster_Message；
Cmd_type among the Cluster_Message is used to distinguish type of message, and for synchronous operation message, this value is set to LVM_SYNC_TYPE.And cmd_index be distributed the virtual village of formula just the system synchronization module which kind of operation is used for distinguishing be, the IOCTL that its value is set in the table accordingly orders.The Data territory is the sync_info_t structured fingers, and the relevant information of LVM action need is provided.
The structure of Sync_info_t is as follows:
typedef?struct
{
char
*command；
int?num；
char
*info[128]；
}syn_info_t；
The name that promptly operate in the command territory of this structure, num is the item number in the info territory, info is some information relevant with this operation.
After the synchronous group of planes server node of quilt is received and is required synchronous Message, the interface function sync_do_request (Cluster_Message that synchronization module provides
*Msg) will be called.Sync_do_request upgrades according to the kernel data of the information among the msg to the distributed virtualization storage system, and the updating steps flow process is as follows:
(1) Sync_do_request at first distinguishes the type of synchronous operation, request be VG synchronously or LV synchronous, the former then calls vg_sync (Cluster_Message
*Msg) handle, the latter then calls lv_sync (Cluster_Message
*Msg) carry out the synchronous operation of kernel data;
(2) Vg_sync and lv_sync at first come to obtain pointing to according to vg_name the vgda file pointer vg (vg_cfgrestore function) of Volume Group from biography.If LV is synchronous, then also need from vg, obtain pointing to the pointer lv (lv_get_index_by_name function) of LogicalVolume according to lv_name;
(3) after obtaining vg and lv, to vg-〉pv[p]-pv_dev and lv-〉lv_current_pe[le] .dev upgrades, the process of upgrading is finished (sync_update_dev function) by searching the DEV table, sync_update_dev from Agent Agent, obtain equipment after the Major on the book server node and Minor number to dev assignment again.
(4) operation of kernel data synchronously.Finish by send the IOCTL order to kernel; Utilize the relevant function that the Lvmlib storehouse provides,, can be sent to kernel module to the vg or the lv that are correlated with so that data are upgraded to LVM kernel transmit operation order by calling these functions as vg_create etc.
(5) finish renewal to kernel data after, CREATE, operations such as REMOVE and RENAME also need make some corresponding modification to system file, and are as shown in table 4:
Table 5-4 cLVM needs retouching operation
| The IOCTL order | The modification that needs |
| VG_CREATE | Under/dev/, create a VG_NAME catalogue, and create the character device file of one/dev/VG_NAME/group.And in/etv/lvmtab, increase Volume Group inlet, the vgda backup file of establishment/dev/lvmtab.d/VG_NAME. |
| VG_REMOVE | Deletion/dev/VG_NAME/ catalogue, the inlet of this Volume Group among the deletion/dev/lvmtab.Deletion/dev/lvmtab.d/VG_NAME. |
| VG_RENAME | Be VG_REMOVE earlier, VG_CREATE is to the modification of system file then. |
| VG_MERGE | Similar VG_RENAME. |
| VG_SPLIT | Similar VG_RENAME. |
| LV_CREATE | Create a new block device file/dev/VG_NAME/LV_NAME. |
| LV_REMOVE | Deleted block device file/dev/VG_NAME/LV_NAME |
| LV_RENAME | Block device file/dev/VG_NAME/LV_NAME renames as/dev/VG_NAME/LV_NEW_NAME |
(6) last, it is ack_Message that sync_do_request fills msg, turns back to initial cluster nodes, notifies its synchronous operation successful information.
The function ratio of management and configuration module is more single, only kernel module and synchronization module is carried out some simple management and configuration operations, and the data communication facility between each node machine is provided.The management of cluster nodes machine and the state transition graph of configuration module be as shown in Figure 8:
Management and configuration module provide two interface functions to be used for data transmission to central authorities' control and administration module and synchronization module, and they are:
Send(Host，Data)；
Recv(Host，Data)；
By these two communication functions, be between the synchronization module of different machines and carry out data communication.
Native system has carried out the validity test of function on the TH-MSNS network store system of department of computer science, Tsinghua university high-performance research institute.The distributed virtualization storage system has been installed on the group of planes of 8 nodes of front end, realized the virtualized function of the network storage resource under the cluster environment by this system, the storage and uniform of a plurality of main frames is provided, realized the virtual store pond, the function of dynamically arbitrarily distributing and revise the logic magnetic disc size of disk space promptly is provided, logic magnetic disc data image and snapshot functions are provided simultaneously, have fully verified validity of the present invention.
Claims (1)
1. the method for distributed virtualization storage in the storage area network, it is characterized in that: based on by a distributed node group of planes and an any supervisor of appointment wherein, network disk and optical fiber switch and press the distributed type assemblies network store system that network store system (SAN) is formed, supervisor console offers the user Telnet service of landing and carries out the management function of distributed virtualization storage system, and central authorities' control in the management node machine and administration module are submitted to functional or to the logic magnetic disc LV of the logical device file system use of the conduct reality in the distributed group of planes, kernel metadata in the logic magnetic disc volume set VG is carried out corresponding operation requests.The core of this method is on unit virtualization system basis, and design is the method for the virtualization system metadata between each node machine synchronously, keeps the consistance of kernel metadata and disk metadata; Simultaneously by setting management node, the virtualization operations and the behavior of unified management distributed virtual storage system offer the virtual storage of the virtual storage view realization of the unified unanimity of file system on each node machine based on unified management module, meta data block and the control and management platform of Network of Workstation.The above method contains following steps successively:
(1) modular structure
The distributed virtualization system management configuration module of central control and administration module, abbreviation management configuration module, the metadata synchronization module and the kernel module of abbreviation synchronization module are set in the management node machine; In each distributed node machine, be provided with respectively by above-mentioned synchronization module, module of the same name that management configuration module is identical with the kernel module content; Wherein
(1.1) central authorities' control and administration module:
All storage resources on the Control Network: this module by reading configuration file and the requester network disk obtains node configuration information, generates the informational table of nodes of implicit nodes serial number information, LV information and VG information when starting, also claim the resource information table;
For carrying out following operation, the user provides interface:
Increase and minimizing to the establishment of the operation of logic magnetic disc LV, block file LV or deletion, LV capacity;
Operation to logic magnetic disc volume set VG: Query Information;
To network disk is the operation of physical disk PV: the metadata VGDA corresponding to the logic magnetic disc volume in the inquiry PV;
Two interfaces that send and receive are provided to management configuration module;
Be that the control kernel module carries out functional performance, provide IOCTL function interface with kernel module to kernel module; The different operating parameter value that this interface sends is corresponding to the different operating that will carry out VG, LV.
(1.2) kernel module
After intercepting and capturing all operations that central authorities control and administration module sends, the actual physical address of the logical space map addresses of the I/O of the file system that transmits operation, to realize virtual storage to network disk to logic magnetic disc LV, logic magnetic disc volume set VG;
(1.3) metadata synchronization module
When its distributed node machine is made virtual storage operation to network storage resource, be consistent to the kernel module metadata of each node of synchronization module and network disk metadata synchronization;
Synchronization module is provided for each distributed node machine kernel module unit number interface function synchronously to management configuration module, is designated as sync_init_request;
Synchronization module is provided for when certain distributed node machine carries out virtual storage to network storage resource to kernel module, provides to carry out wheel synchronization type operational order ground interface function, is designated as sync_do_request;
Described synchronous operation comprises VG data sync and LV data sync, and the data of synchronous operation are divided into two parts: the renewal of the modification of user's space address date and kernel module metadata;
(1.4) distributed virtual China storage administration configuration module
It is provided with transmission interface function S end (Host respectively, Data), (Host Data), offers central control and administration module, the metadata synchronization module usefulness of communicating by letter with kernel module to accept interface function Recv, wherein Host is a node ID, is used to indicate node;
(2) carry out virtual storage successively according to the following steps:
(2.1) start central authorities' control and administration module, the TCP watcher thread is created in initialization.
(2.2) user lands Service Management by telnet, sends operation requests to central authorities' control and administration module
(2.3) after the TCP thread is received the users broadcasting request,, and call the Tcp_request_dispatch function for each user creates a TCP processing threads
(2.4) behind the Tcp_request_dispatch function decision operation command type, classification is handled
(2.5) if information inquiring command is then inquired about the informational table of nodes of these central authorities control and administration module, Query Result is returned the user through Ethernet
(2.6) if the functional performance order, then directly call the IOCTL interface function of kernel module by management configuration module, carry out the operation that central authorities' control and administration module are formulated, operating result is returned management configuration module, return the user through central authorities' control and administration module again; Simultaneously, by optical fiber switch the vg of amended VGDA table is delivered to each distributed node machine, correspondingly, the logical address of VG, LV is mapped to network disk, and revise the network disk metadata by the vg table of amended VGDA through management configuration module;
(2.7) if the wheel synchronization type operational order then by the interface function of management configuration module invoke synchronous module, is carried out following operation:
(2.7.1) earlier by interface function sync_init_rdquest difference type of message, i.e. still LV is synchronous synchronously for the type of synchronous operation: VG, calls different functions more respectively and handles;
(2.7.2) when handling, from the file that biography is come, be that the VG name reaches the pointer that points to VG earlier according to vg_name, if LV is synchronous, be the LV name obtains pointing to LV from the VG table pointer then according to lv_name, then, respectively according to pointer position new logical addresses more;
(2.7.3) synchronous kernel data:, finish renewal to the VGDA metadata by send the IOCTL order to kernel module;
(2.7.4) operation that system file is created respectively, deletes, revised or rename by VG and LV;
(2.7.5) read the VGDA metadata from kernel module, sync_init_request sends to management configuration module by interface function, synchronously keeps the consistance of distributed virtualization storage system kernel metadata and network disk metadata.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101030449A CN1254748C (en) | 2003-10-31 | 2003-10-31 | Method for accessing distributed and virtualized storage in local network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101030449A CN1254748C (en) | 2003-10-31 | 2003-10-31 | Method for accessing distributed and virtualized storage in local network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1540524A true CN1540524A (en) | 2004-10-27 |
| CN1254748C CN1254748C (en) | 2006-05-03 |
Family
ID=34333198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2003101030449A Expired - Fee Related CN1254748C (en) | 2003-10-31 | 2003-10-31 | Method for accessing distributed and virtualized storage in local network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1254748C (en) |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006089479A1 (en) * | 2005-02-23 | 2006-08-31 | Zhang, Jinkui | A data managing method in a network storage system and the network storage system based on the method |
| CN1299207C (en) * | 2005-01-21 | 2007-02-07 | 清华大学 | Large scale resource memory managing method based on network under SAN environment |
| CN1304961C (en) * | 2005-03-11 | 2007-03-14 | 清华大学 | Memory virtualized management method based on metadata server |
| CN1322427C (en) * | 2005-02-25 | 2007-06-20 | 清华大学 | Universal method for dynamical management of storage resource under Windows platform |
| CN100347692C (en) * | 2005-05-31 | 2007-11-07 | 清华大学 | Implementing method of virtual intelligent controller in SAN system |
| CN100354833C (en) * | 2005-01-14 | 2007-12-12 | 英业达股份有限公司 | Protection method for SAN system in abnormal condition |
| CN100356308C (en) * | 2005-11-21 | 2007-12-19 | 余波 | Method for realizing network virtual magnetic disk |
| CN100405777C (en) * | 2006-07-27 | 2008-07-23 | 清华大学 | Cashe method based on target device internal storage in ether net storage regional network |
| CN100423491C (en) * | 2006-03-08 | 2008-10-01 | 杭州华三通信技术有限公司 | Virtual network storing system and network storing equipment thereof |
| CN100464535C (en) * | 2006-05-09 | 2009-02-25 | 国家数字交换系统工程技术研究中心 | Dynamic loading and control method for router virtual drive module |
| CN1794207B (en) * | 2004-12-20 | 2010-05-26 | 国际商业机器公司 | Method and system for implenmenting cache consistency |
| CN101232422B (en) * | 2008-01-18 | 2010-06-02 | 北京交通大学 | Network storage system based on graticule technique |
| WO2010078768A1 (en) * | 2008-12-29 | 2010-07-15 | 成都市华为赛门铁克科技有限公司 | Virtual storage method and device |
| CN101894066A (en) * | 2010-04-28 | 2010-11-24 | 北京同有飞骥科技有限公司 | Disk array virtualization-based network storage management software testing method |
| CN102036419A (en) * | 2009-09-30 | 2011-04-27 | 华为技术有限公司 | Network node configuration information processing method, network node and communication system |
| US8296476B2 (en) | 2008-12-24 | 2012-10-23 | Chengdu Huawei Symantec Technologies Co., Ltd. | Storage method, storage system, and controller |
| CN102868754A (en) * | 2012-09-26 | 2013-01-09 | 北京联创信安科技有限公司 | High-availability method, node device and system for achieving cluster storage |
| CN102970204A (en) * | 2012-10-24 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Distributed switch system based on xen virtualization platform and achieving method thereof |
| CN103118073A (en) * | 2013-01-08 | 2013-05-22 | 华中科技大学 | Virtual machine data persistence storage system and method in cloud environment |
| CN103268252A (en) * | 2013-05-12 | 2013-08-28 | 南京载玄信息科技有限公司 | Virtualization platform system based on distributed storage and achieving method thereof |
| CN103392166A (en) * | 2011-04-27 | 2013-11-13 | 株式会社日立制作所 | Information storage system and storage system management method |
| CN104219318A (en) * | 2014-09-15 | 2014-12-17 | 北京联创信安科技有限公司 | Distributed file storage system and method thereof |
| CN106648474A (en) * | 2017-01-01 | 2017-05-10 | 国云科技股份有限公司 | Virtual machine disk recovery method based on logical volume |
| CN106713465A (en) * | 2016-12-27 | 2017-05-24 | 北京锐安科技有限公司 | Distributed storage system |
| CN107104911A (en) * | 2017-04-06 | 2017-08-29 | 清华大学 | The dividing method and sending method of UDP message bag |
| CN107608857A (en) * | 2017-09-25 | 2018-01-19 | 郑州云海信息技术有限公司 | A kind of SAN storage health status inspection method, device and readable storage medium storing program for executing |
| CN107888638A (en) * | 2016-09-30 | 2018-04-06 | 中移(苏州)软件技术有限公司 | A kind of distributed storage method and device of the Internet storage Local Area Network |
| CN109857335A (en) * | 2017-11-30 | 2019-06-07 | 株式会社日立制作所 | System and its control method and storage medium |
| WO2021088587A1 (en) * | 2019-11-08 | 2021-05-14 | 华为技术有限公司 | Data access method and apparatus, and storage medium |
| CN114089920A (en) * | 2021-11-25 | 2022-02-25 | 北京字节跳动网络技术有限公司 | Data storage method and device, readable medium and electronic equipment |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101277317B (en) * | 2008-04-29 | 2010-10-13 | 北京锐安科技有限公司 | Method for equalization and distribution of distributed processing system data |
-
2003
- 2003-10-31 CN CNB2003101030449A patent/CN1254748C/en not_active Expired - Fee Related
Cited By (38)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1794207B (en) * | 2004-12-20 | 2010-05-26 | 国际商业机器公司 | Method and system for implenmenting cache consistency |
| CN100354833C (en) * | 2005-01-14 | 2007-12-12 | 英业达股份有限公司 | Protection method for SAN system in abnormal condition |
| CN1299207C (en) * | 2005-01-21 | 2007-02-07 | 清华大学 | Large scale resource memory managing method based on network under SAN environment |
| WO2006089479A1 (en) * | 2005-02-23 | 2006-08-31 | Zhang, Jinkui | A data managing method in a network storage system and the network storage system based on the method |
| US8145873B2 (en) | 2005-02-23 | 2012-03-27 | Yaolong Zhu | Data management method for network storage system and the network storage system built thereof |
| CN1322427C (en) * | 2005-02-25 | 2007-06-20 | 清华大学 | Universal method for dynamical management of storage resource under Windows platform |
| CN1304961C (en) * | 2005-03-11 | 2007-03-14 | 清华大学 | Memory virtualized management method based on metadata server |
| CN100347692C (en) * | 2005-05-31 | 2007-11-07 | 清华大学 | Implementing method of virtual intelligent controller in SAN system |
| CN100356308C (en) * | 2005-11-21 | 2007-12-19 | 余波 | Method for realizing network virtual magnetic disk |
| CN100423491C (en) * | 2006-03-08 | 2008-10-01 | 杭州华三通信技术有限公司 | Virtual network storing system and network storing equipment thereof |
| US8612561B2 (en) | 2006-03-08 | 2013-12-17 | Hangzhou H3C Technologies Co., Ltd. | Virtual network storage system, network storage device and virtual method |
| CN100464535C (en) * | 2006-05-09 | 2009-02-25 | 国家数字交换系统工程技术研究中心 | Dynamic loading and control method for router virtual drive module |
| CN100405777C (en) * | 2006-07-27 | 2008-07-23 | 清华大学 | Cashe method based on target device internal storage in ether net storage regional network |
| CN101232422B (en) * | 2008-01-18 | 2010-06-02 | 北京交通大学 | Network storage system based on graticule technique |
| US8296476B2 (en) | 2008-12-24 | 2012-10-23 | Chengdu Huawei Symantec Technologies Co., Ltd. | Storage method, storage system, and controller |
| WO2010078768A1 (en) * | 2008-12-29 | 2010-07-15 | 成都市华为赛门铁克科技有限公司 | Virtual storage method and device |
| CN102036419B (en) * | 2009-09-30 | 2013-04-24 | 华为技术有限公司 | Network node configuration information processing method, network node and communication system |
| CN102036419A (en) * | 2009-09-30 | 2011-04-27 | 华为技术有限公司 | Network node configuration information processing method, network node and communication system |
| CN101894066A (en) * | 2010-04-28 | 2010-11-24 | 北京同有飞骥科技有限公司 | Disk array virtualization-based network storage management software testing method |
| CN101894066B (en) * | 2010-04-28 | 2012-11-28 | 北京同有飞骥科技股份有限公司 | Disk array virtualization-based network storage management software testing method |
| CN103392166A (en) * | 2011-04-27 | 2013-11-13 | 株式会社日立制作所 | Information storage system and storage system management method |
| CN102868754A (en) * | 2012-09-26 | 2013-01-09 | 北京联创信安科技有限公司 | High-availability method, node device and system for achieving cluster storage |
| CN102970204A (en) * | 2012-10-24 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Distributed switch system based on xen virtualization platform and achieving method thereof |
| CN102970204B (en) * | 2012-10-24 | 2017-09-01 | 曙光信息产业(北京)有限公司 | A kind of distribution switch system and its implementation based on xen virtual platforms |
| CN103118073B (en) * | 2013-01-08 | 2015-07-22 | 华中科技大学 | Virtual machine data persistence storage system and method in cloud environment |
| CN103118073A (en) * | 2013-01-08 | 2013-05-22 | 华中科技大学 | Virtual machine data persistence storage system and method in cloud environment |
| CN103268252A (en) * | 2013-05-12 | 2013-08-28 | 南京载玄信息科技有限公司 | Virtualization platform system based on distributed storage and achieving method thereof |
| CN104219318A (en) * | 2014-09-15 | 2014-12-17 | 北京联创信安科技有限公司 | Distributed file storage system and method thereof |
| CN104219318B (en) * | 2014-09-15 | 2018-02-13 | 北京联创信安科技股份有限公司 | A kind of distributed file storage system and method |
| CN107888638A (en) * | 2016-09-30 | 2018-04-06 | 中移(苏州)软件技术有限公司 | A kind of distributed storage method and device of the Internet storage Local Area Network |
| CN106713465A (en) * | 2016-12-27 | 2017-05-24 | 北京锐安科技有限公司 | Distributed storage system |
| CN106648474A (en) * | 2017-01-01 | 2017-05-10 | 国云科技股份有限公司 | Virtual machine disk recovery method based on logical volume |
| CN107104911A (en) * | 2017-04-06 | 2017-08-29 | 清华大学 | The dividing method and sending method of UDP message bag |
| CN107608857A (en) * | 2017-09-25 | 2018-01-19 | 郑州云海信息技术有限公司 | A kind of SAN storage health status inspection method, device and readable storage medium storing program for executing |
| CN109857335A (en) * | 2017-11-30 | 2019-06-07 | 株式会社日立制作所 | System and its control method and storage medium |
| CN109857335B (en) * | 2017-11-30 | 2022-04-19 | 株式会社日立制作所 | System, control method thereof, and storage medium |
| WO2021088587A1 (en) * | 2019-11-08 | 2021-05-14 | 华为技术有限公司 | Data access method and apparatus, and storage medium |
| CN114089920A (en) * | 2021-11-25 | 2022-02-25 | 北京字节跳动网络技术有限公司 | Data storage method and device, readable medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1254748C (en) | 2006-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1540524A (en) | Method for accessing distributed and virtualized storage in local network | |
| JP7053682B2 (en) | Database tenant migration system and method | |
| EP3210103B1 (en) | Disk partition stitching and rebalancing using a partition table | |
| US9405487B2 (en) | Media aware distributed data layout | |
| US7676628B1 (en) | Methods, systems, and computer program products for providing access to shared storage by computing grids and clusters with large numbers of nodes | |
| JP5393515B2 (en) | Server image migration | |
| US9563469B2 (en) | System and method for storage and deployment of virtual machines in a virtual server environment | |
| US8504670B2 (en) | Virtualized data storage applications and optimizations | |
| US9348842B2 (en) | Virtualized data storage system optimizations | |
| WO2013004136A1 (en) | Distributed storage method, device and system | |
| DE102012218269B4 (en) | Interface for managing data movement in a storage system with thin provisioning | |
| CN1955912A (en) | Method for distributing resource in large scale storage system | |
| CN1726454A (en) | A distributed network attached storage system | |
| CN105027122A (en) | Compression and deduplication layered driver | |
| WO2019226365A1 (en) | Scalable multi-tier storage structures and techniques for accessing entries therein | |
| US10936616B2 (en) | Storage-side scanning on non-natively formatted data | |
| CN1848851A (en) | Computer and method for establishing light weight catalog access data protocol exchange format fragment | |
| US8819088B2 (en) | Implementing storage management functions using a data store system | |
| US20160070495A1 (en) | Logical replication mapping for asymmetric compression | |
| WO2022026974A1 (en) | Translation techniques for distributed datasets | |
| Goswami et al. | Graphmap: Scalable iterative graph processing using nosql | |
| US20220342888A1 (en) | Object tagging | |
| US20200012456A1 (en) | Stale block resynchronization in nvm based systems | |
| EP3867753B1 (en) | Intelligent pooling of isolated hierarchical runtimes for cloud scale databases in multi-tenant environment | |
| US11593310B2 (en) | Providing writable streams for external data sources |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20060503 Termination date: 20111031 |