CN1304961C - Memory virtualized management method based on metadata server - Google Patents
Memory virtualized management method based on metadata server Download PDFInfo
- Publication number
- CN1304961C CN1304961C CNB2005100114094A CN200510011409A CN1304961C CN 1304961 C CN1304961 C CN 1304961C CN B2005100114094 A CNB2005100114094 A CN B2005100114094A CN 200510011409 A CN200510011409 A CN 200510011409A CN 1304961 C CN1304961 C CN 1304961C
- Authority
- CN
- China
- Prior art keywords
- virtual
- meta data
- metadata
- data server
- virtual volume
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The present invention relates to a storage virtualization management method based on a meta data server, which belongs to the field of storage area network virtualization. The present invention is characterized in that virtualization agent modules arranged on front end host computers achieve the conversion from virtual logical volumes to actual physical volumes; one or a plurality of meta data servers utilize different storage devices to form the physical volumes by using global unique identifiers, and thus, a storage pool is constructed; meta data which reflects the parameters of the storage devices is used to form the virtual volumes; management modules are used to operate the meta data; the communication contact between the virtualization agent modules and the management module is built through the Ethernet so as to provide required meta data information for the virtualization agent modules; the virtualization agent modules combined the previously received global unique identifier information of the required physical volumes to construct corresponding virtual devices in a kernel. The present invention can complete various virtualization functions and has little influence on the throughput rate and the response time of a management system.
Description
Technical field
Memory virtualized management method based on meta data server belongs to the storage administration field, relates in particular to storage area network (storage area network) virtual management field wherein.
Background technology
Storage Virtualization is meant to be separated concrete memory device with server OS, a kind of technology of unified storage pool and virtual volume is provided for the storage user.Storage Virtualization management in the storage area network is a gordian technique giving full play to storage area network potential.Traditional Host Based virtualized management method adopts symmetrical structure, any one node in the group of planes can carry out the operation of virtual management, lean on internodal communication to safeguard the consistance of metadata, the extensibility that this has increased the burden of main frame and has limited system.In addition, owing to adopted symmetrical structure, this method not to support the front end main frame to have different operating system isomery storage area network environment.
The present invention proposes a kind of new memory virtualized management method, effectively solved the problems referred to above.
Summary of the invention
The object of the present invention is to provide a kind of virtualized method of storage resources that is applicable to the storage area network logical volume transfer method and storage network system of kinds of platform, realize physically disperseing storage resources shared and unified management in logic, the forward end main frame provides virtual volume (LV).Design and the method for expressing of virtual metadata and the communication pattern between meta data server and the virtual agency that focuses on meta data server of the present invention.
The invention is characterized in: it is to realize that this system contains in the virtual management system that is made of following equipment:
Various types of front end main frames, i.e. application server, the virtual proxy module of this front end main frame are realized the virtual conversion of actual physical dish and the metadata that reads the respective virtual volume from following meta data server of being rolled onto;
Meta data server, one or more is arranged, link to each other with above-mentioned each front end main frame through Ethernet according to ICP/IP protocol, the storage resources unification of different company, different medium in the storage area network is built into physical volume, storage pool and virtual volume, and, realize the management and the configuration of storage resources for above-mentioned each front end main frame provides virtual volume;
Memory device, the in store metadata of head, each memory device has a unique general identifications, represents with Globally Unique Identifier;
Optical fiber switch is HBA by the fiber buss adapter, links to each other with meta data server with above-mentioned each front end main frame, also link to each other with above-mentioned memory device simultaneously by optical fiber interface,
Described memory virtualized management method based on meta data server contains following steps successively:
Step 1: structure software module
Virtual agency is virtual proxy module, it is a layer between file system/database layer and the bottom layer driving on each front end main frame, it also is kernel module, be responsible for the mapping of logical address to physical address, communicate by letter with the form of control stream by the communication interface of configuration and the communication interface of above-mentioned meta data server, obtain the configuration information of virtual volume and virtual volume is presented to topmost paper system/database from above-mentioned meta data server; When file system/database sent the I/O of virtual volume operation, this virtual proxy module intercepted this I/O operation, and logical address is mapped to physical address, form data stream after, again the I/O operation is sent to the bottom layer driving device,
The software of meta data server operates on the linux system, is made up of one group of user space program, comprises with lower module:
Management interface, the interface of carrying out administration order is provided to the keeper, wherein contain and create storage pool, virtual volume, authorize of the order of a front end main frame to the rights of using of virtual volume, this management interface also provides the required virtual volume and the configuration information of storage pool to the keeper in addition, and this management interface is to realize by the API that following metadata management module provides;
Host interface, it be responsible for above-mentioned each front end main frame on communication interface communicate by letter according to ICP/IP protocol, when the front end host-initiated requires to be written into the metadata of virtual volume in the meta data server, perhaps when the configuration of virtual store changes, meta data server is notified virtual proxy module the information of metadata by communication interface, virtual proxy module also by host interface to following monitoring module report oneself state with to the operating position of virtual volume;
The metadata management module, execution is to the operation task of metadata, a plurality of physical volumes are organized into storage pool, remaining space on the managing physical volume, generate the metadata of virtual volume according to the mapping mode of virtual volume, again the information of metadata is recorded on the physical volume, this metadata management module also has one group of API for other module invokes, described metadata is used the section tissue, contain: logic initial sector, number of sectors, target phase type and target phase descriptor, one section contiguous sector on the corresponding virtual volume of described target phase; Described target phase type is divided into linear and two kinds of bands, and a complete virtual volume metadata is made up of virtual volume head, paragraph header and plurality of sections,
Monitoring module, monitor in real time described management system each ingredient status information and notify the metadata management module, also monitor all kinds of faults simultaneously and take suitable recovery measure;
Step 2: the virtual proxy module of initialization
Above-mentioned meta data server sends to virtual proxy module to the Globally Unique Identifier of all required memory devices of virtual agency by the mode of creating the physical volume order, this proxy module is all memory devices in the scan storage areas network again, locate the required memory device of this virtual proxy module by the Globally Unique Identifier information that reads the memory device assigned address, and set up in Globally Unique Identifier and the kernel corresponding relation of the device number of expression memory device, be kept at and look into after preparing against in the table, finish the memory device registration, behind the host-initiated, this proxy module is beamed back unique host identification of main frame and identifying code to meta data server, and writes in the local file;
Step 3: meta data server is distributed to main frame to required virtual volume according to keeper's order and is used:
Step 3.1: meta data server sends to virtual proxy module to the Globally Unique Identifier information of the required physical volume of main frame according to keeper's order;
Step 3.2: virtual proxy module is according to the Globally Unique Identifier request scanning disk that step 3.1 provides, and finds the corresponding physical volume, the tabulation of in kernel, setting up corresponding physical volume, and send to meta data server and to reply;
Step 3.3: meta data server sends to virtual proxy module to the metadata information of distributing to the virtual volume of main frame, and this proxy module constructs virtual unit according to described metadata information in kernel, and sends to meta data server and to reply;
Step 3.4: the corresponding virtual unit that main frame uses step 3.3 to provide, and use the metadata that obtains from meta data server that the I/O order of mailing to virtual volume is redirected;
Step 4: when many front end main frames use virtual unit, when being the metadata of virtual volume, carry out according to the following steps as if this virtual volume of the online change of action need that disposes:
Step 4.1: meta data server determines to use the front end main frame of the virtual volume of the online change metadata of needs according to metadata, sends the request of the virtual volume of the online change metadata of these needs that locks to the virtual proxy module of these front end main frames;
Step 4.2: after corresponding virtual proxy module is received the described request of locking of step 4.1, being redirected in all I/O command queues but also uncompleted I/O order mail to the bottom memory device and carry out, all are follow-up this I/O order that need change the metadata virtual volume is postponed carried out, and no longer redirects; And send to meta data server and to reply;
Step 4.3: meta data server is after the success response of receiving all virtual volumes that locks, the metadata management module of this meta data server generates new metadata as required, sends the request of replacement virtual volume again to the front end main frame that has these virtual volume rights of using that lock;
Step 4.4: after virtual proxy module is received the described request of step 4.3, the old metadata of the virtual volume that locks is replaced with new metadata;
Step 4.5: meta data server sends unlocking request to virtual proxy module after the success response of receiving all replacement virtual volumes;
Step 4.6: virtual proxy module redirects all I/O that is postponed orders and carry out with new metadata.
Advantage of the present invention is as follows:
(1) metadata management separates with the map addresses function in the Storage Virtualization management, is finished by meta data server and virtual agency (virtual agency) respectively, has improved the extensibility of system, and supports the heterogeneous platform of front end main frame.
(2) all metadata managements are all finished by the meta data server unification, and virtual agency obtains required metadata information by communicating by letter with meta data server, has realized single point of management, has reduced management complexity, has improved the controllability and the security of system.
(3) use virtual volume is locked and the method for release has solved the problem of the metadata of the virtual volume that dynamic change shared by a plurality of main frames.
(4) virtual agency after obtaining metadata, no longer need with meta data server mutual, can the independent access memory device, reduced influence to greatest extent to system performance.
The present invention tests in department of computer science, Tsinghua university high-performance calculation technical institute.The result shows, can finish various virtualization based on the memory virtualized management method of meta data server, and can disregard to the loss that storage system is brought.
Respectively from the IO throughput, carry out average reaction time two aspects to the test of Storage Virtualization management system.Relatively determine the extra delay that the Storage Virtualization management system is brought by what have or not two kinds of following two indexs of situation of management system.Test environment is made up of 1 meta data server, 4 front end main frames, 1 optical fiber switch and 5 optical fiber disks.Meta data server adopts the two cpu servers of 32 Intel Xeon 2.4GHZ, internal memory 1GB, and operating system is Linux, the kernel version is 2.4.26.The optical fiber disk is the ST3146807FC of Seagate company.Front end main frame and meta data server are connected on the optical fiber switch by the optical channel card Emulex LP982 of the 2GB of emulex corp bandwidth.We adopt the iometer of intel company as testing tool, test 4 throughputs that the front end main frame is operated with 4KB data block sequential read, also tested the Linux server carries out the sequential read operation to physical disks and virtual volume with different data block sizes a response time in addition.Test result is seen Fig. 8, Fig. 9.
Fig. 8 has shown the comparison of throughput, and Fig. 9 has shown the comparison of response time.From as a result we as can be seen, virtualization layer can be ignored to the throughput and the influence of response time of system.
Description of drawings
Fig. 1. the hardware structure diagram of Storage Virtualization management system.
Fig. 2. the software architecture diagram of Storage Virtualization management system.
Fig. 3. the software configuration of meta data server.
Fig. 4. the virtual volume metadata structure.
Fig. 5. the process synoptic diagram of online change metadata.
Virtual agency's realization level and module relation diagram under Fig. 6 .Linux platform.
Virtual agency's realization level and module relation diagram under Fig. 7 .Windows platform.
Fig. 8. the throughput to physical disks and virtual volume visit compares.
Fig. 9. the response time to physical disks and virtual volume visit compares.
Figure 10. program flow chart of the present invention.
Embodiment
The virtual management system is that virtual agent software on the application server is formed by meta data server and each front end main frame mainly.Meta data server mainly is responsible for the storage resources unification of different company, different medium in the storage area network is built into storage pool, and provides virtual volume as required for each application server.And the management of responsible storage resources and configuration provide capacity extension, senior function such as data migtation.Virtual agency on each application server mainly realizes the address translation of virtual Logical Disk to the actual physics dish, and the metadata that reads the respective virtual volume from meta data server.In order to eliminate the Single Point of Faliure that a meta data server brings, can form cluster by two or many meta data servers.The hardware configuration of Storage Virtualization management system as shown in Figure 1.
Front end main frame and meta data server all are connected on the fiber optic network storage area network by the fiber buss adapter, and memory device also is connected on the storage area network by optical fiber interface.Connect by Ethernet between meta data server and the front end main frame.Metadata management module on the meta data server is finished the operation to metadata, and the address mapping module completion logic address on the virtual agency is to the mapping of physical address.Communication module on front end main frame and the meta data server communicates by TCP/IP.The software configuration of Storage Virtualization management system as shown in Figure 2.
The processing of metadata is responsible for specially by meta data server, and it " sees " real physical store, and sets up physical volume, storage pool and virtual volume thereon.Metadata is kept at the head of memory device, obtains metadata by the scanning memory device when meta data server starts and they are kept in the internal memory, is responsible for safeguarding simultaneously their consistance when metadata changes.Virtual agency on the front end main frame is a layer between file system/database layer and the bottom layer driving, and it is responsible for the mapping of logical address to physical address.Virtual agency communicates by letter with meta data server by configuration interface, obtains the configuration information of virtual volume and virtual volume is presented to the file system or the database on upper strata from meta data server.When file system or database sent I/O operation to virtual volume, virtual agency intercepted and captured this I/O operation, and logical address is translated into physical address, then the I/O operation is sent to bottom layer driving.
All are to the management of storage resources, and the configuration of virtual volume and distributing all at the enterprising line operate of meta data server.Each application server does not relate to the operation of any metadata, the metadata that is kept on the memory device is sightless for their part, the metadata that they can only obtain the virtual volume relevant with oneself from meta data server has realized single point of management, has improved security.In native system, data stream is separated with control stream (metadata streams).So, basic no datat read-write postpones on data path, has utilized the bandwidth of fiber optic network to greatest extent, better performances.In addition, virtual agency realizes having and the irrelevant characteristics of bottom hardware on the upper strata that HBA drives, can support the HBA card of different vendor.
Meta data server is in charge of whole virtual storage system, comprises the memory device and the front end main frame that are connected on the storage area network.The running software of meta data server is made up of one group of user space program on linux system, comprises 4 modules: management interface, and communication interface, metadata management module and monitoring module, the relation of intermodule is as shown in Figure 3.
Management interface provides the interface of carrying out administration order to the keeper, these orders comprise creates storage pool, virtual volume, authorize the rights of using of a front end main frame to virtual volume etc., in addition, it also offers the configuration information of required virtual volume of user management and storage pool.It finishes these functions by the API that calls the metadata management module and provide.
Host interface be responsible for and the front end main frame on virtual agency communicate by TCP/IP.When the metadata of virtual volume or the configuration of virtual store that need be written into it when the front end host-initiated changed (for example the online dilatation of virtual volume or have new physical volume to add storage pool), meta data server was notified virtual agency the information of metadata by communication interface.Virtual agency also by host interface to monitoring module report oneself state with to the operating position of virtual volume.
The metadata management module is finished the real task to metadata operation, is the nucleus module of meta data server.It is organized into storage pool to a plurality of physical volumes, and the remaining space on the managing physical volume generates the metadata of virtual volume according to the mapping mode of virtual volume.It also is responsible for being recorded in the information of metadata on the physical volume and conforming maintenance.It provides one group of API for other module invokes.
Monitoring module is finished the system status monitoring function, the status information of each ingredient of the system that obtains that it is real-time, and these ingredients comprise memory device, optical fiber switch, optical fiber link or the like.When the configuration of system changes, for example there is new memory device to join in the system or when having memory device to shift out, monitoring module can be found this incident and notify the metadata management module to change the configuration of storage pool.In addition, when certain part of system breaks down, as optical fiber link make mistakes, magnetic disk media is made mistakes etc., monitoring module is found these faults and is taked suitable recovery measure.
In theory, all memory devices in the network of network of meta data server addressable storage zone.But, if memory device need add the storage pool of being safeguarded by meta data server, and then offer each server use, then need to register this equipment, and carry out corresponding initialization process.Meta data server is finished the registration of memory device by creating the physical volume order.This order writes as memory device uniquely identified Globally Unique Identifier in certain fixed position of memory device, i.e. Globally Unique Identifier, and a certain size space to be preserved of memory device head used for depositing metadata.
Virtual memory management adopts Globally Unique Identifier as the unified sign of memory device between meta data server and virtual agency.Globally Unique Identifier is the random character string sequence of one 32 byte, and in the information that transmits between meta data server and virtual agency, physical equipment is represented with Globally Unique Identifier.Be created as in the process of physical volume at a memory device, Globally Unique Identifier is generated and is write on the assigned address of physical equipment by meta data server.In virtual agency's initialization procedure, meta data server sends to virtual agency to the Globally Unique Identifier of all required memory devices of this virtual agency.All memory devices in the virtual scan agent storage area network, locate its required memory device by the Globally Unique Identifier information of fetch equipment assigned address, and set up the corresponding relation of the device number of indication equipment in Globally Unique Identifier and the kernel, be kept in the table, in order to searching later on.The advantage of this method is, by identifier is write on the equipment, guaranteed that the location of equipment and the connected mode and the link position of equipment have nothing to do, when changing, the link position of memory device (moves on to another port) from a port, virtual agency still can this equipment of correct positioning, the dirigibility that has improved storage system.
During the virtual proxy load of each application server, not directly read metadata, need give out a contract for a project and obtain one's own metadata information to meta data server from hard disk.Meta data server will this virtual agency metadata table return this virtual agency, virtual agency fills its inner core in view of the above.In general, to shine upon the metadata to virtual agency be to be unit with the virtual volume to meta data server.
Meta data server is passed to the metadata section of being to use (segment) tissue of each virtual volume of virtual agency.Section corresponds to a target phase with one section contiguous sector of virtual volume, and target phase is one section continuous sector on one (linearity) or a plurality of (bar branch) memory device.Organizing of each section is as follows:
(logic initial sector, number of sectors, target phase type, target phase descriptor)
The target phase type has two kinds: linear (linear) and bar branch (striped).Each back all has target phase descriptor separately, and its organized formats is as follows:
Linear:
(target device sign, initial sector)
Such as: linear/dev/sda 0
The bar branch:
(the bar mark, stripe size, target device 1, initial sector 1, target device 2, initial sector 2 ...)
Such as: striped 2 32/dev/sda 0/dev/sdb 0 will shine upon the logical block (chunk) of first 16K size according to following principle:
LV?chunk?1->sda，chunk?1
LV?chunk?2->sdb，chunk?1
LV?chunk?3->sda，chunk?2
LV?chunk?4->sdb，chunk?2
Integrate, the metadata table of a possible virtual volume is as follows:
0?1028160?linear/dev/sda?0
1028160?3903762?linear/dev/sdb?0
Above-mentioned virtual volume metadata table is coupling together sdb and sdb linearity.Perhaps
0?2056320?striped?2?32/dev/sda?0/dev/sdb?0
2056320?2875602?linear/dev/sdb?1028160
Above-mentioned virtual volume metadata table is that logical block 0-2056320 bar branch is mapped on disk sda and the disk sdb, and stripe size is 16K.Logical block 2056320-2875602 then is 819282 sectors of linear mapping to disk sdb since 1028160 sectors.
Meta data server sends to the complete structure of metadata of virtual agency's a virtual volume and sees Fig. 4.
Wherein, the virtual volume head comprises virtual volume for information about, as rolls up size, access limit etc.; Paragraph header has indicated the number of the section that is connect thereafter; The form of each section as previously mentioned.
This segmentation method for expressing and the mapping table method for expressing (piece that virtual volume is divided into the plurality of fixed size, the piece of identical size on the corresponding memory device of each piece) compares, significantly reduced the length of the metadata of representing that virtual volume is required, shortened that metadata is stored and on meta data server in time of transmission over networks.
Meta data server and virtual agency communicate by Ethernet by communication interface separately.Communication is base unit with the session, and a complete session is made up of two parts:
1. request: the end by traffic model sends a request package, packet header+data (if there is) to an other end
2. reply: when an other end is asked, just carry out this request, and return the execution result of a packet header+data (if there is)
If communication failure is just thought in the change of not replying in official hour, no longer wait for, think this communication process failure.
The request of all communication all is to represent as RQST_XXXXX, and replying of all communication all is to represent as RESP_XXXXX.Main request type comprises:
The RQST_LOGIN main frame logs on meta data server
The RQST_LOGOUT main frame withdraws from from meta data server
RQST_SETPV is provided with the required physical volume of main frame
RQST_SETLV is provided with the metadata of the virtual volume of main frame
RQST_RESETLV resets the metadata of virtual volume
RQST_LOCKLV locks virtual volume
RQST_UNLOCKLV release virtual volume
A main frame will add in the virtual storage system, need register to meta data server.Behind the host-initiated, virtual agency sends registration packet automatically and returns to unique host identification of main frame to meta data server, i.e. Host ID and an identifying code, i.e. host key.Virtual agency can deposit this sign in local file.This main frame of later all and meta data server are mutual, all distinguish main frame with host identification.
After main frame was finished registration, meta data server just can be distributed to main frame to virtual volume according to keeper's order and use.Meta data server at first sends the SETPV request package to virtual agency, and the Globally Unique Identifier information of the used physical volume of main frame is sent to virtual agency.Virtual agency asks scanning disk according to this, finds the corresponding physical volume, the tabulation of in kernel, setting up all physical volumes, and to meta data server transmission SETPV respond packet.Meta data server sends the SETLV request package to virtual agency then, and the metadata information of the virtual volume of distributing to main frame is sent to virtual agency.Virtual agency constructs corresponding virtual unit according to metadata information in kernel, and sends the SETLV respond packet to meta data server.Afterwards, main frame just can use this virtual unit.In use, virtual agency utilizes the metadata that obtains from meta data server that the I/O order of mailing to virtual volume is redirected, unless the metadata of virtual volume changes, otherwise virtual agency does not need to communicate by letter with meta data server.
In many front ends main frame shared storage environment of storage area network, a virtual volume may be distributed to a plurality of front end main frames and use.In order to satisfy the demand of uninterrupted service, when configuration operation need change the metadata (being the mapping relations of virtual volume and physical volume, as the virtual volume dilatation) of this virtual volume, can not interrupt the service that the front end main frame provides.Therefore must control the visit of virtual volume the front end main frame in this process, to guarantee the consistance of data.The consistance of data when the virtual management system adopts a kind of metadata of lock mechanism assurance virtual volume to change.Meta data server can the forward end main frame virtual agency send locking and unlocking request to virtual volume.Virtual agency receive to a virtual volume lock the request after, the I/O order that has been redirected in all I/O command queues but has not also finished is addressed to the bottom memory device to be carried out, all follow-up I/O orders to virtual volume are postponed to be carried out, and is not redirected.After virtual agency receives unlocking request to a virtual volume, all postponed the I/O order redirected with new metadata, and it is medium pending to be reentered into I/O command queue.
The process of online change virtual volume metadata as shown in Figure 5.At first meta data server can determine which front end main frame this virtual volume has been distributed to and used according to metadata, send the request of this virtual volume that locks then to the virtual agency of these front end main frames, virtual agency receives that this request back execution adds latching operation and beams back response to meta data server.Meta data server is after the success response of receiving all virtual volumes that locks, its metadata management module generates new metadata as required, send the request of replacement virtual volume again to the front end main frame that has these virtual volume rights of using, virtual agency receives that the request back replaces with new metadata to the old metadata of virtual volume.At last, meta data server sends the request of release virtual volume to virtual agency after the success response of receiving all replacement virtual volumes, virtual agency all postponed I/O order be redirected and execution with new metadata.
Owing to have only after the old metadata of all virtual agencies' respective virtual volume all replaces with new metadata, meta data server just can send unlocking request, causes the inconsistent situation of data so can avoid the different virtual agency to use different metadata that virtual volume is conducted interviews.In addition, when certain link in this process produced mistake, meta data server returned to the consistance that the preceding state of command execution guarantees metadata with a kind of mode of rollback.
Virtual agency on each server mainly realizes the virtual address translation that is rolled onto the actual physics dish, and reads the metadata of response virtual volume from meta data server, and dynamically some status informations of virtual volume is returned to meta data server.Virtual agency mainly supports Linux and windows platform.Realization level and the module relationship of virtual agency under Linux and the windows platform are seen Fig. 6, Fig. 7.
Virtual agency under Linux and the windows platform is made up of two parts: the communication module of user's attitude and the address mapping module of kernel state.Communication module obtains the metadata information of virtual volume from meta data server, and these information are imported kernel module.Kernel module is a level between file system layer and the block device Drive Layer, is responsible for carrying out map addresses according to the metadata of virtual volume.After metadata information was imported into, unless metadata changes, virtual agency no longer needed to communicate with meta data server, can the independent access memory device, reduced influence to greatest extent to system performance.
Claims (1)
1. based on the memory virtualized management method of meta data server, it is characterized in that this method is to realize that this system contains in the virtual management system that is made of following equipment:
Various types of front end main frames, i.e. application server, the virtual proxy module of this front end main frame are realized the virtual conversion of actual physical dish and the metadata that reads the respective virtual volume from following meta data server of being rolled onto;
Meta data server, one or more is arranged, link to each other with above-mentioned each front end main frame through Ethernet according to ICP/IP protocol, the storage resources unification of different company, different medium in the storage area network is built into physical volume, storage pool and virtual volume, and, realize the management and the configuration of storage resources for above-mentioned each front end main frame provides virtual volume;
Memory device, the in store metadata of head, each memory device has a unique general identifications, represents with globally unique identifier;
Optical fiber switch is HBA by the fiber buss adapter, links to each other with meta data server with above-mentioned each front end main frame, also link to each other with above-mentioned memory device simultaneously by optical fiber interface,
Described memory virtualized management method based on meta data server contains following steps successively:
Step 1: structure software module
Virtual agency is the agent module, it is a layer between file system/database layer and the bottom layer driving on each front end main frame, it also is kernel module, be responsible for the mapping of logical address to physical address, communicate by letter with the form of control stream by the communication interface of configuration and the communication interface of above-mentioned meta data server, obtain the configuration information of virtual volume and virtual volume is presented to topmost paper system/database from above-mentioned meta data server; When file system/database sent the I/O of virtual volume operation, this virtual proxy module intercepted this I/O operation, and logical address is mapped to physical address, form data stream after, again the I/O operation is sent to the bottom layer driving device,
The software of meta data server operates on the linux system, is made up of one group of user space program, comprises with lower module:
Management interface, the interface of carrying out administration order is provided to the keeper, wherein contain and create storage pool, virtual volume, authorize of the order of a front end main frame to the rights of using of virtual volume, this management interface also provides the required virtual volume and the configuration information of storage pool to the keeper in addition, and this management interface is to realize by the API that following metadata management module provides;
Host interface, be responsible for above-mentioned each front end main frame on communication interface communicate by letter according to ICP/IP protocol, when the front end host-initiated requires to be written into the metadata of virtual volume in the meta data server, perhaps when the configuration of virtual store changes, meta data server is notified virtual proxy module the information of metadata by communication interface, virtual proxy module also by host interface to following monitoring module report oneself state with to the operating position of virtual volume;
The metadata management module, execution is to the operation task of metadata, a plurality of physical volumes are organized into storage pool, remaining space on the managing physical volume, generate the metadata of virtual volume according to the mapping mode of virtual volume, again the information of metadata is recorded on the physical volume, this metadata management module also has one group of API for other module invokes, described metadata is used the section tissue, contain: logic initial sector, number of sectors, target phase type and target phase descriptor, one section contiguous sector on the corresponding virtual volume of described target phase; Described target phase type is divided into linear and two kinds of bands, and a complete virtual volume metadata is made up of virtual volume head, paragraph header and plurality of sections,
Monitoring module, monitor in real time described management system each ingredient status information and notify the metadata management module, also monitor all kinds of faults simultaneously and take suitable recovery measure;
Step 2: the virtual proxy module of initialization
Above-mentioned meta data server sends to virtual proxy module to the globally unique identifier of all required memory devices of virtual agency by the mode of creating the physical volume order, this proxy module is all memory devices in the scan storage areas network again, locate the required memory device of this virtual proxy module by the globally unique identifier's information that reads the memory device assigned address, and set up in globally unique identifier and the kernel corresponding relation of the device number of expression memory device, be kept at and look into after preparing against in the table, finish the memory device registration, behind the host-initiated, this proxy module is beamed back unique host identification of main frame and identifying code to meta data server, and writes in the local file;
Step 3: meta data server is distributed to main frame to required virtual volume according to keeper's order and is used:
Step 3.1: meta data server sends to virtual proxy module to globally unique identifier's information of the required physical volume of main frame according to keeper's order;
Step 3.2: globally unique identifier's request scanning disk that virtual proxy module provides according to step 3.1, find the corresponding physical volume, the tabulation of in kernel, setting up corresponding physical volume, and send to meta data server and to reply;
Step 3.3: meta data server sends to virtual proxy module to the metadata information of distributing to the virtual volume of main frame, and this proxy module constructs virtual unit according to described metadata information in kernel, and sends to meta data server and to reply;
Step 3.4: the corresponding virtual unit that main frame uses step 3.3 to provide, and use the metadata that obtains from meta data server that the I/O order of mailing to virtual volume is redirected;
Step 4: when many front end main frames use virtual unit,, carry out according to the following steps if during the metadata of online this virtual volume of change of action need of configuration:
Step 4.1: meta data server determines to use the front end main frame of the virtual volume of the online change metadata of needs according to metadata, sends the request of the virtual volume of the online change metadata of these needs that locks to the virtual proxy module of these front end main frames;
Step 4.2: after corresponding virtual proxy module is received the described request of locking of step 4.1, being redirected in all I/O command queues but also uncompleted I/O order mail to the bottom memory device and carry out, all are follow-up this I/O order that need change the metadata virtual volume is postponed carried out, and no longer redirects; And send to meta data server and to reply;
Step 4.3: meta data server is after the success response of receiving all virtual volumes that locks, the metadata management module of this meta data server generates new metadata as required, sends the request of replacement virtual volume again to the front end main frame that has these virtual volume rights of using that lock;
Step 4.4: after virtual proxy module is received the described request of step 4.3, the old metadata of the virtual volume that locks is replaced with new metadata;
Step 4.5: meta data server sends unlocking request to virtual proxy module after the success response of receiving all replacement virtual volumes;
Step 4.6: virtual proxy module redirects all I/O that is postponed orders and carry out with new metadata.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100114094A CN1304961C (en) | 2005-03-11 | 2005-03-11 | Memory virtualized management method based on metadata server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100114094A CN1304961C (en) | 2005-03-11 | 2005-03-11 | Memory virtualized management method based on metadata server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1664793A CN1664793A (en) | 2005-09-07 |
| CN1304961C true CN1304961C (en) | 2007-03-14 |
Family
ID=35035895
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100114094A Expired - Fee Related CN1304961C (en) | 2005-03-11 | 2005-03-11 | Memory virtualized management method based on metadata server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1304961C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9965184B2 (en) | 2015-10-19 | 2018-05-08 | International Business Machines Corporation | Multiple storage subpools of a virtual storage pool in a multiple processor environment |
Families Citing this family (43)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI359377B (en) * | 2005-04-05 | 2012-03-01 | Ibm | System and method for providing execute-in-place f |
| EP1798934A1 (en) * | 2005-12-13 | 2007-06-20 | Deutsche Thomson-Brandt Gmbh | Method and apparatus for organizing nodes in a network |
| US7603392B2 (en) * | 2006-06-05 | 2009-10-13 | International Business Machines Corporation | System, method and computer program product for storing transient state information |
| CN101495973A (en) * | 2006-06-09 | 2009-07-29 | 迅龙国际集团有限公司 | N^N data management, access, storage, transmission, exchange and retrieval system (data concentration management) |
| JP4331220B2 (en) * | 2007-03-07 | 2009-09-16 | 株式会社東芝 | Storage device with autonomous management function of unused physical area |
| US8005993B2 (en) * | 2007-04-30 | 2011-08-23 | Hewlett-Packard Development Company, L.P. | System and method of a storage expansion unit for a network attached storage device |
| CN101118558B (en) * | 2007-07-10 | 2011-05-25 | 青岛海信移动通信技术股份有限公司 | Converting method and apparatus for resource file |
| JP5252953B2 (en) * | 2008-02-22 | 2013-07-31 | キヤノン株式会社 | COMMUNICATION DEVICE, COMMUNICATION METHOD, PROGRAM, AND STORAGE MEDIUM |
| JP5374056B2 (en) * | 2008-02-26 | 2013-12-25 | 株式会社バッファロー | Data management method |
| JP5072692B2 (en) * | 2008-04-07 | 2012-11-14 | 株式会社日立製作所 | Storage system with multiple storage system modules |
| US8484204B2 (en) * | 2008-08-28 | 2013-07-09 | Microsoft Corporation | Dynamic metadata |
| CN101582092B (en) * | 2009-06-12 | 2011-04-20 | 中兴通讯股份有限公司 | Method and device for realizing the store of date in memory |
| CN101621537B (en) * | 2009-07-24 | 2012-09-05 | 北京航空航天大学 | Hyperspectral mass data parallel processing system with low cost |
| CN101753617B (en) * | 2009-12-11 | 2013-05-08 | 中兴通讯股份有限公司 | Cloud storage system and method |
| CN101788889B (en) * | 2010-03-03 | 2011-08-10 | 浪潮(北京)电子信息产业有限公司 | Memory virtualization system and method |
| CN102118442A (en) * | 2011-02-21 | 2011-07-06 | 清华大学 | Method and device for accessing Web resources |
| CN102143228A (en) * | 2011-03-30 | 2011-08-03 | 浪潮(北京)电子信息产业有限公司 | Cloud storage system, cloud client and method for realizing storage area network service |
| CN102202087B (en) * | 2011-04-25 | 2015-04-01 | 中兴通讯股份有限公司 | Method for identifying storage equipment and system thereof |
| TWI614670B (en) * | 2013-02-12 | 2018-02-11 | Lsi公司 | Chained, scalable storage system and method of accessing data in a chained, scalable storage system |
| US9619404B2 (en) | 2013-04-16 | 2017-04-11 | International Business Machines Corporation | Backup cache with immediate availability |
| US9329938B2 (en) | 2013-04-16 | 2016-05-03 | International Business Machines Corporation | Essential metadata replication |
| US9104332B2 (en) | 2013-04-16 | 2015-08-11 | International Business Machines Corporation | Managing metadata and data for a logical volume in a distributed and declustered system |
| US9423981B2 (en) | 2013-04-16 | 2016-08-23 | International Business Machines Corporation | Logical region allocation with immediate availability |
| US9298398B2 (en) | 2013-04-16 | 2016-03-29 | International Business Machines Corporation | Fine-grained control of data placement |
| US9298617B2 (en) | 2013-04-16 | 2016-03-29 | International Business Machines Corporation | Parallel destaging with replicated cache pinning |
| US9104597B2 (en) | 2013-04-16 | 2015-08-11 | International Business Machines Corporation | Destaging cache data using a distributed freezer |
| CN103209219A (en) * | 2013-04-23 | 2013-07-17 | 深圳市京华科讯科技有限公司 | Distributed cluster file system |
| CN104636077B (en) * | 2013-11-15 | 2017-10-10 | 中国电信股份有限公司 | Network control techology storage system and method for virtual machine |
| CN103902354A (en) * | 2014-04-11 | 2014-07-02 | 浪潮电子信息产业股份有限公司 | Method for rapidly initializing disk in virtualization application |
| CN104238963B (en) * | 2014-09-30 | 2017-08-11 | 华为技术有限公司 | A kind of date storage method, storage device and storage system |
| US10209899B2 (en) * | 2015-11-06 | 2019-02-19 | Microsoft Technology Licensing, Llc | Storage virtualization offload |
| CN107122123A (en) * | 2016-02-24 | 2017-09-01 | 湖南百里目科技有限责任公司 | A kind of new Storage Virtualization gateway direct mode operation method |
| CN106919346B (en) * | 2017-02-21 | 2019-01-22 | 无锡华云数据技术服务有限公司 | A kind of shared Storage Virtualization implementation method based on CLVM |
| CN110300960B (en) * | 2017-02-28 | 2023-04-04 | 株式会社日立制作所 | Information system, management program, and program replacement method for information system |
| CN107154848A (en) * | 2017-03-10 | 2017-09-12 | 深圳市盾盘科技有限公司 | A kind of data encryption based on CPK certifications and storage method and device |
| CN107291402A (en) * | 2017-07-05 | 2017-10-24 | 长沙开雅电子科技有限公司 | A kind of new transparent transmission implementation method of Storage Virtualization |
| CN107888689B (en) * | 2017-11-16 | 2019-04-30 | 无锡地铁集团有限公司 | Locking resource allocation method based on shared storage |
| CN111259010A (en) * | 2020-01-14 | 2020-06-09 | 北京智睿博信息技术有限公司 | Management method and system for associated configuration information of storage area network equipment |
| EP4160422A4 (en) * | 2020-07-02 | 2023-12-06 | Huawei Technologies Co., Ltd. | Method for using intermediate device to process data, computer system, and intermediate device |
| CN112241365B (en) * | 2020-07-24 | 2022-09-30 | 国家计算机网络与信息安全管理中心 | Method and device for efficient metadata construction and RPC performance improvement for Namenode |
| CN112968965B (en) * | 2021-02-25 | 2022-12-09 | 网宿科技股份有限公司 | Metadata service method, server and storage medium for NFV network node |
| US11943115B2 (en) * | 2022-04-05 | 2024-03-26 | International Business Machines Corporation | Locally debugging remote deployment of microservices |
| CN116360696B (en) * | 2023-03-31 | 2024-01-23 | 北京志凌海纳科技有限公司 | Distributed block storage system, medium, and electronic device thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1506844A (en) * | 2002-11-19 | 2004-06-23 | �Ҵ���˾ | Hierarchy storage management method and apparatus using dynamic content table and content table collection |
| CN1540524A (en) * | 2003-10-31 | 2004-10-27 | 清华大学 | Method for accessing distributed and virtualized storage in local network |
| US6839812B2 (en) * | 2001-12-21 | 2005-01-04 | Intel Corporation | Method and system to cache metadata |
-
2005
- 2005-03-11 CN CNB2005100114094A patent/CN1304961C/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6839812B2 (en) * | 2001-12-21 | 2005-01-04 | Intel Corporation | Method and system to cache metadata |
| CN1506844A (en) * | 2002-11-19 | 2004-06-23 | �Ҵ���˾ | Hierarchy storage management method and apparatus using dynamic content table and content table collection |
| CN1540524A (en) * | 2003-10-31 | 2004-10-27 | 清华大学 | Method for accessing distributed and virtualized storage in local network |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9965184B2 (en) | 2015-10-19 | 2018-05-08 | International Business Machines Corporation | Multiple storage subpools of a virtual storage pool in a multiple processor environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1664793A (en) | 2005-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1304961C (en) | Memory virtualized management method based on metadata server | |
| CN104603739B (en) | The block level of parallel memorizing is accessed | |
| US9501233B2 (en) | Providing snapshots of virtual storage devices | |
| US7877545B2 (en) | Online restriping technique for distributed network based virtualization | |
| US8706837B2 (en) | System and method for managing switch and information handling system SAS protocol communication | |
| US8631150B2 (en) | Method and apparatus for in-band management of storage devices | |
| US7478177B2 (en) | System and method for automatic reassignment of shared storage on blade replacement | |
| CN100396065C (en) | A method for implementing iSCSI memory system | |
| CN100347692C (en) | Implementing method of virtual intelligent controller in SAN system | |
| CN1623136A (en) | Methods and apparatus for implementing virtualization of storage within a storage area network through a virtual enclosure | |
| CN105471695A (en) | Storage area network virtualization management method | |
| US20060272015A1 (en) | Virtual devices and virtual bus tunnels, modules and methods | |
| CN102340530B (en) | The method and system of a kind of memory space adapter and Data Migration | |
| US20020049825A1 (en) | Architecture for providing block-level storage access over a computer network | |
| CN1688982A (en) | Multi-protocol storage appliance that provides integrated support for file and block access protocols | |
| CN1766851A (en) | Systems and methods for data storage management | |
| JP2017091546A (en) | Storage device including multiple partition for multimode operation and method for operating storage device | |
| CN107170474A (en) | Expansible the storage box, computer implemented method and computer readable storage means | |
| CN101689148A (en) | offloading storage operations to storage hardware | |
| JP2003248555A (en) | A system for partitioning data storage area network associated data library using element address | |
| JP2003263397A (en) | Method for providing peripheral device virtual functionality overlay system | |
| CN1766885A (en) | Systems and methods for supporting managed data | |
| CN1655129A (en) | Universal method for dynamical management of storage resource under Windows platform | |
| CN1818890A (en) | Rnic-based offload of iscsi data movement function by initiator | |
| CN103870202A (en) | Distributed storage method and system of block device |
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: 20070314 Termination date: 20110311 |