CN101501656A - System for archival storage of data - Google Patents
System for archival storage of data Download PDFInfo
- Publication number
- CN101501656A CN101501656A CNA2006800312365A CN200680031236A CN101501656A CN 101501656 A CN101501656 A CN 101501656A CN A2006800312365 A CNA2006800312365 A CN A2006800312365A CN 200680031236 A CN200680031236 A CN 200680031236A CN 101501656 A CN101501656 A CN 101501656A
- Authority
- CN
- China
- Prior art keywords
- storage medium
- auxilliary
- storage system
- data cell
- data
- 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.)
- Pending
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A secondary storage system for maintaining data units transferred from a primary storage system is provided. The secondary storage system includes secondary storage media. Not all of the secondary storage media are powered on at the same time. The secondary storage media includes at least one storage medium that is always in the powered-on mode. Metadata is stored in one or more of at least the one storage medium in the powered-on mode. The metadata includes at least one attribute of a data unit stored in a secondary storage medium that is in the lower power mode of operation than at least the one storage medium that is always in the powered-on mode.
Description
The application requires the right of priority of following application, and as what propose in detail among the application, its content merges as a reference at this:
The U.S. Provisional Patent Application sequence number No.60/722 of " the SYSTEM FOR ARCHIVALSTORAGE OF DATA " by name that submitted on September 29th, 2005,215, and the U.S. Provisional Patent Application sequence number No.60/730 of " the USER INTERFACE FORARCHIVAL STORAGE OF DATA " by name of submission on October 25th, 2005,288.
Technical field
Specific embodiment relates in general to data-storage system, relates more specifically to filing system.
Background technology
Data are backed up or file duplicate normally and important.File can discharge main storage system to hold extra data.File can also make data lose, damage or occur can being resumed after the mistake.Can also improve the system effectiveness of the data of asking without frequentation.
Typical filing system uses array of disk drives as its main storage system.Filing system is duplicated or transferred to data from main storage system.This filing system is bigger than Entry-level System usually, speed is slow and cost is low.For example, filing system can use tape drive, slow disk drive, CD-ROM driver to wait and store data.In other words, it is lower and consume less power filing system can be designed to the cost of each storage unit.Creating effective filing system must be very careful, so that the storage between Entry-level System and the filing system and obtaining can not be disturbed the integrated operation that is subjected to the computer system that this filing system design supports.
For the smooth operation of polytype computer utility, the System Administrator Management archival task, check, organize and recover history file and catalogue and carry out the ability of other function extremely important.
Summary of the invention
According to various embodiments, a kind of auxilliary storage system is provided, be used to keep shift the data cell of coming from main storage system.Should auxilliary storage system comprise auxilliary storage medium, described auxilliary storage medium is not all to be in simultaneously to add power mode.In addition, auxilliary storage medium comprises and is at least one storage medium that adds power mode all the time.Auxilliary storage medium also comprises and is stored in the metadata that is in all the time at least one or the more a plurality of storage medium that adds power mode.This metadata comprises at least one attribute that is in the data cell in the auxilliary storage medium that at least one storage medium that adds power mode other are in low-power operating mode all the time except described.
According to embodiment, provide a kind of being used for to shift the method that the data cell of coming remains on auxilliary storage system from main storage system.Auxilliary storage system comprises and is not in the auxilliary storage medium that adds power mode simultaneously.In addition, auxilliary storage medium comprises and is at least one storage medium that adds power mode all the time.This method comprises: determine the metadata in one or more data cell in the auxilliary storage medium.This metadata comprises the attribute that is in the data cell at least one auxilliary storage medium under the lower power mode except being at least one storage medium that adds power mode all the time.In addition, this method comprises metadata store in being at least one storage medium that adds power mode all the time.This attribute allow to determine be in lower power mode under at least one auxilliary storage medium in the relevant information of data cell.
Description of drawings
Hereinafter will describe a plurality of embodiment of the present invention in conjunction with the accompanying drawings, the accompanying drawing that is provided is unrestricted the present invention for illustration, and wherein similar Reference numeral is represented similar components, in the accompanying drawings:
Fig. 1 shows the block scheme according to the general structure of the archival data storage system that links to each other with client device of a plurality of embodiment.
Fig. 2 shows the block scheme according to the processing module in the frame of embodiment.
The block scheme of the auxilliary storage system that is used for storage data units that provides according to embodiment is provided Fig. 3.
Fig. 4 shows the block scheme that is used for filing system that the data unit is filed according to embodiment.
Fig. 5 shows the process flow diagram that is used for data cell is remained on the method for auxilliary storage system according to a plurality of embodiment.
Fig. 6 shows and is used to provide process flow diagram about the method for the information of data cell according to embodiment.
Fig. 7 shows the diagram according to scalable (scalable) filing system of embodiment.
Can carry out various modifications and alterative version to the present invention, and in accompanying drawing and appended detailed description, show specific embodiments of the invention as example.However, it should be understood that accompanying drawing and detailed description are not intended to the present invention is limited to specific embodiment as described herein.The disclosure is intended to cover all modifications, the equivalent and alternative in the scope of the invention that falls into the claims qualification.
Embodiment
One or more embodiment of the present invention is described below.These that will describe below it should be noted that and any other embodiment are exemplary, and all are intended to example and unrestricted the present invention.
Embodiments of the invention provide a kind of method, system and computer program that is used for the data archiving storage system.This data archiving storage system be used for from the multiple archiving files of main storage system in auxilliary storage system, multiple file is got the main storage system and is managed these files from auxilliary storage system.
Fig. 1 shows the block scheme according to the general structure of the archival data storage system that links to each other with client device of a plurality of embodiment.Archival data storage system 100 comprises client 102, network 104, switch (switch) 106 and filing system 108.Archival data storage system 100 can comprise a plurality of clients and a plurality of filing system.A plurality of clients can communicate by network 104 and a plurality of filing systems.The example of network 104 comprises mobile network, individual territory net (PAN), Local Area Network, Metropolitan Area Network (MAN) (MAN), internet and wide area network (WAN), but is not limited to this.In an embodiment, network 104 can be the one or more combination in the above-mentioned network.
Main storage system can be stored the data cell such as file and catalogue.May have restriction for the data area that can be stored in the main storage system, the max cap. that for example is stored on the hard disk can be the 80G byte.Data cell can be archived to filing system 108 from main memory unit.In an embodiment, the Ethernet switch of gigabit can link to each other network 104 with filing system 108.Filing system 108 comprises frame (rack) 110 and auxilliary storage system 112.The data file of filing can be stored in the auxilliary storage system 112.Frame 110 can be used for implementing multiple operation, for example the data cell that is stored in filing system 108 places is filed or obtains.Frame 110 can also power up the auxilliary storage medium that is in low-power mode.Frame 110 has one or more processing modules, below in conjunction with Fig. 2 processing module is described in detail.
May be slower from the auxilliary storage medium visit data unit that is in low-power operating mode than the situation that this auxilliary storage medium is in powering state.In an embodiment, filing system 108 is based on independent/inexpensive redundant arrays of disks (RAID) system that is subjected to power management or be subjected to extensive non-movable disk array (MAID) system of power management.
In being subjected to the storage system of power management,, once only a limited number of memory device is powered up according to admissible maximum power dissipation or " power budget ".For example, United States Patent (USP) 7 at " Method andApparatus for Power Efficient High-capacity Storage System " by name, 035, in 972 (propose in detail, its content merges as a reference at this) the RAID system that is subjected to power management has been described at this document.
In an embodiment, can use I/O (I/O) to engage the MAID part visit data unit of (coalescing) system from system.This technology powers up and cuts off the power supply by I/O request rearrangement has been avoided unnecessary for the cluster (rather than according to they original received orders) of visiting identical drivers simultaneously.
The metadata that is stored in the data cell on the auxilliary storage system 112 can be stored in the first auxilliary storage medium that powers up all the time.This metadata can comprise one or more attributes of data cell.Even when the second auxilliary storage medium is in low-power operating mode, this metadata also can be used for checking the attribute of the data cell of storing at the second auxilliary storage medium.
Metadata represents to can be used for identifying the data cell attribute of this data cell.The attribute of data cell comprises size of the establishment of the owner of title, data cell of data cell or author, data cell and/or up-to-date modification date, data cell etc.In an embodiment, can receive inquiry or the request that the data cell that is stored in the auxilliary storage system is filed or obtained.Can submit this inquiry to by using the graphic user interface (GUI) in the client 102.For example, can be from the data cell being stored at least one of main storage system and auxilliary storage system 112 search have all data cells of extension name " .txt ".In addition, even when one or more disk drives of storage data units are in low-power operating mode, also can provide being stored in checking of data cell on the auxilliary storage medium.The metadata that is used for storing the filing system 100 of data can be stored in the first auxilliary storage medium that powers up all the time.This metadata can be stored and is stored in the second relevant information of data cell of assisting on the storage medium that is in low-power operating mode.For checking the data cell that is stored on the second auxilliary storage medium, can this second auxilliary storage medium not powered up.This metadata is used to provide the attribute that is stored in the data cell on the auxilliary storage medium.Use this attribute to create and check view, and do not need the second auxilliary storage medium is powered up.The filing system 100 that is used to store data can come multiple operation is carried out in the data unit by means of metadata, and does not need the second auxilliary storage medium is not powered up.Yet,, need power up the second auxilliary storage medium in order to read in the content of the data cell of storing on the second auxilliary storage medium.In addition, can be by means of the metadata that is stored on the first auxilliary storage medium that powers up all the time, search data unit in the second auxilliary storage medium.For search is stored in second data cell of assisting on the storage medium that is in low-power operating mode, do not need the second auxilliary storage medium is powered up.
Fig. 2 shows the block scheme according to the processing module in the frame 110 of embodiment.Frame 100 comprises the processing module such as metadata access library (MAL) 202, file-archiver 204 and power management module 206.MAL 202 can storing metadata, and metadata comprises on directory level checks, identifies and carry out attribute and the multiple parameter that master data is handled the necessary data cell of operation.Check that view can provide the different tissues to data.The master data that can carry out at filing system 108 places is handled operation and can be included as the archival task designated data unit, data cell is got main memory unit from auxilliary storage system 112, or the like.
Metadata can be used for carrying out to being stored in the inquiry of the data cell of assisting storage system 112 by file-archiver 204.File-archiver 204 can also move the original user data position of data file from main storage system or transfer to auxilliary storage system 112, and keeps the raw data file former state constant.In another embodiment, can the filing system 100 that be used to store data be configured, so that deletion is archived to the data file of auxilliary storage system 112 from main storage system by file-archiver 204 from main storage system.
In addition, file-archiver 204 is used the metadata that is stored in the data cell on the first auxilliary storage medium that powers up all the time.Aforesaid metadata comprises and is stored in the relevant information of data cell that is on the auxilliary storage medium of second under the low-power operating mode.The information of the data cell on being stored in the second auxilliary storage medium, metadata also comprises the position of data cell.When file-archiver 204 receives the request of checking data cell, can be by means of metadata, show the detailed content that is stored in the data cell on the second auxilliary storage medium to the user of the filing system 100 that is used to store data.For checking, do not need the second auxilliary storage medium is powered up about for the information of data cell.Except the information of data cell, can also be by means of the position of metadata to user's video data unit of the filing system 100 that is used to store data.Similarly, receive at file-archiver 204 places at data cell read request the time, file-archiver 204 is come the position of identification data unit by means of metadata.Then, begin the second auxilliary storage medium of storage data units is powered up from low-power operating mode, can the reading of data unit so that be used in the user of the filing system 100 of storage data.
The second auxilliary storage medium that is in low-power operating mode is not powered up, can check data cell yet.Yet, when obtaining the data cell that is stored on the second auxilliary storage medium in response to inquiry, may need the second auxilliary storage medium is powered up.In an embodiment, power management module 206 can dispose be used for will be in the second auxilliary storage medium of low-power operating mode be transformed into and add power mode.Before receiving, can power up the second auxilliary storage medium at the request that is stored in the data cell in the auxilliary storage medium.
In an embodiment, frame 110 can also comprise network file system(NFS) (NFS) client 208, nfs server 210, file-archiver read-only file system (FARFS) 212, administration interface 214, Virtual File System (VFS) 216, the file system 218 such as unix filesystem (UFS) and fiber channel drivers 220.FARFS 212 is piling up in the operating system that is embedded on the VFS (stackable) file system layers.NFS client 208 can send duplicates or moves to the request of assisting storage system 112 with the data cell such as data file from main storage system.The request of filing or obtaining to the data unit can be handled by nfs server 210.The user that administration interface 214 allows to be used to store the filing system 100 of data checks metadata.Administration interface 214 can also make the user can check the result of inquiry, and this inquiry is to carry out to obtain data cell.Then, the user can be from administration interface 214 selection results, and visit corresponding data cell.In an embodiment, fiber channel drivers 220 can connect optical fiber interconnections, so that in operation frame 110 is linked to each other with auxilliary storage system 112.One or more stand module can link to each other with file system 218 with VFS 216 on function, to carry out alternately with auxilliary storage system 112.In addition, fibre channel interconnect can be installed the connection of multi-to-multi.
Fig. 3 shows the block scheme according to the auxilliary storage system that is used for storage data units of embodiment.Auxilliary storage system 112 comprises the first and second auxilliary storage mediums that can be used for storage data units.The first auxilliary storage medium is in all the time and adds power mode.On the other hand, the second auxilliary storage medium can be in low-power operating mode in preset time, and the second auxilliary storage medium is entered add power mode.The first auxilliary storage medium can comprise the one or more racks that are used for storage data units.For example, the first auxilliary storage medium shown in comprises and is in first rack 302 that powers up operator scheme all the time.Similarly, the second auxilliary storage medium can also comprise the one or more racks that are used for storage data units.In Fig. 3, the one or more racks that are used for storage data units are shown data rack 304.The number that However, it should be understood that the data rack that can comprise in the second auxilliary storage medium can be greater or less than the number shown in Fig. 3.
The metadata store of data cell is in being in the first auxilliary storage medium that adds power mode all the time.All the time being in the first auxilliary storage medium that powers up operator scheme can also storage data units.Metadata can comprise basic file attributes, for example establishment of the title of data cell, data cell and/or revise date, the size of data cell, the type of data cell, or the like.In addition, the realization demand according to specific can define more multiattribute, and these attributes can be associated with data cell.These attributes can also be appended on the data cell with a part as metadata.For example, when carrying out inquiry, comprise be associated with data cell, may be very useful as the author of the part of the metadata of data cell or founder's title.Can also be in conjunction with key word that can the identification data unit, as the metadata of data cell.The key word of data cell and other attribute can be defined by the user, and can be included in the metadata of data cell.For example, content that can the definition of data unit, thus when even for example file content such as authentic document content is archived on the second auxilliary storage medium that is in low-power operating mode, also can carry out keyword search to the data cell of filing.Like this, a large amount of (for example 1000 GB) data cell can be archived on the second auxilliary storage medium that is in low-power operating mode, still can carry out multiple basic function simultaneously the data unit.
In an embodiment, the metadata of data cell can comprise versioned (versioning) information, and this versioned information can be used for providing the information about data cell.For example, a plurality of versions of same file can be archived on the auxilliary storage system 112.Can specify the job description of filing or obtaining task, so that all copies that system can storage data units, perhaps system can only keep ' n ' individual copy (wherein, n 〉=1 of data cell; And n is an integer).When reaching ' n ' individual copy threshold, during the redaction of each this data cell of in filing system 108, filing, just can delete version the earliest.
In an embodiment, when a plurality of request of receiving at auxilliary storage system 112 places from client 104, may need the mechanism that resequences, with to the file that receives at auxilliary storage system 112 places or obtain and ask to sort.This rearrangement mechanism can be configured in the auxilliary storage system 112, so that a plurality of requests from one or more clients are resequenced.Request can be divided in proper order first order request, second order request, the 3rd order request etc.First order request can allow the part in a plurality of requests sequentially to visit first and second first storage mediums of assisting in the storage mediums.In addition, second order request can allow second storage medium that another part visit first and second in a plurality of requests is assisted in the storage mediums.Can resequence to a plurality of requests, begin the number of times that same storage medium is powered up with restriction from low-power operating mode.In addition, can dispose the rearrangement that a plurality of requests are carried out, to optimize powering up and the number of times that cuts off the power supply to same storage medium.For the change number of times of the power rating that reduces storage medium, and strengthen power budget simultaneously, this is essential, and the serviceable life of storage medium has typically been shortened in the change of the power rating of storage medium.
In another embodiment of the present invention, can be at auxilliary storage system 112 configuring high speed caching mechanisms.Can utilize following mode to come configuring high speed caching mechanism: the file that will visit recently is cached on the first auxilliary storage medium that powers up all the time.This cache organization allows the faster visit of carrying out to often accessed data cell.Simultaneously, this cache organization has reduced frequently powering up and cut off the power supply the second auxilliary storage medium that is in low-power operating mode in preset time.
In addition, can be in auxilliary storage system 112 place configuration file archive device mechanisms.One or more data cells in being stored in the second auxilliary storage medium are during by frequent access, and this document archive device mechanism is divided into the particular-data unit group with these data cells.The above-mentioned one or more data cells that are stored in the second auxilliary storage medium that is in low-power operating mode can be cached on the first auxilliary storage medium, thereby minimize frequently powering up and cut off the power supply the second auxilliary storage medium.
In an embodiment, can also dispose and add electrical mechanisms, be used for based on to the search of metadata and the auxilliary storage medium that will be in low-power operating mode be transformed into from low-power operating mode and add power mode.Can dispose this in the following manner and add electrical mechanisms: before auxilliary storage system 112 places receive request at data cell, can change the power mode of auxilliary storage medium.This adds electrical mechanisms and can allow the second auxilliary storage medium need be optimized from the number of times that low-power operating mode begins to be powered.Yet, still can in being in the auxilliary storage medium of low-power operating mode, carry out search to the data unit.
Fig. 4 shows the block scheme that is used for filing system that the data unit is filed according to embodiment.The filing system 100 that is used to store data can comprise that file-archiver 402, network file system(NFS) (NFS) server 404, metadatabase (MDL) 406, network connect (network-attached) storage (NAS) high-speed cache 408, administration interface 410 and auxilliary storage system 112.File-archiver 402 can link to each other with nfs server 404 on function.Nfs server 404 can be visited the data file in the main storage system and is stored in metadata among the MDL406.File-archiver 402 can move or copy to NAS high-speed cache 408 from main storage system with data file.In an embodiment, NAS high-speed cache 408 can be stock (off-shelf) the NAS box (box) that is embedded into as high-speed cache in the filing system 108.File-archiver 402 can determine to be stored in the metadata of the data file in the NAS high-speed cache 406.This metadata can be stored among the MDL 406.In addition, file-archiver 402 can use the metadata and the data cell that are stored in the NAS high-speed cache 408 to search for.For example, the filing system 100 that is used for storing data can dispose and be used to obtain the title that auxilliary storage system 112 has all data cells of extension name ' .mpg '.In an embodiment, can implement the policy of deferring to (compliance policy) so that data cell is archived to auxilliary memory device 112 from NAS high-speed cache 408.In an embodiment, can dispatch, so that data cell is archived to auxilliary storage system 112 from NAS high-speed cache 408 data cell that is stored in the NAS high-speed cache 408.
In another embodiment, when finishing the archival task of data cell, can in NAS high-speed cache 408, create config directory from client 104 to auxilliary storage system 112.This config directory can have the information relevant with the structure that data cell has been filed.In addition, this config directory can comprise and optionally defers to configuration (compliance-configuration) data.Deferring to configuration data can specify the file structure that is associated with archival task and defer to policy.This config directory can also be used for multiple data unit more is archived to auxilliary storage system 112 from client 104 by file-archiver 402.
According to embodiment, the policy of deferring to can be created and manage to administration interface 410.The example of administration interface 410 comprises graphic user interface (GUI), Command Line Interface, UNIX command interface etc.The policy of deferring to can comprise defers to configuration or rule.The policy of deferring to can be stored in the NAS high-speed cache 408.The policy of deferring to can comprise a plurality of policy set, therefore can different policy set be applied to different data cell set based on user's preference.Deferring to the data service that the example of policy can be based in the network 104 comes the file of data unit is dispatched.
Fig. 5 shows the process flow diagram that is used for data cell is remained on the method for auxilliary storage system according to a plurality of embodiment.Data cell such as data file can be archived to auxilliary storage system 112 from main storage system.Auxilliary storage system 112 comprises the first auxilliary storage medium and the second auxilliary storage medium.The first auxilliary storage medium powers up all the time, and the second auxilliary storage medium is in low-power operating mode and can be powered based on needs simultaneously.In step 502, determine metadata at the one or more data files that are stored in the auxilliary storage system 112.One or more attribute that provides about the data cell of the information of data cell is provided this metadata.In an embodiment, can also comprise user definition information and versioned information in the metadata of data cell.
In step 504, the metadata store of data cell is being at least one storage medium (i.e. the first auxilliary storage medium) that adds power mode all the time.Be stored in one or more attribute in the metadata can also comprise be stored at least one the auxilliary storage medium (i.e. the second auxilliary storage medium) that is in low-power operating mode on the relevant information of data cell.In an embodiment, the filing system that is used to store data can receive and be used for inquiry that the data cell that is stored on the first and second auxilliary storage mediums is filed and obtained.This inquiry can be according to one or more attribute that can identify the data cell that is stored in the auxilliary storage system 112.In addition, one or more attribute that provides in this inquiry is used to provide the information about data cell.Even when the second auxilliary storage medium is in low-power operating mode, also can provide information at filing system 100 places that are used to store data about data cell.Can provide information in real time at filing system 100 places that are used to store data based on this inquiry about data cell.
Can also determine to be in one or more disk drive of low-power operating mode at filing system 100 places that are used to store data.Can also be appointed as and to be archived to the second auxilliary storage medium that is in low-power operating mode being stored in data cell on the client 104.In an embodiment, can determine to be in all the time first unallocated space of assisting in the storage medium that adds power mode.In addition, the storage of metadata can be based on determined, first unallocated space of assisting in the storage medium.For example, the unallocated space of determining the 20G byte on the first auxilliary storage medium that adds power mode can be in.Can be with in the metadata store of the 12G byte unallocated space on the first auxilliary storage medium.
In an embodiment, data file can be moved to the dish that another is subjected to power management from a dish that is subjected to power management, the file that perhaps will be read or obtain visit in the lump is divided into one group.Before reorientating data cell, metadata is upgraded to reflect the reposition of data cell, this makes that the mobile of data cell is sightless for the user.When the same data sheet tuple of frequent access, this measure can make the power consumption of auxilliary storage system 112 very effective.
Fig. 6 shows and is used to provide process flow diagram about the method for the information of data cell according to another embodiment.Can determine information by the inquiry that is used to obtain data cell in the data cell of the first and second auxilliary storage medium places storages of the filing system 100 of storage data.This inquiry can be at the data cell that is stored in the first auxilliary storage medium place that powers up all the time, or at the data cell that is stored in the second auxilliary storage medium place that is in low-power operating mode simultaneously.
In step 602, the user interface from client 102 receives inquiry.This inquiry can receive this request from GUI or Command Line Interface.In step 604, determine to be stored in first metadata of assisting in the storage medium that powers up all the time based on this inquiry.The filing system 100 that is used to store data can determine to be stored in the metadata of the data cell on the auxilliary storage system 112.This metadata can comprise and be stored in the second relevant information of data cell of assisting on the storage medium that is in low-power operating mode.
At step 606 place, one or more attribute of data file is used to provide the information about data cell.For example, can use the title of data cell and the size of data cell to determine to be stored in second data cell of assisting in the storage medium that is in low-power operating mode.Can use GUI or Command Line Interface that checking the data unit is provided.In addition, can use GUI or Command Line Interface to obtain and be stored in second data cell of assisting in the storage medium that is in low-power operating mode.In order to obtain data cell, may need the second auxilliary storage medium is begun to power up from low-power operating mode.
In an embodiment, GUI can also be used for by creating new metadata tree from main metadata tree (main metadatatree) xcopy.Can be according to the metadata of the data file of different tree constructions, the specified data unit check view.In this manner, data cell can be reorganized for the new view of checking, to serve particular demands.In this process, do not change main metadata tree.Can the metadata tree that each is new be rendered as the network file system(NFS) that is separated with main metadata tree, different restrict access can be disposed to the different views of checking thus.The view of checking that can present in an embodiment, the data cell that is stored in auxilliary storage system 112 places by the graphic user interface in the client 102 (GUI).
Fig. 7 shows the figure according to the scalable filing system of embodiment.At multiple user's request, may need filing system 108 upgrade (scale up).The example of user's request comprises the load that improves filing system 108, the speed that filing system 108 is carried out function etc.Based on multiple user's request, filing system 108 can upgraded aspect the Request Processing speed, for example improves the speed obtain data, data archiving, to inquire about etc.In addition, filing system 108 can increase its data storage capacity by using a plurality of memory devices.This scalable filing system can comprise a plurality of frames and a plurality of auxilliary storage rack such as first rack 708, second rack 710, the 3rd rack 712 such as first frame 702, second frame 704, the 3rd frame 706.In an embodiment, a plurality of frames and a plurality of rack can be positioned at the diverse geographic location place.
One or more file-archiver in one or more frame in a plurality of frames can be visited the metadata that is stored in the first auxilliary storage medium.This metadata can be stored in more than one, be in the auxilliary storage medium that adds power mode.This metadata comprises and is stored in the first and second relevant information of data cell of assisting in the storage mediums.The first auxilliary storage medium can power up all the time, and the second auxilliary storage medium can be in low-power operating mode simultaneously.
In a plurality of embodiment, can exist than frame shown in Figure 7 and rack is Duoed or few frame and rack.In a plurality of frames one or more can be implemented in the processor node such as server.Can adopt the Ethernet switch 714 of gigabit that first frame 702, second frame 704 and the 3rd frame 706 are linked to each other with network 102.Can a plurality of frames be linked to each other with a plurality of racks by FC switch 716.In a plurality of frames one or more can be visited one or more in a plurality of racks.Compare with the single frame of storage medium, a plurality of frames can provide bigger bandwidth.
In addition, filing system 108 performed work disposal can be distributed on a plurality of frames.For example, in filing system shown in Figure 7 108, work disposal can be distributed in first frame 702, second frame 704 and the 3rd frame 706.In an embodiment, can use in one of processor node the GUI that occurs or Command Line Interface to initiate to be used to store the task at filing system 100 places of data.This task can be the archival task of creating at being stored in the data cell on the auxilliary storage system 112 or obtain task.
Follow the new task of obtaining, processor node can be checked mailbox, to determine the busy extent of other processor node by the state of checking current effective task.In an embodiment, mailbox can be stored in the treatment facility (not shown among Fig. 7), and this treatment facility can be computing machine or server.This mailbox can be the storage system of frequent updating in the filing system 108.Then, the processor interface with new task is divided into the subtask with new task, and defines the node of underusing is distributed in the subtask by place the subtask in one or more mailboxes of other node.Node periodically monitors the progress of other node by sharing the status information that the mailbox location place checks other node.Stopping owing to node failure under the situation of task, one of other node can be responsible for this task, and can obtain finish this not complete operation entitlement or can restart the closing operation of appointing with the failure of irrecoverable form.
In an embodiment, by utilizing the priority orders of when processor node is installed in the document filing system 108, being distributed, perhaps alternatively, based on first module, obtained to indicate the proprietorial shared lock (lock) of task by arbitration scheme, can determine which processor node to take over task by.Therefore, the processor node behind the cluster provides upgradeable bandwidth when high availability (HA) structure is provided, and wherein the failure of single processor node can not cause task termination.
The number of the hard disk that the needs maintenance powers up can change along with the interpolation of metadata.When filing system 108 needs and the second relevant content-data of data cell of assisting on the storage medium that is stored in the operator scheme that is in low-power consumption if can being predicted, and when started the second auxilliary storage medium that is identified before visit.For example, if carry out search by the key word that use is stored in the metadata, and the hunting zone narrowed down in 100 or still less the result, then before this result's of visit visit, system can power up the second auxilliary storage medium that comprises with the corresponding data cell of this result.Power up can be automatically, by user's control or undertaken by other means.
In a plurality of embodiment of the present invention, can use different system architectures.For example, do not need to adopt frame/rack/module/equipment layout among Fig. 1.Can adopt suitable structure arbitrarily, utilize a plurality of features of embodiments of the invention.Here the concrete unit of the data of indication or type be only as example, and can replace with suitable data type or data volume arbitrarily.For example, although embodiments of the invention are described, can a part or the group or the out of Memory unit of file, piece, sector, dish will be applied to like the feature class of the present invention with respect to file management.Can use any type of content, for example image, audio frequency, executable program code, text, numeric data etc.
System described in the present invention or its random component can be implemented as the form of computer system.The typical case of computer system comprises multi-purpose computer, the microprocessor that is programmed, microcontroller, peripheral integrated circuit component and the miscellaneous equipment or the equipment layout that can realize the formation step of the inventive method.Function as described herein can be as required realizes with the form of hardware, software or both combinations.As required, can change other details of concrete programming language, statement, grammer or software or software description.
Although invention has been described with respect to specific embodiments of the invention, these embodiment are description of this invention and unrestricted.For example, it is evident that the occurrence of parameter can be with described here different with scope.
Although used the term such as " memory device ", " disk drive ", also can use the memory device that is suitable for any type of the present invention.For example, can also use disk drive, magnetic driven device etc.Can also use different existing and following memory technologies, those technology of creating such as using magnetic, solid-state, light, bioelectricity, nanometer engineering or other technologies.
Storage unit can be positioned at computer-internal or be positioned at the separate housing computing machine outside, that link to each other with computing machine.Other assembly in storage unit discussed herein, controller and the system can be included in single position or be dispersed in the diverse location place.These assemblies can interconnect by the arbitrarily suitable means such as network, communication link or other technologies.For example, although concrete function open to discussion for example operates in or is positioned at ad-hoc location and time place, can locate to provide function at diverse location and time usually.For example, can on sorter controller not at the same level, provide function such as the data protection step.Can use the RAID of any type to arrange or configuration.
In the description here, provide a large amount of details, the example of assembly and/or method for example is so that provide complete understanding to embodiments of the invention.Yet those skilled in the relevant art will recognize, under the situation of one or more in not having these details, also can implement the embodiment of the invention; Perhaps can use other device, system, composite set, method, assembly, material, parts and/or analog etc. to put into practice embodiments of the invention.In other cases, do not specify or describe in detail known structure, material or operation, to avoid the making aspect of the embodiment of the invention unclear.
" processor " or " processing " comprises anyone, hardware and/or software systems, mechanism or assembly of deal with data, signal or out of Memory.Processor can comprise having general CPU (central processing unit), a plurality of processing unit, be used to realize system or other system of the special circuit of function.Processing needn't be subject to the geographic position or have time restriction.For example, processor can " in real time ", " off-line ", carry out its function in modes such as " one-tenth batch modes ".In addition, can carry out specific processing section with the place by different (or identical) disposal system at different time.
In whole instructions, reference to " embodiment ", " embodiment " or " specific embodiment " is meant that described in conjunction with the embodiments concrete feature, structure or characteristic comprise at least one embodiment of the present invention, and needn't comprise in all embodiments.Therefore, these phrases are in the use at the diverse location place of whole instructions and do not mean that they necessarily relate to same embodiment's.In addition, can adopt arbitrarily suitable mode combines concrete feature, structure or characteristic in any specific embodiment of the present invention with one or more other embodiment.Should be understood that basis the benefit gained from others' wisdom here, describing also here, other variant, the modification of the embodiments of the invention of example are possible, and regard it part of the spirit and scope of the present invention as.
What it is also understood that is, the one or more elements described in the drawings/figures can perhaps be used according to concrete, even can remove as required under specific circumstances or present more to separate or integrated mode realizes, and becomes and can not operate.Realization can be stored in program on the machine readable media or code with allow computing machine carry out in the said method arbitrary method also within the spirit and scope of the present invention.
In addition, only the random signal arrows in the drawings/figures should be thought as exemplary, and unrestricted, unless otherwise indicated.In addition, employed here term " or " be intended to usually represent " and/or ", unless otherwise noted.Also the combination of assembly or step is considered as noting, wherein predict term or make separate or the ability of combination unclear.
As here in description and employed in whole claim subsequently, " one " and articles such as " one " and definite article comprise plural form, unless spell out in addition in the context.In addition, as here in description and employed in whole claim subsequently, " ... among " the meaning comprise " ... within " and " ... on ", unless spell out in addition in the context.
To the aforementioned description of illustration embodiment of the present invention (be included in summary described in) is not to be intended to exhaustive or the present invention is limited to precise forms disclosed herein.Those skilled in the relevant art will be familiar with and be understood that, here to the description of specific embodiments of the invention and example just to example, and various equivalent modifications also can be within the spirit and scope of the present invention.As shown here, according to aforementioned description, can carry out these to the present invention and revise, and these modifications all to comprise within the spirit and scope of the present invention example embodiment of the present invention.
Therefore, although invention has been described with reference to specific embodiment of the present invention, scope, various change and the replacement revised have been stipulated here aforementioned disclosing.To be understood that in some instances, will adopt some features of embodiments of the invention, and correspondingly do not use further feature, this does not deviate from scope and spirit of the present invention.Therefore, can carry out multiple modification, so that concrete condition or material are applicable to base region of the present invention and spirit.Stipulated that the present invention is not limited in the following claim employed specific project and/or is used to realize the disclosed specific embodiment of optimal mode of the present invention as imagination, the present invention can comprise all embodiment and the equivalent thereof that falls in the claims scope.
Claims (21)
1, a kind of auxilliary storage system is used to keep shift the data cell of coming from main storage system, and described auxilliary storage system comprises:
Auxilliary storage medium, wherein, described auxilliary storage medium is not all to be in simultaneously to add power mode, described auxilliary storage medium comprises and is at least one storage medium that adds power mode all the time; And
Metadata, be stored in described being in all the time among at least one storage medium that adds power mode one or more, wherein, described metadata comprises at least one attribute that is in the data cell in the auxilliary storage medium that at least one storage medium that adds power mode other are in low-power operating mode all the time except described.
2. auxilliary storage system according to claim 1 also comprises:
Administration interface is used to allow the user to check described metadata.
3. auxilliary storage system according to claim 1 also comprises:
Administration interface is used to allow the user to check Query Result, and wherein said inquiry is to use described metadata to obtain data from described auxilliary storage system.
4. auxilliary storage system according to claim 3, wherein, described metadata comprises the user definition information that is used to show described Query Result.
5. auxilliary storage system according to claim 3, wherein, described metadata comprises the versioned information that is used to show described Query Result.
6. auxilliary storage system according to claim 1 also comprises:
Administration interface is used to allow the user based on user's request of using described metadata, dynamically checks data in the storage system with the different tissues form
7. auxilliary storage system according to claim 1 also comprises:
File-archiver is used, be used for data cell is moved to second storage system from main storage system, wherein be stored in metadata in first storage system, make that be transparent for the visit of the data cell on the auxilliary storage system for the user of the data of first storage system by use.
8. auxilliary storage system according to claim 1, wherein, data cell comprises file.
9. auxilliary storage system according to claim 1 also comprises:
Administration interface, configuration is used to use described metadata video data unit on directory level.
10. auxilliary storage system according to claim 1 also comprises:
Rearrangement mechanism, configuration is used for according to rearranging described a plurality of request with reception at the first different order of second order of a plurality of requests of data cell, wherein, described first order allows a part in described a plurality of request sequentially to visit same storage media in the described auxilliary storage medium.
11. auxilliary storage system according to claim 10 wherein, than the situation of described a plurality of requests not being resequenced, limits powering up of described same storage media and cuts off the power supply the rearrangement of described a plurality of requests.
12. auxilliary storage system according to claim 1 also comprises:
Cache organization, configuration are used for data cell is cached at and describedly are in the storage medium that adds power mode all the time, so that visit faster.
13. auxilliary storage system according to claim 1 also comprises:
File-archiver mechanism, configuration is used for when the data cell of the storage medium of determining described auxilliary storage system is often visited in the lump described data cell being divided into one group.
14. auxilliary storage system according to claim 1 also comprises:
Add electrical mechanisms, configuration is used for being transformed into from low-power mode based on the auxilliary storage medium that the search to described metadata will be in low-power mode and adds power mode, wherein, before the request that receives at the data cell in the described auxilliary storage medium, change described auxilliary storage medium.
15. one kind is used for keeping from the method for the next data of main storage system transfer in the auxilliary storage system that comprises auxilliary storage medium, wherein, described auxilliary storage medium is not all to power up simultaneously to be in to add power mode, described auxilliary storage medium comprises and is at least one storage medium that adds power mode all the time that described method comprises:
Determine metadata at the one or more data cells in the auxilliary storage medium of described auxilliary storage system, wherein, described metadata comprises the attribute that is in the data cell at least one auxilliary storage medium that at least one storage medium that adds power mode other are in low-power mode all the time except described; And
Described metadata store is at least one storage medium that adds power mode all the time described, wherein, described attribute allow to determine and described at least one auxilliary storage medium that is in low-power mode in the relevant information of data cell.
16. method according to claim 15 also comprises:
Reception is from the inquiry at interface; And
Use the attribute of at least one data cell in one or more data cells that information about data cell is provided.
17. method according to claim 16, wherein, the user definition information that is used to provide about the information of data cell is provided described attribute.
18. method according to claim 17 also comprises at the data cell in the one or more data cells in described at least one storage medium that is in low-power mode, and the response to described inquiry is provided in real time.
19. method according to claim 16, wherein, the versioned information that is used to provide about the information of data cell is provided described attribute.
20. method according to claim 16 also comprises:
Determining which storage medium is in adds power mode; And
Described metadata store is in the storage medium that adds power mode determined.
21. method according to claim 16 also comprises:
Determine to be in and have how many unappropriated open spaces on the storage medium that adds power mode; And
Based on described unallocated space, determine the memory location of described metadata.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US72221505P | 2005-09-29 | 2005-09-29 | |
| US60/722,215 | 2005-09-29 | ||
| US60/730,288 | 2005-10-25 | ||
| US11/540,494 | 2006-09-28 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101501656A true CN101501656A (en) | 2009-08-05 |
Family
ID=40947440
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2006800312365A Pending CN101501656A (en) | 2005-09-29 | 2006-09-29 | System for archival storage of data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101501656A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488128A (en) * | 2019-12-30 | 2020-08-04 | 北京浪潮数据技术有限公司 | Method, device, equipment and medium for updating metadata |
| CN111552439A (en) * | 2020-04-24 | 2020-08-18 | 北京云宽志业网络技术有限公司 | Data storage method, device, system, electronic equipment and storage medium |
-
2006
- 2006-09-29 CN CNA2006800312365A patent/CN101501656A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488128A (en) * | 2019-12-30 | 2020-08-04 | 北京浪潮数据技术有限公司 | Method, device, equipment and medium for updating metadata |
| CN111488128B (en) * | 2019-12-30 | 2022-03-22 | 北京浪潮数据技术有限公司 | Method, device, equipment and medium for updating metadata |
| CN111552439A (en) * | 2020-04-24 | 2020-08-18 | 北京云宽志业网络技术有限公司 | Data storage method, device, system, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101743546B (en) | Hierarchical storage management for a file system providing snapshots | |
| US20070079086A1 (en) | System for archival storage of data | |
| CN101689129B (en) | File system mounting in a clustered file system | |
| CN103473250B (en) | For preserving the method and system of the past state of file system nodes | |
| US7546486B2 (en) | Scalable distributed object management in a distributed fixed content storage system | |
| US8738575B2 (en) | Data recovery in a hierarchical data storage system | |
| US6658589B1 (en) | System and method for backup a parallel server data storage system | |
| CN100419664C (en) | Incremental backup operations in storage networks | |
| CN1311358C (en) | Efficient search for migration and purge candidates | |
| CN100416508C (en) | Copy operations in storage networks | |
| US20070174580A1 (en) | Scalable storage architecture | |
| JP5722962B2 (en) | Optimize storage performance | |
| US20070220029A1 (en) | System and method for hierarchical storage management using shadow volumes | |
| CN1804810A (en) | Method and system of redirection for storage access requests | |
| US20080021902A1 (en) | System and Method for Storage Area Network Search Appliance | |
| JP2004062344A (en) | Method for destaging storage device system, disk control device, storage device system, and program | |
| CN101258497A (en) | A method for centralized policy based disk-space preallocation in a distributed file system | |
| CN101427251A (en) | Configurable views of archived data storage | |
| NO326041B1 (en) | Procedure for managing data storage in a system for searching and retrieving information | |
| EP1960918A2 (en) | Systems and methods for data management | |
| CN1770115A (en) | Recovery operations in storage networks | |
| CN101147118A (en) | Methods and apparatus for reconfiguring a storage system | |
| CN102165448A (en) | Storage tiers for database server system | |
| WO2003071429A1 (en) | Flexible and adaptive read and write storage system architecture | |
| CN104025058A (en) | Content selection for storage tiering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090805 |