Connect public, paid and private patent data with Google Patents Public Datasets

Backup data storage system comprising a memory and a management method thereof

Download PDF

Info

Publication number
CN103019966B
CN103019966B CN 201110282863 CN201110282863A CN103019966B CN 103019966 B CN103019966 B CN 103019966B CN 201110282863 CN201110282863 CN 201110282863 CN 201110282863 A CN201110282863 A CN 201110282863A CN 103019966 B CN103019966 B CN 103019966B
Authority
CN
Grant status
Grant
Patent type
Application number
CN 201110282863
Other languages
Chinese (zh)
Other versions
CN103019966A (en )
Inventor
简宏名
王哲人
彭怡华
Original Assignee
乔鼎资讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Grant date

Links

Abstract

本发明涉及一种数据储存系统及其管理方法。 The present invention relates to a data storage system and a management method. 所述数据储存系统包含N个储存装置、一备份内存以及一控制器,其中,N为一自然数。 The data storage system includes N storage means, a backup memory and a controller, wherein, N is a natural number. 每一储存装置包含一各自的写入高速缓存。 Each storage device comprises a respective write cache. 一旦该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内的数据。 Once the data storage system suffered a power outage, the backup data is stored in the memory is still stored within it. 所述控制器用以将从一存取要求单元传送的多笔数据执行一预定运算以产生多笔写入数据,传送这些写入数据至该N个储存装置的写入高速缓存,复制这些写入数据至该备份内存内,并且响应该N个储存装置传送一写入确认讯息将复制至该备份内存内的这些写入数据标示为”有效”。 The controller is configured from a multi-pen unit access request performs a predetermined operation is transmitted to generate a plurality of write data, the write data transfer to the write cache of the N storage device, copy the write the backup data into the memory, and in response store the N write device transmits a confirmation message will be copied to the backup data is written in the memory designated as "active."

Description

一种包含备份体内存的数据储存系统及其管理方法 Backup data storage system comprising a memory and a management method thereof

技术领域 FIELD

[0001] 本发明关于一种数据储存系统(data storage system)及其管理方法(managingmethod),尤其是针对符合磁盘阵列(redundant array of independent drives,RAID)架构的数据储存系统。 [0001] The present invention relates to a data storage system (data storage system) and a management method (managingmethod), especially for compliance with the disk array (redundant array of independent drives, RAID) architecture data storage system. 并且特别地,本发明数据储存系统包含备份内存,以避免数据储存系统遭受电力中断时欲写入的数据消失不见,并被保护。 And in particular, the data storage system of the present invention comprises a backup memory, a data storage system in order to avoid suffering the data to be written when the power interruption disappear, and protected.

背景技术 Background technique

[0002] 随着用户储存数据量日益增多,符合磁盘阵列(RAID)架构的数据储存系统(亦称为磁盘阵列(RAID)系统)已广泛地被采用来储存大量数据。 [0002] As the user data storage capacity increasing, in line with disk arrays (RAID) architecture of the data storage system (also known as disk arrays (RAID) systems) have widely been used to store large amounts of data. 磁盘阵列(RAID)系统能提供高可用性(highavailability)、高效能(high performance)或大容量(high volume)的数据储存空间给主机(host)。 Disk array (RAID) systems can provide high availability (highavailability), high performance (high performance) or bulk (high volume) data storage space to the host (host).

[0003] 磁盘阵列系统所利用的不同技术被称为RAID级别,也被区分为:RAID 0、RAID 1、RAID2、RAID 3, RAID 4, RAID 5以及RAID 6。 [0003] The disk array system utilized by different techniques is referred to as RAID levels, is also divided into areas: RAID 0, RAID 1, RAID2, RAID 3, RAID 4, RAID 5 and RAID 6. 每一种RAID级别技术各有各的优点及缺点。 Each RAID level techniques have advantages and disadvantages.

[0004] 现有的磁盘阵列系统的构成,包含一磁盘阵列控制器(RAID controller)以及由复数个磁盘驱动器(disk drive)所构成的磁盘阵列。 [0004] The configuration of a conventional disk array system comprising a disk array controller (RAID controller) and a plurality of disk array of a magnetic disk drive (disk drive) is constituted. 磁盘阵列控制器系连接至每一个磁盘驱动器,并由该磁盘阵列控制器将该磁盘阵列定义成RAID O, RAID 1、RAID 2、RAID 3、RAID 4、RAID 5或RAID 6之一个或多个逻辑磁盘驱动器(logical disk drive)而达成。 Based disk array controller is connected to each disk drive, the disk array controller is defined by the disk array into a RAID O, RAID 1, RAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 of one or more logical disk drive (logical disk drive) and reach.

[0005] 请参阅图1,一现有的符合一RAID架构之数据储存系统I的架构绘示于图1中。 [0005] Referring to FIG. 1, a conventional RAID architecture of infrastructure in line with a data storage system I is shown in FIG. 该数据储存系统I包含多个储存装置12以及一控制器10。 I The data storage system comprises a plurality of storage devices 12 and a controller 10.

[0006] 于一具体实施例中,该多个储存装置12系规范成一磁盘阵列架构,例如,RAID 0、RAIDU RAID 2、RAID 3、RAID 4、RAID 5 或RAID 6 等。 [0006] In one particular embodiment, the plurality of storage means 12 into a disk array system architecture specification, for example, RAID 0, RAIDU RAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 and so on.

[0007] 该控制器10系能产生(重建)与待读取数据相同的冗余数据(redundant data)。 [0007] The system controller 10 can generate the same (reconstructed) data to be read with the redundant data (redundant data). 例如,于RAID5架构下,该控制器10系透过互斥或(XOR)运算来产生冗余数据。 For example, under a RAID5 architecture, the system controller 10 to generate the redundant data through the exclusive-OR (XOR) operation.

[0008] 于实际应用中,每一个储存装置12可以是一磁带机(tape drive)、一磁盘驱动器(disk drive)、一内存装置(memory drive)、一光储存记录装置(optical storagedrive),或是于同一颗磁盘驱动器中对应于单一读写头之一个扇区,或是其它等效的储存 [0008] In practical applications, each of the storage device 12 may be a tape (tape drive), a magnetic disk drive (disk drive), a memory device (memory drive), an optical storage recording apparatus (optical storagedrive), or is a disk drive in the same corresponding to a single sector of the head, or other equivalent storage

目.ο Head .ο

[0009] 同样示于图1,该控制器10系连接至每一个储存装置12。 [0009] Also shown in FIG. 1, the system controller 10 is connected to each of the storage means 12. 图1并且绘示一存取要求单元(applicat1n I/O request unit) 2o该存取要求单元2经由一传输接口14连接至该控制器10。 And FIG. 1 illustrates a access request unit (applicat1n I / O request unit) 2o the access request unit 2 is connected to the controller 10 via a transmission interface 14. 于实际应用中,该存取要求单元2可以是网络计算机、迷你计算机、大型主机、笔记型计算机,或是需要读取数据储存系统I中之数据的任何电子设备,例如,手机、个人数字助理(personaldigital assistant,PDA)、数字录像设备、数字音乐播放器,等。 In practical applications, the access request unit 2 may be a network computer, a mini computer, mainframe, laptop computer, or any electronic device needs to read the data in the data storage system I, e.g., a mobile phone, a personal digital assistant (personaldigital assistant, PDA), digital video recording device, digital music player, and so on.

[0010] 当该存取要求单元2为一部独立的电子设备时,其可透过储存局域网络(SAN)、局域网络(LAN)、序列先进技术(serial ATA,SATA)接口、光纤信道(FC)、小型计算机标准接口(SCSI)等传输接口,或是PCI Express等输出入(I/O)接口与该数据储存系统I相连接。 [0010] When the access request is a separate unit 2 of the electronic device, which is transparent storage area network (SAN), Local Area Network (LAN), serial advanced technology (serial ATA, SATA) interface, a Fiber Channel ( the FC), small computer standard interface (SCSI) and other transmission interface, or PCI Express, input and output (I / O) interface I connected to the data storage system. 此外,当该存取要求单元2系一特殊集成电路组件,或是其它能够送出输出入读取要求的等效装置,其能依据其它装置之命令(或请求)而送出数据读取要求至该控制器10,进而透过该控制器10读取这些储存装置12中的数据。 Further, when the access unit 2 requires a special integrated circuit assembly line, or other equivalent means can be read out into the output requirements, which can be based on the command (or request) and the other device sends the data read request to the the controller 10, and thus the read data storage means 12 through the controller 10.

[0011] 该数据储存系统I中的该控制器10以及这些储存装置12可以同时安置于单一箱体内,也可以独立分开而安置于不同的箱体内。 [0011] The data storage system of the I controller 10 and the storage device 12 can be disposed simultaneously in a single housing, it may be disposed independently and separately on different cabinets. 实务上,该控制器10可以透过光纤信道(FC)、小型计算机系统接口(SCSI)、序列连接SCSI (SAS)、序列ATA(SATA)、平行ATA(PATA,或称IDE)等传输接口与这些储存装置12相连接。 In practice, the controller 10 may pass through a fiber channel (the FC), small computer system interface (SCSI), serial connection SCSI (SAS), serial ATA (SATA), parallel ATA (PATA, also known as IDE) interfaces, etc. and a transmission the storage device 12 is connected. 若这些储存装置12为磁盘驱动器的话,每一个储存装置12可以是FC、SAS、SCS1、SATA, PATA等各种不同接口型态的磁盘驱动器。 When the storage device 12 such as a disk drive devices, each storage device 12 may be FC, SAS, SCS1, SATA, PATA interfaces and other various patterns of the disk drive. 该控制器10系可为磁盘阵列控制器,或是具有能够产生冗余数据之数据储存系统的控制器。 The system controller 10 may be a disk array controller, or a controller having a data storage system capable of generating the redundant data is.

[0012] 同样示于图1,每一储存装置12包含一各自的写入高速缓存(write cache) 122。 [0012] Also shown in Figure 1, each storage device 12 comprises a respective write cache (write cache) 122. 每一个储存装置12被区分成多个数据区段(block)。 Each storage device 12 is divided into a plurality of segments of data (block) area. 从容错的观点来看,可分为使用者数据区段以及同位数据区段两类。 From the viewpoint of fault tolerance can be divided into user data sectors and parity data segment two. 使用者数据区段储存一般的使用者数据。 Usually the user data sectors stored user data. 同位数据区段则储存多余的一组同位数据,以当有容错需求时,供反算使用者数据之用。 With the bit data storage section with a set of redundant bits, to when fault tolerance requirements for the user data using the inverse operator. 存在于不同储存装置之相对应使用者数据区段与同位数据区段形成一个储存分割带(stripe),其中同位数据区段中的数据系由使用者数据区段中的数据执行互斥或(XOR)运算所得的结果。 Exist in different storage devices is formed with the same bit of the data section corresponding to a user data storage segment is divided band (stripe), wherein the data bits in the data section with the data from the system performs exclusive or user data sectors ( XOR) calculation results obtained. 须声明的是,本领域的普通技术人员皆了解,同位数据区段中之数据之计算,除可使用互斥或(XOR)运算外,也包含各式各样的同位(parity)运算或类似的运算技术,只要存在以下关系:相同储存分割带中之任一个数据区段中之数据可由其相对应数据区段之数据计算而得。 Represent that is, those of ordinary skill in the art are understood to calculate the parity data of the data segment, may be used in addition to exclusive or (XOR) operation, but also contains a variety of the same bit (Parity) calculation or the like computing technology, as long as the following relationship exists: dividing the data storage tape according to any of the same data section of a can therefrom corresponding to the data calculated from segment.

[0013] 使用者端欲将数据写入该数据储存系统I里,该控制器10经由该传输接口14接收该存取要求单元2传送的多笔数据。 [0013] user terminal to write data into the data storage system I, the multi-pen unit 2 of the data transfer controller 10 receives the transmission interface 14 via the access. 接着,该控制器10将这些接收到的数据执行一预定运算(例如,互斥或(XOR)运算)以产生多笔写入数据。 Next, the controller 10 performs the received data a predetermined operation (e.g., exclusive OR (XOR) operation) to generate a plurality of write data. 接着,该控制器10传送这些写入数据至该多个储存装置12的写入高速缓存122。 Next, the controller 10 transmits the write data to the plurality of write cache storage apparatus 12 122. 这些写入数据储存至该多个储存装置12之写入高速缓存122内后,该多个储存装置12回复该控制器10 —写入确认讯息。 These write data is stored to the plurality of storage devices 12 to write the cache 122, the plurality of storage devices 12 reply to the controller 10 - confirmation message is written. 此时,该控制器10即认定这些写入数据已经写入该多个储存装置12内,而不会保护这些数据。 At this time, i.e., the controller 10 finds the write data has been written to the plurality of storage devices 12, but does not protect the data. 但是,储存在该多个储存装置12之写入高速缓存122内的这些写入数据未必全数已写入该多个储存装置12本身的记录媒体(media)内。 However, stored in the plurality of storage devices 12 to write the write data in the cache 122 may not be fully written recording medium 12 (Media) itself of the plurality of storage devices. 一旦该数据储存系统I遭受不预期的电力中断时,仍储存在该多个储存装置12之写入高速缓存122的这些写入数据即会消失不见。 Once the data storage system I suffer not expected power outages, still stored in the plurality of storage devices write 12 of these write data cache 122, that will disappear.

发明内容 SUMMARY

[0014] 因此,本发明提供一种数据储存系统及其管理方法。 [0014] Accordingly, the present invention provides a data storage system and a management method. 并且特别地,本发明之数据储存系统包含备份内存,以避免数据储存系统遭受电力中断时欲写入的数据消失不见,并被保护。 And in particular, the data storage system of the present invention comprises a backup memory, a data storage system in order to avoid suffering the data to be written when the power interruption disappear, and protected.

[0015] 根据本发明之一较佳具体实施例的数据储存系统,其包含N个储存装置、备份内存以及一控制器,其中N为一自然数。 [0015] The data storage system of the present invention, one of preferred embodiments, which contains N storage means, the backup memory and a controller, wherein N is a natural number. 每一储存装置包含一各自的写入高速缓存。 Each storage device comprises a respective write cache. 一旦该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内之数据。 Once the data storage system suffered a power outage, the backup data is stored in the memory is still stored within it. 该控制器用以接收一存取要求单元传送的多笔数据,将这些接收到的数据执行一预定运算以产生多笔写入数据,传送这些写入数据至该N个储存装置之写入高速缓存,复制这些写入数据至该备份内存内,并且响应该N个储存装置传送一写入确认讯息将复制至该备份内存内的这些写入数据标示为有效。 The controller is configured to receive an access request transfer unit is a multi-pen, the implementation of these received data to a predetermined operation generate a plurality of write data, the write data transfer to the writing of the N cache storage means , copy the backup data is written into the memory, and in response store the N write device transmits a confirmation message will be copied to the backup data is written in the memory designated as active. 并且,当该控制器判断该备份内存之储存空间低于一门坎值或复制这些写入数据至该备份内存内之地址重复时,该控制器发布一清理快取命令(flush cachecommand)至该N个储存装置,并且响应该N个储存装置传送一清理快取完成讯息清除储存于该备份内存内之写入数据。 Further, when the controller determines that the storage space of the backup memory is below a threshold value or to copy the write data of the address in the backup memory when repeated, the cache controller issues a command to clean (flush cachecommand) to the N a storage means, and storage means in response to the N transmitting a clear complete message cleanup cache the write data stored in the memory of the backup.

[0016] 于一具体实施例中,当该数据储存系统开机时,该控制器判断该备份内存内是否存有标示有效的写入数据。 [0016] In one particular embodiment, when the data storage system boot, the controller determines whether there marked valid write data in the backup memory. 若判断结果为肯定,该控制器将该备份内存内标示有效的写入数据写入该N个储存装置。 If the determination result is affirmative, the controller sets the designated backup memory valid write data of the N storage means. 并且,在完成将这些标示有效的写入数据写入该N个储存装置之后,该控制器清除储存于该备份内存内的写入数据。 And, after completion of these marked valid write data of the N storage means, the controller clears the write data stored in the backup memory.

[0017] 于一具体实施例中,该预定运算可以是一互斥或(XOR)运算、一镜像运算(mirroringoperat1n)、一无运算(NO operat1n)、一切分运算(slice operat1n)、一有限加洛瓦场(Galois field, GF)本质多项式运算、一有限场的数学最大距离可分码(maximum distance separable code)运算、其它同位(parity)运算或类似的运算技术。 [0017] In one particular embodiment, the predetermined operation may be an exclusive or (XOR) operation, a mirroring operation (mirroringoperat1n), a no operation (NO operat1n), all operation points (slice operat1n), plus a limited Luwa field (Galois field, GF) polynomial operation essence, a finite mathematical field maximum distance separable codes (maximum distance separable code) operation, with the other bit (Parity) calculation, or similar computing technology.

[0018] 于一具体实施例中,该备份内存系一非挥发性内存或一配置一独立电源之挥发性内存。 [0018] In one particular embodiment, the backup memory system or a non-volatile memory, a volatile memory configuration of an independent power source.

[0019] 于一具体实施例中,该备份内存系仅电连接至该控制器。 [0019] In one particular embodiment, the backup memory only lines electrically connected to the controller. 于另一具体实施例中,该备份内存系该控制器内之一写入高速缓存。 In another particular embodiment, the controller is within one line of the write cache memory backup. 于另一具体实施例中,该备份内存系电连接在该控制器与该N个储存装置之间。 In another particular embodiment, the backup memory is electrically coupled between the system controller and the N storage means.

[0020] 根据本发明之一较佳具体实施例之管理方法针对一数据储存系统之管理方法。 The method for managing a management method for a data storage system of [0020] According to one embodiment of the present invention is particularly preferred. 该数据储存系统包含多个储存装置以及一备份内存。 The data storage system comprises a plurality of storage devices and a backup memory. 每一储存装置包含一各自的写入高速缓存。 Each storage device comprises a respective write cache. 当该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内的数据。 When the data storage system suffered a power outage, the backup data is stored in the memory is still stored within it. 根据本发明之较佳具体实施例的管理方法,首先,接收一存取要求单元传送的多笔数据。 Management method according to a preferred embodiment of the present invention, first, an access request receiving unit sends the multi-pen. 接着,根据本发明之管理方法将这些接收到的数据执行一预定运算以产生多笔写入数据。 Subsequently, these received data performs a predetermined calculation based on the management method of the present invention to generate a plurality of write data. 接着,根据本发明之管理方法判断该备份内存的储存空间是否低于一门坎值,或判断复制这些写入数据至该备份内存内之地址是否重复。 Next, according to the management method of the present invention determines whether the backup memory storage space is below a threshold value, or determines the address to copy the write data within the backup memory is repeated. 若判断结果为否定者,根据本发明之管理方法则传送这些写入数据至该N个储存装置的写入高速缓存,并复制这些写入数据至该备份内存内。 If the determination result is negative persons, according to which the write data is transmitted to the management method of the present invention the N write cache storage means, and to copy the write data into the backup memory. 并且,根据本发明的管理方法响应该N个储存装置传送一写入确认讯息,将复制至该备份内存内的这些写入数据标示为有效。 And, means for transmitting the N storing a write acknowledgment message management method according to the present invention in response to these write copy the data in the backup memory marked as valid.

[0021] 若该备份内存之储存空间低于一门坎值或复制这些写入数据至该备份内存内之地址重复,根据本发明之管理方法则发布一清理快取命令至该N个储存装置。 [0021] If the storage space of the backup memory is below a threshold value or copy the write data into the backup memory address within the repeat, the cache issues a command to the cleaning of the N storage device based on the management method of the present invention. 并且,根据本发明的管理方法响应该N个储存装置传送一清理快取完成讯息,清除储存于该备份内存内的写入数据。 Further, the N storage means to transmit a complete message cleanup cache management method according to the present invention in response, clears the write data stored in the backup memory.

[0022] 当该数据储存系统开机时,根据本发明之管理方法判断该备份内存内是否存有标示有效的写入数据。 [0022] When the data storage system boot, according to the management method of the present invention determines whether there marked valid write data in the backup memory. 若判断结果为肯定者,根据本发明的管理方法则将该备份内存内标示有效的写入数据写入该N个储存装置,并且在完成将这些标示有效的写入数据写入该N个储存装置之后,清除储存于该备份内存内的写入数据。 If the determination result is affirmative who designated the backup memory inside the valid write data of the N storage device management method according to the present invention, and upon completion of those marked valid write data storage of the N after means clearing the write data stored in the backup memory.

[0023] 关于本发明之优点与精神可以藉由以下的实施方式及所附图式得到进一步的了解。 [0023] With regard to the advantages of the present invention may be by the spirit of the following embodiments and the accompanying drawings provide further understanding.

附图说明 BRIEF DESCRIPTION

[0024] 下面结合附图对本发明作进一步说明: [0024] DRAWINGS The present invention is further described:

[0025]图1为一现有的符合RAID架构的数据储存系统的架构示意图。 [0025] FIG. 1 is a diagram of a data storage system architecture of a conventional RAID architectures meet.

[0026]图2为根据本发明一较佳具体实施例数据储存系统的架构示意图。 [0026] FIG. 2 is a schematic diagram of architecture of a data storage system in accordance with one preferred embodiment of the present invention.

[0027]图3为根据本发明较佳具体实施例数据储存系统其备份内存之一变形的示意图。 [0027] FIG. 3 is a schematic view of one embodiment of memory, its backup data storage system modified according to the particular preferred embodiment of the present invention.

[0028]图4为根据本发明较佳具体实施例数据储存系统其备份内存之另一变形的示意图。 [0028] FIG. 4 is a schematic diagram of another embodiment of the memory which backup data storage system modified according to the particular preferred embodiment of the present invention.

[0029] 图5为根据本发明一较佳具体实施例管理方法的流程图。 [0029] FIG. 5 is a flowchart of the management method according to a preferred particular embodiment of the present invention.

[0030] 图6为根据本发明之较佳具体实施例管理方法的另一流程图。 [0030] FIG 6 is a flowchart illustrating another embodiment according to the management method particularly preferred embodiment of the present invention.

[0031] 元件符号说明: [0031] Description of Symbols elements:

[0032]【主要组件符号说明】 [0032] The main component symbol DESCRIPTION

[0033] 1、3:数据储存系统 10、30:控制器 [0033] 1,3: data storage system 10, 30: Controller

[0034] 12,32:储存装置 122、322:写入高速缓存 [0034] 12, 32: storage device 122, 322: Write Cache

[0035] 14、34:传输接口 36:备份内存 [0035] 14, 34: transmission interface 36: the backup memory

[0036] 2、4:存取要求单元 5:管理方法 [0036] 2,4: management: access request unit 5

[0037] S50〜S70:流程步骤 [0037] S50~S70: Process Steps

具体实施方式 detailed description

[0038] 本发明提供一种数据储存系统及其管理方法。 [0038] The present invention provides a system for data storage and management. 并且特别地,根据本发明数据储存系统包含备份内存,以避免数据储存系统遭受电力中断时欲写入的数据消失不见,并被保护。 And in particular, the data storage system according to the present invention comprises a backup memory, a data storage system in order to avoid suffering the data to be written when the power interruption disappear, and protected. 以下藉由对本发明的较佳具体实施例的详细说明,藉以充分解说关于本发明的特征、精神、优点以及实施上的可行性。 The following detailed description of preferred by the specific embodiment of the present invention, in order to fully explain, spirit, advantages, and embodiments regarding the feasibility of the present invention features.

[0039] 请参阅图2,根据本发明之一较佳具体实施例的数据储存系统3的架构绘示于图2中。 [0039] Referring to FIG 2, a data storage system in accordance with one embodiment of the present invention, particularly preferred embodiment of the structure 3 shown in FIG. 该数据储存系统3包含N个储存装置32、一控制器30以及一备份内存36,其中N为一自然数(1,2,3,……)。 The data storage system 3 comprises N storage means 32, a controller 30 and a backup memory 36, where N is a natural number (1,2,3, ......).

[0040] 于一具体实施例中,该N个储存装置32规范成一磁盘阵列架构,例如,RAID 0、RAID URAID 2、RAID 3、RAID 4、RAID 5 或RAID 6 等。 [0040] In one particular embodiment, the N storage means 32 into a disk array architecture specification, for example, RAID 0, RAID URAID 2, RAID 3, RAID 4, RAID 5 or RAID 6 and so on.

[0041] 该控制器30系能产生(重建)与待读取数据相同的冗余数据。 [0041] The system controller 30 can generate (reconstruct) the same data as redundant data to be read. 例如,于RAID 5架构下,该控制器30系透过互斥或(XOR)运算来产生冗余数据。 For example, in RAID 5 architecture, the system controller 30 to generate the redundant data through the exclusive-OR (XOR) operation.

[0042] 于一具体实施例中,每一个储存装置32可以是一磁带机、一磁盘驱动器、一内存装置、一光储存记录装置,或是于同一颗磁盘驱动器中对应于单一读写头之一个扇区,或是其它等效的储存装置。 [0042] In one particular embodiment, each storage device 32 may be a tape drive, a disk drive, a memory device, an optical storage recording apparatus, or corresponding to a single head on the same a disk drive in the a sector, or other equivalent storage device.

[0043] 同样示于图2,该控制器30连接至每一个储存装置32。 [0043] Also shown in FIG. 2, the controller 30 is connected to each of the storage device 32. 图2并且绘示一存取要求单元4。 And FIG. 2 shows an access unit in claim 4. 该存取要求单元4经由一传输接口34连接至该控制器30。 The access request unit 4 is connected to the controller 30 via a transmitting interface 34. 于实际应用中,该存取要求单元4可以是网络计算机、迷你计算机、大型主机、笔记型计算机,或是需要读取数据储存系统3中之数据的任何电子设备,例如,手机、个人数字助理、数字录像设备、数字音乐播放器,等。 In practical applications, the access request unit 4 may be a network computer, a mini computer, mainframe, laptop computer, or a need to read any data storage system data 3 of the electronic device, e.g., a mobile phone, a personal digital assistant , digital video recording device, digital music player, and so on.

[0044] 当该存取要求单元4为一部独立的电子设备时,其系可透过储存局域网络(SAN)、局域网络(LAN)、序列先进技术(serial ATA,SATA)接口、光纤信道(FC)、小型计算机标准接口(SCSI)等传输接口,或是PCI Express等输出入(I/O)接口与该数据储存系统3相连接。 [0044] When the access request unit 4 is a separate electronic device through the storage area network may be based (SAN), Local Area Network (LAN), serial advanced technology (serial ATA, SATA) interface, a Fiber Channel (the FC), small computer standard interface (SCSI) and other transmission interface, or PCI Express, input and output (I / O) interface 3 connected to the data storage system. 此外,当该存取要求单元4为一特殊集成电路组件,或是其它能够送出输出入读取要求的等效装置,其能依据其它装置之命令(或请求)而送出数据读取要求至该控制器30,进而透过该控制器30读取这些储存装置32中的数据。 Further, when the access request unit 4 is a special integrated circuit components, or other equivalent means can be read out into the output requirements, which can be based on the command (or request) and the other device sends the data read request to the the controller 30, and thus the read data storage means 32 through the controller 30.

[0045] 该数据储存系统3中的该控制器30以及这些储存装置32可以同时安置于单一箱体内,也可以独立分开而安置于不同的箱体内。 [0045] The data storage system 3 of the controller 30 and the storage device 32 can be disposed simultaneously in a single housing, it may be disposed separately and independently in different cabinets. 实务上,该控制器30可以透过光纤信道(FC)、小型计算机系统接口(SCSI)、序列连接SCSI (SAS)、序列ATA(SATA)、平行ATA(PATA,或称IDE)等传输接口与这些储存装置32相连接。 In practice, the controller 30 may pass through a fiber channel (the FC), small computer system interface (SCSI), serial connection SCSI (SAS), serial ATA (SATA), parallel ATA (PATA, also known as IDE) interfaces, etc. and a transmission the storage device 32 is connected. 若这些储存装置32为磁盘驱动器的话,每一个储存装置32可以是FC、SAS、SCS1、SATA, PATA等各种不同接口型态的磁盘驱动器。 When the storage device 32 is a disk drive devices, each storage device 32 may be FC, SAS, SCS1, SATA, PATA interfaces and other various patterns of the disk drive. 该控制器30系可为磁盘阵列控制器,或是具有能够产生冗余数据之数据储存系统的控制器。 The system controller 30 may be a disk array controller, or a controller having a data storage system capable of generating the redundant data is.

[0046] 同样示于图2,每一储存装置32包含一各自的写入高速缓存322。 [0046] Also shown in FIG. 2, each storage device 32 comprises a respective write cache 322. 每一个储存装置32被区分成多个数据区段。 Each storage device 32 is divided into a plurality of segments of data regions. 从容错的观点来看,可分为使用者数据区段以及同位数据区段两类。 From the viewpoint of fault tolerance can be divided into user data sectors and parity data segment two. 使用者数据区段储存一般的使用者数据。 Usually the user data sectors stored user data. 同位数据区段则储存多余的一组同位数据,以当有容错需求时,供反算使用者数据之用。 With the bit data storage section with a set of redundant bits, to when fault tolerance requirements for the user data using the inverse operator. 存在于不同储存装置之相对应使用者数据区段与同位数据区段形成一个储存分割带,其中同位数据区段中的数据系由使用者数据区段中的数据执行互斥或(XOR)运算所得的结果。 The presence of different storage means corresponding to the user data and parity data section forms a storage section of the divided bands, with the data bits in the data section of the data-based user data section performs exclusive OR (XOR) operation the results obtained. 须声明的是,本领域的普通技术人员皆了解,同位数据区段中的数据之计算,除可使用互斥或(XOR)运算外,针对各种RAID架构,也可以是镜像运算、无运算(NO operat1n)、切分运算、有限加洛瓦场(Galois field,GF)本质多项式运算、有限场的数学最大距离可分码(maximum distance separable code)运算、其它同位运算或类似的运算技术,只要存在以下关系:相同储存分割带中之任一个数据区段中之数据可由其相对应数据区段之数据计算而得。 Is to be stated, those of ordinary skill in the art are understood, the parity data is calculated in the data section, can be used in addition to exclusive or (XOR) operation, the RAID architectures for various, but may be a mirror operation, no operation (NO operat1n), segmentation operation, Gallois finite field (Galois field, GF) polynomial operation nature, the maximum distance separable code mathematical finite field (maximum distance separable code) operation, with the other bit operations, or similar computing technology, As long as the following relationship: dividing the data storage tape according to any of the same data section of a can from its corresponding segment of the data calculated from data.

[0047] 使用者端欲将数据写入该数据储存系统3里,该控制器30经由该传输接口34接收该存取要求单元4传送的多笔数据。 [0047] user terminal to write data into the data in the storage system 3, the controller 30 receives the access request transmitted plurality pen unit 4 via the transmission interface 34. 接着,该控制器30将这些接收到的数据执行一预定运算(例如,互斥或(XOR)运算)以产生多笔写入数据。 Next, the controller 30 performs the received data a predetermined operation (e.g., exclusive OR (XOR) operation) to generate a plurality of write data. 接着,该控制器30传送这些写入数据至该N个储存装置32之写入高速缓存322。 Next, the controller 30 transmits the write data is written to the storage device 32 of the N cache 322. 特别地,该控制器30复制这些写入数据至该备份内存36内。 In particular, the controller 30 writes data to the copy in the backup memory 36.

[0048] 这些写入数据储存至该N个储存装置32之写入高速缓存322内后,该N个储存装置32回复该控制器30 —写入确认讯息。 After [0048] The write data is stored to the N storage means 32 of the write cache 322, the N storage devices 32 reply to the controller 30 - Write the confirmation message. 此时,该控制器30响应该N个储存装置32传送之写入确认讯息将复制至该备份内存36内之这些写入数据标示为有效,仍保护这些写入数据。 At this time, the write controller 30 in response to the confirmation message 32 transmits the N storage means to the copy of the written data in the backup memory 36 is marked valid, the write data is still protected. 但是,储存在该N个储存装置32之写入高速缓存322内之这些写入数据未必全数已写入该N个储存装置32本身的记录媒体内。 However, the N is stored in the storage device 32 writes the write data of the cache 322 may not be within full recording medium 32 have been written into the storage device of the N itself.

[0049] 并且,当该控制器30判断该备份内存36之储存空间低于一门坎值或复制这些写入数据至该备份内存36内之地址重复时,该控制器30发布一清理快取命令至该N个储存装置32。 [0049] Then, when the controller 30 determines that the backup memory 36 of the storage space falls below a threshold value or copy the write data into the address memory 36 within the backup when repeated, the controller 30 issues a command to cache clean-up to the N storage means 32. 该N个储存装置32收到该清理快取命令后,即会将这些写入高速缓存322内的写入数据写入该N个储存装置32本身的记录媒体内。 After the N storage means 32 receives the command cache cleanup, i.e., the write data will be written in the cache 322 is written in the recording medium 32 of the N storage device itself. 这些高速缓存322内的这些写入数据全数写入该N个储存装置32本身的记录媒体后,该N个储存装置32即回复该控制器30 —清理快取完成讯息。 These data is written in these full write cache 322 of the N recording medium 32 storing the device itself, the N storage means 32 i.e., the reply controller 30 - preparation complete message cache. 此时,该控制器30即响应该清理快取完成讯息,清除储存于该备份内存36内的写入数据。 At this time, the cleaning controller 30 that is responsive to the cache complete message, clearing the write data stored in the backup memory 36.

[0050] 特别地,一旦该数据储存系统3遭受不预期的电力中断时,该备份内存36仍保存储存在其内之数据。 [0050] In particular, once the data storage system 3 are not expected to suffer a power interruption, the backup memory 36 is still stored in the data which is stored within. 当该数据储存系统3开机时,该控制器30判断该备份内存36内是否存有标示有效的写入数据。 When the boot data storage system 3, the controller 30 determines that the designated backup memory 36 whether there is a valid write data. 若判断结果为肯定,该控制器30将该备份内存36内标示有效的写入数据写入该N个储存装置32。 If the determination result is affirmative, marked valid write data storage device 32 of the N within the backup memory 30 of the controller 36. 并且,在完成将这些标示有效的写入数据写入该N个储存装置32之后,该控制器30清除储存于该备份内存36内的写入数据。 Further, the completion of these marked valid write data of the N 32 after storage devices, the controller 30 clears the write data stored in the backup memory 36. 藉此,根据本发明之数据储存系统3可以避免遭受电力中断时欲写入的数据消失不见,并且确保欲写入的数据写入该N个储存装置32内。 Accordingly, the data to be written when a power interruption data disappear according to the present invention suffered storage system 3 can be avoided, and to ensure that the data to be written is written in the storage device 32 of the N.

[0051] 于一具体实施例中,该备份内存36为一非挥发性内存(如Flash Memory),或是一配置一独立电源(例如,电池或不断电系统)之挥发性内存。 [0051] In one particular embodiment, the backup memory 36 is a non-volatile memory (e.g. Flash Memory), a separate power supply or a configuration (e.g., a battery or uninterruptible power supply) of the volatile memory.

[0052] 于一具体实施例中,如图2所示,该备份内存36仅电连接至该控制器30。 [0052] In one particular embodiment, shown in Figure 2, the backup memory 36 is only electrically connected to the controller 30.

[0053] 于另一具体实施例中,如图3所不,该备份内存36系该控制器30内之一写入高速缓存。 [0053] In another particular embodiment, not shown in FIG 3, the backup memory 36 of the system 30 one of the write cache controller. 图3中具有与图2相同号码标记的组件,有相同或类似的功能以及运作,在此不做赘述。 Numerals in FIG. 3 having the same number of components of FIG. 2, the same or similar function and operation, and is not repeated herein.

[0054] 于另一具体实施例中,如图4所示,该备份内存36为电连接在该控制器30与该N个储存装置32之间。 [0054] In another particular embodiment, shown in Figure 4, the backup memory 36 is electrically connected between the controller 30 and the N storage means 32. 图4中具有与图2相同号码标记的组件,有相同或类似的功能以及运作,在此不做赘述。 In FIG. 4 components having the same number of numerals in FIG. 2, the same or similar function and operation, and is not repeated herein.

[0055] 请参阅图5,图5为绘示根据本发明之一较佳具体实施例之管理方法2的流程图。 [0055] Referring to FIG. 5, FIG. 5 is a schematic flowchart illustrating the management method of Example 2 according to one embodiment of the present invention is particularly preferred. 根据本发明的管理方法5系针对例如图2、图3或图4所示的数据储存系统3的管理方法。 For example 2, the data storage system shown in FIG. 3 or FIG. 4 according to the management method for the three-based management method of the present invention. 该数据储存系统3的架构已于上文中详述,在此不再赘述。 The data storage system architecture 3 was detailed above, are not repeated here.

[0056] 如图5所示,根据本发明的管理方法5首先执行步骤S50,接收一存取要求单元4传送的多笔数据。 [0056] As shown in FIG 5, according to the management method of the present invention 5 first executes step S50, the access request receiving unit 4 transmits a plurality of pieces of data.

[0057] 接着,根据本发明的管理方法5执行步骤S52,将这些接收到的数据执行一预定运算,以产生多笔写入数据。 [0057] Next, the management method of the present invention 5 executes step S52, the received data to perform a predetermined operation, to generate a plurality of write data.

[0058] 于一具体实施例中,该预定运算可以是一互斥或(XOR)运算、一镜像运算、一无运算、一切分运算、一有限加洛瓦场(Galois field, GF)本质多项式运算、一有限场的数学最大距离可分码运算、其它同位运算或类似的运算技术。 [0058] In one particular embodiment, the predetermined operation may be an exclusive or (XOR) operation, an image operation, a non-operation of all sub-operation, a finite nature Gallois field (Galois field, GF) polynomial operations, mathematical computing the maximum distance Separation code is a finite field, or other similar computing parity calculation techniques.

[0059] 接着,根据本发明的管理方法5执行步骤S54,判断该备份内存36之储存空间是否低于一门坎值。 [0059] Next, the management method of the present invention step 5 S54, determines whether or not the backup memory 36 of the storage space falls below a threshold value.

[0060] 若步骤S54的判断结果为否定者,根据本发明的管理方法5执行步骤S56,判断复制这些写入数据至该备份内存36内的地址是否重复。 [0060] If the determination result in step S54 is negative persons, according to the management method of the present invention is performed 5 step S56, the write data is determined to copy the address in the backup memory 36 whether to repeat.

[0061] 若步骤S56之判断结果为否定者,根据本发明的管理方法5执行步骤S58,传送这些写入数据至该N个储存装置32的写入高速缓存322,并复制这些写入数据至该备份内存36内。 [0061] If the result of decision at step S56 is negative persons, according to the management method of the present invention is 5 executes step S58, the write data to the write transfer of the N storage devices 322 cache 32, and writes data to the copy the backup memory 36.

[0062] 在步骤S58之后,根据本发明的管理方法5执行步骤S60,响应该N个储存装置32传送一写入确认讯息,将复制至该备份内存36内的这些写入数据标示为有效。 Transfer [0062] After the step S58, the management method according to the present invention is 5 executes step S60, the response of the N storage means 32 a write confirmation message, the data is written to the copy in the backup memory 36 is marked valid.

[0063] 若步骤S54之判断结果为肯定者,或步骤S56之判断结果为肯定者,根据本发明之管理方法5执行步骤S62,发布一清理快取命令至该N个储存装置32。 [0063] If the determination in step S54 is affirmative result of those, or a determination in step S56 is affirmative result of those, according to the management method of the present invention performing step S62, the cache issues a command to the cleaning of the N storage means 32. 此时,这些高速缓存322内的这些写入数据全数写入该N个储存装置32本身的记录媒体后,该N个储存装置32即回复一清理快取完成讯息。 At this time, the data is written in the write cache 322 full of the N recording medium 32 storing the device itself, the N storage means 32 i.e. a reply cache cleanup complete message.

[0064] 在步骤S62之后,根据本发明的管理方法5执行步骤S64,响应该N个储存装置32传送之该清理快取完成讯息,清除储存于该备份内存36内的写入数据。 [0064] After the step S62, the management method of the present invention step 5 S64, in response to the transfer complete message cache cleanup of the N storage means 32, clear the write data stored in the backup memory 36. 在步骤S64之后,即完成程序。 After the step S64, the program is completed.

[0065] 请参阅图6,当该数据储存系统3开机时,根据本发明的管理方法5进一步执行步骤S66,判断该备份内存36内是否存有标示有效的写入数据。 [0065] Referring to FIG 6, when the data storage system boot 3, 5 further performs step S66 management method according to the present invention, it is determined that the designated backup memory 36 whether there is a valid write data.

[0066] 若步骤S66的判断结果为肯定者,根据本发明的管理方法5进一步执行步骤S68,将该备份内存36内标示有效的写入数据写入该N个储存装置32。 [0066] If the determination result of step S66 is affirmative were further perform the management method of the present invention 5 step S68, the backup memory 36 designated within the valid write data of the N storage means 32. 在步骤S68之后即在完成将这些标示有效的写入数据写入该N个储存装置之后,根据本发明的管理方法5执行步骤S70,清除储存于该备份内存36内的写入数据。 After the step S68 after the completion of these marked i.e. valid write data of the N storage device management method according to the present invention is 5 executes step S70, the write data stored in the clear of the backup memory 36. 藉此,根据本发明的管理方法5可以避免该数据储存系统3遭受电力中断时欲写入的数据消失不见,并且确保欲写入的数据写入该N个储存装置32内。 Accordingly, the data storage system 3 to be subjected to the data writing disappear when power is interrupted, and the write data to be written to ensure that the N storage means 32 based on the management method of the present invention 5 can be avoided.

[0067] 藉由以上本发明之较佳具体实施例的详细说明,能清楚了解本发明所提供的数据储存系统及其管理方法,能避免数据储存系统遭受电力中断时欲写入的数据消失不见,并被保护。 [0067] The detailed description of the present invention by the above particular preferred embodiment, a clear understanding of the data storage system and its management method of the present invention provides, to avoid the data storage system from the data to be written when the power interruption disappear , and protected.

[0068] 藉由以上较佳具体实施例之详述,希望能更加清楚描述本发明之特征与精神,而并非以上述所揭露的较佳具体实施例来对本发明之范畴加以限制。 [0068] Detailed Description of the preferred embodiment of the above particular embodiments by, hoping to more clearly describe the characteristics and spirit of the invention, and are not disclosed in the above-described specific preferred embodiments to limit the scope of the invention. 相反地,其目的是希望能涵盖各种改变及具相等性的安排于本发明所欲申请的专利范围的范畴内。 Conversely, its purpose is to be able to cover various modifications within the scope of the patent and with equal scope of the present invention arranged in the desired application. 因此,本发明所申请的专利范围的范畴应该根据上述的说明作最宽广的解释,以致使其涵盖所有可能的改变以及具相等性的安排。 Accordingly, the scope of the scope of the patent application of the present invention should be the broadest interpretation of the above description, so as to encompass all the possible changes and arrangements with equality.

Claims (9)

1.一种数据储存系统,与一存取要求单元成电气地耦合,包含: N个储存装置,选择性地输出一写入确认讯息,每一储存装置包含一各自的写入高速缓存与一各自的记录媒体,N为一自然数; 一备份内存,一旦该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内的数据;以及一控制器,用以将从该存取要求单元传送的多笔数据执行一预定运算以产生多笔写入数据,传送这些写入数据至该N个储存装置的写入高速缓存,复制这些写入数据至该备份内存内,并且响应该写入确认讯息,将复制至该备份内存内的这些写入数据标示为”有效”, 其中当该备份内存之储存空间低于一门坎值或复制这些写入数据至该备份内存内之地址重复时,该控制器发布一清理快取命令至该N个储存装置,使该N个储存装置的写入高速缓存在将该些写入数据完全写入该N个 A data storage system, and an access unit to be electrically coupled claim, comprising: N number storage means, for selectively outputting a write confirmation message, each storage device comprises a respective write cache and a each recording medium, N being a natural number; a backup memory, once the data storage system from a power interruption, the backup data stored in the memory remains stored therein; and a controller for the access requirement from the unit multi pen performs a predetermined data transfer operation to generate a plurality of write data, the write data transfer to the write cache of the N storage device, copy the backup data is written into the memory, and in response to the write the confirmation message, which is written to the copy data in the backup memory labeled "valid", wherein the backup memory when the storage space is below a threshold value or copy the write data of the address in the backup memory to repeat, the cache controller issues a command to the cleaning of the N storage means, the write cache storage device of the N fully written write data in the some of the N 存装置的记录媒体后,由该N个储存装置回复该控制器一清理快取完成讯息,该控制器响应该清理快取完成讯息而清除储存于该备份内存内之写入数据。 After the recording medium storage device, the reply to the N storage device controller that a clean-up complete message cache, the cache controller in response to the cleaning and clear complete message within the write data stored in the backup memory.
2.根据权利要求1所述的数据储存系统,其特征在于:当该数据储存系统开机时,若该备份内存内存有标示”有效”的写入数据,该控制器则将该有效的写入数据写回该N个储存装置,并且在完成后,清除存于该备份内存内的写入数据。 2. The data storage system according to claim 1, wherein: when the boot data storage system, the backup memory if the memory is marked "valid" data is written, then the controller of the valid write write data back to the N storage means, and after completion, clear stored in the write data in the backup memory.
3.根据权利要求2所述的数据储存系统,其特征在于:该预定运算选自由一互斥或(XOR)运算、一镜像运算、一无运算、一切分运算、一有限加洛瓦场本质多项式运算以及一有限场的数学最大距离可分码运算所组成的群组中之其一。 The data storage system according to claim 2, wherein: the predetermined operation selected from the group consisting of an exclusive OR (XOR) operation, an image operation, a non-operation of all sub-operation, a finite nature of Gallois Field polynomial arithmetic and mathematics group of a finite field arithmetic maximum distance separable code consisting of the one.
4.根据权利要求2所述的数据储存系统,其特征在于:该备份内存为一非挥发性内存或一配置一独立电源的挥发性内存,该备份内存仅电连接至该控制器、该控制器内的一写入高速缓存,或电连接在该控制器与该N个储存装置之间。 4. The data storage system recited in claim 2, wherein: the backup memory is a non-volatile memory or a volatile configuration memory of a separate power source, the backup memory only electrically connected to the controller, which controls in a write cache, or electrical connection between the controller and the N storage means.
5.一种针对一数据储存系统的管理方法,该数据储存系统与一存取要求单元成电气地耦合,并且包含N个储存装置以及一备份内存,N为一自然数,每一储存装置包含一各自的写入高速缓存与一各自的记录媒体,一旦该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内的数据,该管理方法包含下列步骤: (a)接收该存取要求单元传送的多笔数据; (b)将这些接收到的数据执行一预定运算以产生多笔写入数据; (C)判断该备份内存的储存空间是否低于一门坎值,或判断复制这些写入数据至该备份内存内的地址是否重复; 若步骤(C)判断结果为肯定者,执行下列步骤: 由一控制器发布一清理快取命令至该N个储存装置; 该N个储存装置的写入高速缓存在将该等写入数据完全写入该N个储存装置的记录媒体后,由该N个储存装置回复该控制器一清理快取完成讯 A management method for a data storage system, the data access request to a storage system electrically coupled to the unit, and comprising N storage means and a backup memory, N being a natural number, each storage device comprises a each write cache with a respective recording medium, once the data storage system from a power interruption, the backup data stored in the memory is still stored therein, the management method comprising the steps of: (a) receiving the access request multi-pen unit transmitted; (b) the implementation of these received data is a predetermined write operation to generate a plurality of data; if (C) determining whether the backup memory storage space is below a threshold value, or determines to write copy the data to the address in the backup memory is repeated; if step (C) by determination result is affirmative, the following steps: a cleaning issued by a cache controller to command the N storage means; storage device of the N after the write cache completely written the N recording medium in the other storage device to write data, the reply of the N storage means cleaning the cache controller a complete hearing 息;以及由该控制器响应该N个储存装置传送之该清理快取完成讯息,清除储存于该备份内存内之写入数据; 若步骤(c)判断结果为否定者,执行下列步骤: 传送这些写入数据至该N个储存装置的写入高速缓存,并复制这些写入数据至该备份内存内;以及响应该N个储存装置传送的一写入确认讯息,将复制至该备份内存内之这些写入数据标示为”有效”。 Information; and the clean up is completed messaging cache of the N storage means by the controller in response, clears the write data stored in the backup memory; if step (c) by the determination result is negative, performing the following steps: transmitting these write data to the write cache of the N storage means, and to copy the write data into the backup memory; and in response to a write acknowledgment message transmitted by the N storage means, the backup will be copied into the memory of these write data marked as "effective."
6.根据权利要求5所述的管理方法,其特征在于:当该数据储存系统开机时,执行下列步骤: 判断该备份内存内是否存有标示”有效”的写入资料;以及若判断结果为肯定者,将该备份内存内标示有效的写入数据写入该N个储存装置,并且在完成后,清除储存于该备份内存内的写入数据。 6. The management method as claimed in claim 5, wherein: when the boot data storage system, perform the following steps: determining whether there labeled "valid" is written to the data within the backup memory; and if the judgment result is certainly are, within the designated backup memory valid write data of the N storage means, and after completion, clearing the write data stored in the backup memory.
7.根据权利要求6所述之管理方法,其特征在于:该预定运算选自由一互斥或(XOR)运算、一镜像运算、一无运算、一切分运算、一有限加洛瓦场本质多项式运算以及一有限场的数学最大距离可分码运算所组成的群组中之其一。 The management method of claim 6 claim, wherein: the predetermined operation selected from the group consisting of an exclusive OR (XOR) operation, an image operation, a non-operation, all operation points, essentially a finite field polynomial Gallois the maximum distance separable code group mathematical calculation operation and a limited field consisting of the one.
8.根据权利要求7所述之管理方法,其特征在于:该备份内存为一非挥发性内存或一配置一独立电源之挥发性内存。 8. The management method of claim 7, wherein: the backup memory is a non-volatile memory or a volatile memory configuration of an independent power source.
9.一种数据储存系统,与一存取要求单元成电气地耦合,包含: N个储存装置,选择性地输出一写入确认讯息,每一储存装置包含一各自的写入高速缓存与一各自的记录媒体,N为一自然数; 一备份内存,一旦该数据储存系统遭受电力中断时,该备份内存仍保存储存在其内的数据;以及一控制器,用以将从该存取要求单元传送的多笔数据执行一预定运算以产生多笔写入数据,复制这些写入数据至该备份内存内,并且响应该写入确认讯息,将复制至该备份内存内的这些写入数据标示为”有效”, 其中当该备份内存之储存空间低于一门坎值或复制这些写入数据至该备份内存内之地址重复时,该控制器发布一清理快取命令至该N个储存装置,使该N个储存装置的写入高速缓存在将该些写入数据完全写入该N个储存装置的记录媒体后,由该N个储存装置回复该控制 A data storage system, and an access unit to be electrically coupled claim, comprising: N number storage means, for selectively outputting a write confirmation message, each storage device comprises a respective write cache and a each recording medium, N being a natural number; a backup memory, once the data storage system from a power interruption, the backup data stored in the memory remains stored therein; and a controller for the access requirement from the unit multi pen performs a predetermined data transfer operation to generate a plurality of write data, the write data is copied into said backup memory, and written in response to the confirmation message, which is written to the copy data in the backup memory is designated an "effective", wherein the backup memory when the storage space is below a threshold value or copy the write data of the address in the backup memory to repeat, the cache controller issues a command to the cleaning of the N storage means, so that the N write cache storage means after the data is completely written to write some of the N recording medium storage device, the reply is controlled by the N storage means 一清理快取完成讯息,该控制器响应该清理快取完成讯息而清除储存于该备份内存内之写入数据,当该数据储存系统开机时,若该备份内存内存有标示”有效”的写入数据,该控制器则将该有效的写入数据写回该N个储存装置,并且在完成后,清除存于该备份内存内之写入数据。 A cache clean-up complete message, the controller in response to the completion message cache cleanup cleared within the write data stored in the backup memory, when the data storage system boot, the backup memory if the memory is marked "active" Write data into the controller of the valid write data of the write back of the N storage means, and after completion, clear stored in the backup memory of the write data.
CN 201110282863 2011-09-22 2011-09-22 Backup data storage system comprising a memory and a management method thereof CN103019966B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110282863 CN103019966B (en) 2011-09-22 2011-09-22 Backup data storage system comprising a memory and a management method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110282863 CN103019966B (en) 2011-09-22 2011-09-22 Backup data storage system comprising a memory and a management method thereof

Publications (2)

Publication Number Publication Date
CN103019966A true CN103019966A (en) 2013-04-03
CN103019966B true CN103019966B (en) 2015-08-19

Family

ID=47968591

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110282863 CN103019966B (en) 2011-09-22 2011-09-22 Backup data storage system comprising a memory and a management method thereof

Country Status (1)

Country Link
CN (1) CN103019966B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6243795B1 (en) * 1998-08-04 2001-06-05 The Board Of Governors For Higher Education, State Of Rhode Island And Providence Plantations Redundant, asymmetrically parallel disk cache for a data storage system
CN101521039A (en) * 2008-02-29 2009-09-02 群联电子股份有限公司 Data storage system, controller and method therefor

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7272687B2 (en) * 2005-02-01 2007-09-18 Lsi Corporation Cache redundancy for LSI raid controllers

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6243795B1 (en) * 1998-08-04 2001-06-05 The Board Of Governors For Higher Education, State Of Rhode Island And Providence Plantations Redundant, asymmetrically parallel disk cache for a data storage system
CN101521039A (en) * 2008-02-29 2009-09-02 群联电子股份有限公司 Data storage system, controller and method therefor

Also Published As

Publication number Publication date Type
CN103019966A (en) 2013-04-03 application

Similar Documents

Publication Publication Date Title
US7350101B1 (en) Simultaneous writing and reconstruction of a redundant array of independent limited performance storage devices
US7823009B1 (en) Fault tolerant distributed storage for cloud computing
US7315976B2 (en) Method for using CRC as metadata to protect against drive anomaly errors in a storage array
US20090172335A1 (en) Flash devices with raid
US7890795B1 (en) Auto-adapting cache memory system and memory
US20090327802A1 (en) Storage control device, data recovery device, and storage system
US20080115017A1 (en) Detection and correction of block-level data corruption in fault-tolerant data-storage systems
US20050005191A1 (en) System and method for detecting write errors in a storage device
US20080256420A1 (en) Error checking addressable blocks in storage
US20110191649A1 (en) Solid state drive and method of controlling an error thereof
US20020184481A1 (en) Method, system and program for initializing a storage device comprising multiple storage units through a storage controller
US5896493A (en) Raid algorithm using a multimedia functional unit
US20140089730A1 (en) Storage control device and method for controlling storages
US20110320915A1 (en) Method and system to improve the performance and/or reliability of a solid-state drive
US9128855B1 (en) Flash cache partitioning
Dholakia et al. A new intra-disk redundancy scheme for high-reliability RAID storage systems in the presence of unrecoverable errors
US20100251012A1 (en) Data Volume Rebuilder and Methods for Arranging Data Volumes for Improved RAID Reconstruction Performance
US9135119B1 (en) System and method for data management
US8799571B1 (en) System and method for configuring a device array upon detecting addition of a storage device
US20120260146A1 (en) Data storage device related method of operation
US20090235022A1 (en) Apparatus and method to select a deduplication protocol for a data storage library
US7774643B2 (en) Method and apparatus for preventing permanent data loss due to single failure of a fault tolerant array
US20120054456A1 (en) Migrating an encoded data slice based on an end-of-life memory level of a memory device
US20090271659A1 (en) Raid rebuild using file system and block list
CN101504625A (en) Method for implementing independent disk redundancy array, solid state disk and electronic equipment

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted