TW200413907A - Storage system with memory for storing data - Google Patents

Abstract

A data storage system (110) utilizing a non-volatile integrated circuit based memory (126) for storing data. In some examples, the non-volatile integrated circuit based memory is a magnetoresistive random access memory (MRAM) (126). In some examples where the data storage system (110) includes a hard disk, the non-volatile integrated circuit based memory can be used to store system data such as a master list (127) for the hard disk. In other examples, the non-volatile integrated circuit based memory may be implemented in a removable media storage system (1510) to store operational data associated with a removable media. In other examples, the non-volatile integrated circuit based memory may be used to store user data.

Description

200413907 玖、發明說明·· 【發明所屬之技術領域】 本發明一般關於資料之儲存及特定地，關於抽取式媒體 儲存單位内資料之儲存。 【先前技術】 資料儲存媒體用於資料的儲存。資料儲存媒體之範例包 括，積體電路式儲存媒體，如磁電阻式隨機存取記憶體 (MRAM)，動態隨機存取記憶體（DRAM)，靜態隨機存取記 憶體（SRAM)，快閃記憶體，電子抹除式唯讀記憶體（EEpR〇M) ，可抹式唯讀記憶體（EPROM)，唯讀記憶體（R〇M)，隨機存 取記憶體（RAM)及可程式唯讀記憶體（pR〇M)。非積體電路 1媒體之範例包括，磁性儲存媒體，如磁性硬碟，磁軌， 軟磁盤；光學性系統’如光碟（CD)及數位影音光碟（dvd); ^兹光媒體。上述之非IC式記憶體類型本質上為機械化且 ^需要機械化之操作（如旋轉磁碟，移動磁帶）以揭取資 =媒體之資料存取時間—般顯著地較㈣式媒體存年 著^快。然而’非1C媒體資料儲存之每位S成本-般顯 者地較1C式媒體更低。因此， 大量料。 ㈣式般用於資料的 利用在非1C式媒體的存取日李門 统M m 、 仔取時間限制可降低利用媒體之系 、，死的操作效率。故需要一玄枝 ^ 操作效率。 系、，无，其可為改善非1C式媒體的 更關於抽取式媒體之辟左 ^ ^ ^ ^ ^ 4存早位，當移動抽取式媒體時， 乂自抽取式媒體存取资料 /、行·^貝科一般不以非揮發性地儲 87246 200413907 存於抽取式媒體之資料儲存系統。因此，當再置入抽取式 媒體時應該重新存取該資訊。 【發明内容】 一種硬碟系統，包括： 一硬碟，配置以儲存使用者資料，其中可配置一主列表 以識別至少一資料區塊之實質位置，該資料與使用者資料 檔案相聯；及 一非揮發性積體電路式記憶體，配置非揮發性積體電路 式記憶體以儲·存與硬碟相聯之主列表。 【實施方式】 本申請案已於美國申請專利，申請號10/230,690，1〇/23〇,785 ，10/230,788及 1〇/231，868，申請日 08/29/2002。 下文詳述為實行本發明之模式。述文僅為本發明繪示部 份，而不應以此為限。 圖1為根據本發明之資料儲存系統之方塊圖。依實例所 不，資料儲存系統110包括一非1(3式媒體112及用以執行資料 存取至媒體112之線路。資料儲存系統11〇可藉一 1/〇匯流排 128耦口，以接收來自資訊裝置141之使用者資料以儲存於媒 體112及提供從媒體U2擷取之使用者資料至資訊裝置丨“。 資訊裝置141所使用之資料為使用者資料。資訊裝置141之使 用者貝料範例可包括執行程式檔，資料庫檔，多媒體資料 才田（如曰术，影首，圖片），資訊檔（如文字處理文件或試算 表）及私腦&作者配置文件。在一範例中，資訊裝置ι4ΐ為 個人私細系統及資訊儲存系統丨1〇為裝配於電腦系統之電 87246 200413907 腦周邊（未示）内之硬碟萃絲r 中，匯—^ ^ (或有時稱為硬碟機）。在此範例 ’為付合電腦匯流排標準(如爾準，增強刑 IDE (E-IDE)標準，或小型電 知卞曰鈿』 ^ 月旬系統界面（SCSI)標準）之匯、户 排。在其他實例資訊裝置141 、 匚机 個人影音紀錄器，σ)’㈣媒體㈣器， 立變…〜 術星視訊轉換器，電視遊戲機， 理㈣Α)，呼叫器，行動電令，：車…印’個人數位助 ^ % ^ ^ ^ r , „ 衫首播放器，中央家用通訊 术、.泉态及數位相機。在其他實 顯示器，耳機.，麥克風，j裝置可為喇队， ^ 或，、他可利用或提供儲存於資料· 儲存系統110之使用者資料的裝置。 在實例中所示，媒辦Μ ? 4 _ I 2匕括一用以儲存使用者資料之硬 、、 ^ 2可包括多重硬碟，或其他種類 《記憶體，如光學光碟200413907 发明 、 Explanation of the invention ... [Technical field to which the invention belongs] The present invention relates generally to the storage of data and specifically to the storage of data in a removable media storage unit. [Prior art] A data storage medium is used for data storage. Examples of data storage media include integrated circuit storage media such as magnetoresistive random access memory (MRAM), dynamic random access memory (DRAM), static random access memory (SRAM), and flash memory Memory, electronic erasable read-only memory (EEpROM), erasable read-only memory (EPROM), read-only memory (ROM), random access memory (RAM), and programmable read-only memory Memory (pROM). Examples of non-integrated circuit 1 media include magnetic storage media such as magnetic hard disks, magnetic tracks, and floppy disks; optical systems such as compact discs (CDs) and digital audio-visual discs (dvd); and ^ optical media. The above non-IC type of memory is essentially mechanized and requires mechanized operations (such as rotating disks, moving tapes) to expose funds = media data access time-as significantly longer than traditional media. fast. However, the cost per bit of non-1C media data storage is generally lower than that of 1C media. So a lot of material. Normally used for data utilization. Access to non-1C-type media. Li Mian system Mm, the time limit for fetching can reduce the efficiency of the use of media, and the dead operation. Therefore, it requires a mysterious operation. Yes, no, it can be a better way to improve non-1C media. ^ ^ ^ ^ ^ 4 Save early. When you move removable media, you can access data /, self-extractable media. · ^ Beke generally does not store non-volatile data storage systems 87246 200413907 stored in removable media. Therefore, the information should be re-accessed when the removable media is re-inserted. SUMMARY OF THE INVENTION A hard disk system includes: a hard disk configured to store user data, wherein a master list can be configured to identify a physical location of at least one data block, the data being associated with a user data file; and A non-volatile integrated circuit-type memory configured with a non-volatile integrated circuit-type memory to store and store a master list associated with a hard disk. [Embodiment] This application has been applied for a patent in the United States, application numbers 10 / 230,690, 10 / 23〇, 785, 10 / 230,788 and 10 / 231,868, application date 08/29/2002. The following is a detailed description of the mode for carrying out the present invention. The description is only part of the present invention, and should not be limited to this. FIG. 1 is a block diagram of a data storage system according to the present invention. According to the example, the data storage system 110 includes a non-1 (3-type media 112 and a line for performing data access to the media 112. The data storage system 110 can use a 1/0 bus 128 coupling port to receive The user data from the information device 141 is stored in the media 112 and provides the user data retrieved from the media U2 to the information device. The data used by the information device 141 is user data. The user data of the information device 141 Examples can include executive program files, database files, multimedia data fields (such as art, shadows, pictures), information files (such as word processing documents or spreadsheets), and personal brain & author profile. In one example, The information device ι4 is a personal personal information system and an information storage system. １１ is an electrical device assembled in a computer system. 87246 200413907 The hard disk extracting wire r in the periphery of the brain (not shown), the sink— ^ ^ (or sometimes called hard Disk drive). In this example, 'combined with the computer bus standard (such as the standard, enhanced punishment IDE (E-IDE) standard, or mini-electronics 卞 钿 钿 ^ month system interface (SCSI) standard) meeting , Household row. In other For example, the information device 141, the personal video recorder of),)) '㈣ media player, stand-up ... ~ Shuxing video converter, TV game console, management ㈣), pager, mobile electric order, car ... print' Personal digital assistant ^% ^ ^ ^ r, „T-shirt player, central home communication, spring and digital camera. In other real monitors, headphones, microphones, and j devices, it may be a team, ^ or, and he may use or provide the user data stored in the data storage system 110 device. As shown in the example, the media office M_4_I2 includes a hard disk for storing user data, and may include multiple hard disks, or other types of memory, such as optical discs.

Am、， 果磁T，磁光儲存裝置，或任何其 他用以彳諸存资料之摇雕 /、十 < 媒賤，或上述所列媒體種類之組合。 貨料儲存系統1 10包括一祆 G拾 &制系統111。控制系統ill包括 制資料儲存系統11G操作之處理HU6。可將處理哭 116耦合至匯流排128以由資 卞如里口口 田貝成I置141提供及接收使用者資 : 器116藉頭控信號線及擴大機119而提供-頭控信號 至頭驅動器131以移動_讀寫頭113至資料存取之媒體⑴之 通當磁軌。處理器116亦藉馬達控制線及擴大機122而提供一 馬^ ^制訊號至王軸馬達114以維持馬達114於—丨亙速率。處 里时116藉-寫入信號線及擴大機⑽而提供資料至讀寫頭 3 乂知貝料寫入至一特定區段。處理器i 16藉READ線及擴 大機118接收來自謂寫頭J13之資料。控制系統⑴包括用以 87246 200413907 儲存程式編碼之程式記憶體124,處理器116執行該程式編石、、 以執行其操作。熟習此技藝者將了解，資料餘存系統包^ 其他圖1未示之習用電路及裝置。亦可配置其他種類之儲 系統以實作文中所教示内容。 。予 控制系統111包括可耦合至處理器116之記憶體126。在— 貫例中，於一 1C式記憶體（如MRAM)中實作記憶體。 MRAM為磁性儲存資料之非揮發性記憶體。在mram記憶俨 中，於操作期間可變更儲存於MRAM之資料。在某些 記憶體中，記.憶體中之資料可藉不大於5_5伏特之電壓而運 作。大邵分之MRAM藉小於2伏特之電壓而運作。因此，用 以供給控制系統111電子元件所需電壓之電力可用於運作 程式記憶體126。在一實例中，記憶體124及126實作於相同 之積體電路。在其他實例中，記憶體IN及實作於相同之 $己k、ta陣列。在某些實例中，於儲存系統操作期間，配置 至私式^憶體124及i己憶體126之部份陣列可為不同。在另一 只例中主列表可儲存於記憶體12 6内之不同位置。在一實 例中’處理器116，程式記憶體124及記憶體126連同控制系 統111之其他電子裝置（未示）一起實作於一電路板（未示）。 硬碟112之主列表127為非揮發性地儲存於記憶體126。一 主列表一般包括一條目，其各檔案儲存於儲存媒體丨12。各 條目均包括區塊媒體112上之實體位置，該區塊為儲存該檔 案貨料之弟一區塊。 圖2為以符號表示之主列表及硬碟表面之表層，其繪示將 資料儲存於硬碟之習用方法範例。在圖2之實例中，檔案儲 87246 200413907Am, magneto T, magneto-optical storage device, or any other engraving /, medium &media; or combination of media types listed above. The cargo storage system 110 includes a G & P manufacturing system 111. The control system ill includes a processing HU6 for operation of the data storage system 11G. The processing unit 116 can be coupled to the bus 128 to provide and receive user data from the device such as Likoukou Tianbei Cheng I. 141: The device 116 is provided by the head control signal line and the amplifier 119. Move the read / write head 113 to the magnetic track of the media for data access. The processor 116 also provides a signal via the motor control line and the amplifier 122 to the king axis motor 114 to maintain the motor 114 at the speed. When it is 116, it provides data to the read / write head by using a write-in signal line and an amplifier. 3 It knows that the material is written to a specific section. The processor i 16 receives the data from the predicate write head J13 via the READ line and the amplifier 118. The control system does not include a program memory 124 for storing program codes in 87246 200413907, and the processor 116 executes the program editing to perform its operations. Those skilled in the art will understand that the remaining data system package ^ other conventional circuits and devices not shown in Figure 1. Other types of storage systems can also be configured to implement what is taught in the composition. . The control system 111 includes a memory 126 that can be coupled to the processor 116. In the example, the memory is implemented in a 1C type memory (such as MRAM). MRAM is a non-volatile memory that stores data magnetically. In mram memory, the data stored in MRAM can be changed during operation. In some memories, the data in the memory can be operated with a voltage not greater than 5-5 volts. Da Shao MRAM operates with a voltage of less than 2 volts. Therefore, the power used to supply the voltage required for the electronic components of the control system 111 can be used to operate the program memory 126. In one example, the memories 124 and 126 are implemented in the same integrated circuit. In other examples, the memory IN is implemented in the same $ k, ta array. In some examples, during storage system operation, some of the arrays allocated to the private memory 124 and the private memory 126 may be different. In another example, the master list may be stored in different locations in the memory 126. In one example, the processor 116, program memory 124 and memory 126 are implemented on a circuit board (not shown) together with other electronic devices (not shown) of the control system 111. The master list 127 of the hard disk 112 is non-volatilely stored in the memory 126. A master list generally includes an entry, each file of which is stored on a storage medium. Each entry includes the physical location on the block media 112, which is the block that stores the material of the file. Fig. 2 is a master list represented by symbols and a surface layer on the surface of a hard disk, which illustrates an example of a conventional method for storing data on the hard disk. In the example of Figure 2, the archive stores 87246 200413907

存於硬碟表面212，其位於註明DATA A，DATA B及DATA C 之使用者資料之三區塊，並個別儲存於硬碟表面212之區段 215’區段216及區段217。區段區塊206顯示儲存於區段215 之貝料’ £ 4又區塊208顯不儲存於區段216之資料及區段區塊 210顯示儲存於區段217之資料。各區段區塊（206，207及208) 均包括一檔頭，一儲存於區段（如區段區塊2〇6包括區塊 DATA A)之使用者資料區塊及一包括儲存下一檔案使用者 /貝料區塊之區段實體位置之尾部。例如，區段區塊2〇6之尾 部包括儲存區.塊DATA B之區段2丨6的實體位置。 欲檢視一儲存於磁碟表面之檔案，資料儲存處理器（如， 116)首先需存取主列表2〇5，以便取得儲存於第一使用者資 料區塊（DATA A)之區段215的實體位置。該處理器隨後將資 料存取於區段215以獲得位於該位置及區段216 (為儲存下 二使用者^資料區塊（DATAB))的實體位置之使用者資料。區 段2Π的實體位置位於區段區塊2〇8 (儲存於區段216)之尾部 =段一檔案尾部（E0F)之標記位於區段區塊21〇(儲存於區 段7)&lt;尾#。这植案尾邵之標記指示包括樓案資料的最 終邵分之區段區塊210。 。^用 &lt; 硬碟系統’主列表均儲存於硬碟之一（或更多） 。4 了存取樓案，硬碟㈣之處理器將主列表存取 ^更碟上、，以便獲得標案資料之第一區段區塊的位置。當 新擒案寫入硬石莱時，主列矣由 、 田 R , 刁表中便產生一新條目。此新條 目匕括像存於檔案資料 π ^ m 罘 5塊内《區段的實體位置。 回晨、圖1’因資料儲存系統110之主列表127儲存於一扣式 87246 -10- 200413907 Μ &amp; &amp; β m中’因存取主列表127須存取硬碟112使得主 列表 &lt; 存取時間顯著地減少。因此，可因少去至少一次至 硬碟112之存取而縮短―檔案之存取時間。此外，因主列表 127為非揮發性地儲存於記憶體126中，即使若緊急電源關 閉，主列表亦可自動地更新。此外，因主列表不需由硬碟 、t視即可接著以新樓案之條目回寫至硬碟⑴，是故亦會 減少寫入一新檔案至硬碟的時間。 此外，一般於硬碟系統中，主列表儲存於硬碟表面之區 段。若該硬環.之區段毁壞了’則整個資料儲存系統會變成 無法使用。在某些非似記憶體（如硬碟系統或磁帶驅動系 統）中，資料儲存於非IC式記憶體而毁壞之機率，會較資料 儲存於1C式記憶體（如MRAM記憶體）之毀壞機率更大。因 此，使用這些較可靠記憶體可使資料儲存系統更為可靠。 卜在木些Λ例中，主列表之備份亦可餘存於硬碟中。 在某些如此的實例中，主列表之備份可儲存於任何硬碟112 中之任一位置。且在這些實例中，處理器116可將備份主列 表與原始表相比找出錯誤之處。 圖3為根據本發明之資料存取之實例流程圖。圖3之流程 圖表現由處理器116執行儲存於程式記憶體124之編碼。在 302中，處理斋116接收來自資訊裝置14ι之存取請求。在go# 中，處理态116將該存取判定為讀取存取或為寫入存取。若 該存取為讀取存取，307之處理器116會將主列表127存取於 記憶體126中，以找出儲存請求檔案資料之第一區段的位 置。在310中，處理器116擷取來自硬碟112區段之檔案。若 87246 -11 - 200413907 304中，該存取判定為寫入存耳又，則3〇6中，處理器叫執行 窝入存取且在308中，更新主列表127。 斤圖2之範例中，主列表2G5之各條目包括位於硬縣面212 第一區段之硬碟表面212的實體位置，該硬碟表面212儲存與 條目聯合之標案資料之區塊。然而，纟本發明某些實例中^ 主列表包括各檔案條目之其他資料及/或附加資料。、例如， ，考圖5,主列表中各檔案之條目52〇均包括各區段（其儲存 孩檔案資料之區塊）之位置。參照以下圖5之論述。 次在其他實例·中，可配置處理器110以儲存其他種類之系統 資料於記憶體126。例如，記憶體126可儲存下列列表：說明 媒體112毁壞區段之列表，可於媒體112取得區段之列表，識 刎f又壞區段及用於毁壞區段的代替區段（毁壞區段代替列 表）之列表及保護媒體112上的區段及/或資料·之備份列 ^。亦可配置記憶體Π6以儲存讀取/寫入通道參數，如頭增 盈及伺服參數，如II矩常數。記憶體126可依媒體ιΐ2之媒體 =及儲存資料於媒體112之方法，而儲存其他種類之系統 ^料非揮發性1C式記憶體中之儲存系統資料可使得一資 料儲存系統處理器更快速地存取系統資料，因而可藉資料 儲存系統縮短使用者資料之全部存取時間。 非揮發性記憶體126亦可用以儲存接收來自資訊裝置141 、、使用者貝料。利用記憶體126以错存使用者資料可改善資 料儲存系統之效能，速度及/或效率。 贿圖4為傳遞資料存取之方法流程圖，其中利用一部份記憶 匕126以快取使用者資料。在圖4之實例中，由資訊裝置141 87246 -12- 200413907 接收之使用者資料檔案儲存於記憶體126。因此，若檔案有 一稍後之請求，若與存取媒體112而獲得資料之時間相比， 提供儲存於記憶體126之檔案至資訊裝置141所需時間較少。 參考圖4，處理器116自資訊裝置141接收存取請求。若在 414中，判定存取請求為寫入存取，則在430中，處理器116 將檢視記憶體126是否填滿。若記憶體126沒有填滿，則434 中，處理器116將接收資料寫入至記憶體126。若430中判定 記憶體126為填滿，則在432中，處理器116執行一快取清除 演算法，以將目前儲存於記憶體126之檔案寫入至媒體112，-以便記憶體126為所接收資料騰出空位。在434，自資訊裝置 141接收之資料寫入至記憶體126。在一範例中，快取清除演 算法意味移動記憶體126中最早寫入之資料至媒體112。在另 一實例中，快取清除演算法包括移動近期最少使用之資 料。當快取資料時，另一快取清除演算法會移動一檔案或 同樣大小之複數個檔案。熟習此技藝者該了解根據文中所 教示，亦可實作其他快取清除演算法。 若414中，判定存取為一讀取存取，則416中，處理器.116 將檢視在記憶體126中是否快取請求資料。若在記憶體126 中發現請求資料，則422中，由記憶體126擷取該請求資料且 提供至資訊裝置141。若於記憶體126中無法發現416中所判 定之請求資料，則處理器自媒體112接收請求資料。 在另一實例中，處理器116可判定在預定時間内頻繁地存 取儲存於媒體112之檔案，且處理器116可將該檔案寫入至記 憶體126。因此，可大幅減少提供檔案至資訊裝置141所需之 87246 -13- 200413907 時間。 藉由資料儲存系統之1C式記憶體内快取使用者資料，可 大幅減少提供資料至資訊裝置之存取時間。此外，藉由利 用非揮發性記憶體以快取資料，甚至能在能源損失之情況 下仍可保留儲存於記憶體126之資料。因此，若有一儲存於 記憶體112及記憶體126之檔案備份，則記憶體126中之檔案 備份為較新之備份，由此可除去判定哪一備份為較新備份 的需要。 此外，利用·記憶體126作為一資料寫入之快取，使得資料 儲存系統110可在資料寫入後更快速地執行一接替資料存 取。在某些習用之資料儲存系統中，在資料寫入後資料儲 存系統無法處理資料之存取，直到該資料寫入之資料寫入 至非ic式媒體（如硬碟）。然而，若利用部份記憶體126作為 資料寫入之緩衝器，在第一資料寫入之資料可立即寫入至 1C式記憶體126之後，資料儲存系統丨1〇可立即執行第二資料 存取。因記憶體126為非揮發性記憶體，甚若自資料儲存系 統卿開電源時，使用者資料仍可維持在記憶體126中。因 此’在某些實例中，並無迫切的需要來將儲存於記憶體⑶ 之資料寫人至儲存I统11G，且因此，在資訊裝置i4i無存取 期間内，資料可寫入至媒體112。 在其他實例中，記憶體126中之資料儲存系統可快取許 由資訊裝置141所接收之檔案資料，其中處理器ιΐ6可 次突發寫入以將檔案寫入至硬碟112。因此，因為多重檔 (哭發寫人所需動力會較各標案之多重單—寫人所需= 87246 -14 - 200413907 為夕’可減少藉由硬碟系統i 1〇所消耗之動力。 在貝例中，實作—上文之寫入緩衝系統於圖4之流程 :：於432中《快取清除演算法，配置以確定記憶體126 水遠具有可取得之足夠緩衝空間以用於資料寫入。因 、、在414中右判疋一存取為資料寫入，處理器ye可立即 讀資料寫入至記憶體126。 右在資料儲存系統中實作圖4之方法，其包括抽取式媒體 (見圖15之1512) ’儲存於記憶體⑽（見圖is)之資料可回應 私動媒體之請·求而寫入至媒體丨5 12。 參考圖5及6,回應來自資訊裝置M1之讀取存取請求而由 媒體1職取之資料，亦可在記憶體126中緩衝該資料以縮短 讀取存取所需之時間。參考圖5，使用者資料之檔案训可 諸存表硬碟表面5〇1之複數個區段中（如53丨）。在圖$之範例 中，檔案519包括6個資料之區塊（521，522，523，524，及 6)且各區塊均儲存於磁碟表面501之不同區段内。例如， 區塊50丨儲存於磁碟表面501之區段531。 在某些資料儲存系統中，資料之區塊可儲存於需要資料 餘存系統之一讀窝頭（如113)磁碟表面的非連續區段，以在 多重磁軌之間移回及移出以擷取資料。例如，區段53ι (其 儲存檔案519之資料521之第一區塊）位於磁碟表面5〇1之磁 軌522。區段532 (其儲存資料522之第二區塊）位於磁軌551。 區段533 (其儲存資料523之第三區塊）位於磁軌551。習用 中，擷取檔案之資料，需要一讀取頭（113)移至磁軌552以擷 取儲存於區段531之第一區塊521，接著移至磁軌551以擷取 87246 -15- 200413907 儲存於區段532之第二資料區塊522，且隨後移回至磁軌 以擷取儲存於區段533之第三資料區塊533。習用中，於—讀 取存取期間内，習用方法之擷取時間可因於磁軌間之—讀 取頭（如113)之移動而延遲。 、 本發明之貫例實作非循序之資料讀取以將資料擷取時間 取佳化。在這些實例中，在資料提供至資訊裝置141之前， 可先以非循序次序擷取及在記憶體126中緩衝。參考圖5， 在這些貫例中，主列表127之各檔案條目（如52…包括磁碟表 面501各區段（如531)之位置，該磁碟表面5〇1儲存與檔案條 目駟合之檔案資料區塊。例如，主列表條目52〇包括區段 531 ’ 532 ’ 533 ’ 534，535及 536之各自位置 LI，L2，L3，L4， L5及L6。包括儲存主列表之檔案條目中之檔案資料之各區 段位置可使得系統以非循序性次序擷取資料。因此，可減 少掏取貧料檔案所需之時間。 圖6為藉貝料儲存系統而傳遞一資料存取的實作方法流 私'圖’該資料儲存系統實作非循序之資料擷取。在603，當 處理器116由資訊裝置141接收一檔案之資料存取請求，且在 605中判定該請求存取為一讀取，則處理器116存取與請求檔 案聯合之主列表127條目，以在607中，擷取儲存檔案資料之 區’又位置。在此實例中，主列表127之各條目近似於圖5之 條目520，因其含有儲存與條目聯合之檔案資料的每一區段 位置。 在609中’處理器116將區段（如531，532)之位置（如L1，L2) 分類’以判定用以擷取在區段中資料的最佳路徑。判定擷 87246 -16- 200413907 取“牛中最佳路徑之規則依 如，在某些習用硬磲系統中，鱼在存系統之操作。例 移動相比，磁軌至磁軌間 磁軌中之不同區段間 -方式為進行至下-：:：：要更多的時間。因此， 個區段之構案的所有區/ H储存於—軌磁的複數 磁軌的複數個E p / ’、 g用硬碟系統中，在一 進至所需磁軌的曰較由先一磁軌珂 另-射^々 可因此而判定-最佳路徑。 硬碟上之為宫京了匕括在擷取躓取請求時位於 為傳送請求資料至资“另判疋取佳路徑之因素· 作。.一^/、 ”成裝置（如141)之資料儲存系統的操 ^ π式之目的為將由資訊裝置請求資 、土疋i二請求資料至資訊裝置的時間減到最少。 ，在圖5之S例中’若讀寫頭113位於-外部磁執，則資料 儲存系欠統將首先擴取在磁軌551之資料（如於區段M2,似及 533之貝料）’且接著擷取在磁軌522之資料（如於區段兄1， 及533之貝料）。在一磁軌中擷取資料之特定順序會依據進 入磁軌時的唄寫頭之位置。在一範例中，最佳區段擴取路 徑可為區段532，534，535，531，536及533之順序。 參考回顧圖6，於在609判定最佳擷取路徑後，611中之處 理器116掏取路徑之第一區段中的資料，且在615中緩衝記憶 體126中之資料。上述之擷取路徑，處理器u6首先擷取區段 532中之資料區塊522，且將資料區塊522寫入至記憶體126 (見圖5)。615中，處理器116判定記憶體126中之資料是否包 括接著將傳送至資訊裝置141之資料區塊。因此時記憶體126 87246 -17- 200413907 僅包括第二資料區塊522,且因第一資料區塊521無法傳送至 資訊裝置141，故處理器116回到611以#貞取來自路徑下一區 段（區段534)之資料（區塊524)，且接著進行至613以窝入資料 區塊524至記憶體126。因為無法擷取第一資料區塊52ι，故 直到由區段531擷取區塊521後，處理器116才回到61卜在此 時’資料區塊52；1，522，524及525儲存於記憶體126。因此， 615中，處理器116判定傳送至資訊裝置141之下一資料區塊 (521)為位於記憶體126中。627中，傳送資料區塊521至資訊 裝置141。處理器116接著回到615，其判定下一個傳送之資 料區塊（資料區塊522)在記憶體126中緩衝，且因此627中傳 送貝料區塊522。在此時，因下一傳送之資料區塊並不位於 纪fe、體126中，故處理器H6回到611以由區段536擴取下一資 料區塊526，且6Π中，將區塊526寫入於記憶體丨26。因615 中，資料區塊523 (將傳送之下一區塊）並不位於記憶體126 中，故處理器116回到611以擷取來自區段533之區塊523。因 區塊523現在位於記憶體126中，故處理器116接著傳送區塊 523至資訊裝置141。因剩下的區塊524，525及526亦位於記情 體126中，則處理器116繼續615至627之程序，其中以連鲭: 序將剩下的區塊傳送至資訊裝置141。615中，當沒有待傳送 之資料時，則處理器116會進行到619，因已擷取檔案之全部 區塊，故該操作在623處終止。 田在其他貫例中’相儲存系統可利用其他種類之非循序 最佳擷取方法。例如，可設計處理器116以 資料區塊，㈣，且當區段儲存第一資料區塊二= 87246 -18- 200413907 衝位於相同磁軌之複數個區段 〈所有區塊。處理哭拉莫腺 到該磁軌，其具有儲存下一待 时接耆知 得&amp;貝料區塊之區段（提 料區塊，其不位於儲存第一區塊 尾&lt; ^ #又磁軌），並搞取；5绾 衝儲存於該磁軌之複數個區段中 &amp; . T J不田木 &lt; 所有資料區塊。 因此，在此方式下，讀寫頭⑴可於每次的擴取中由一磁軌 擴取資料。並且在其他實例中，亦可緩衝位於另—種類= $己fe體（如，DRAM或SRAM)之資料。 提供-主列表，其包括含有樓案資料之全部區段位置， 該主列表使圖6實例之資料儲存系統可實作—非猶序概 取。因處理器已知全部檔案之位置，故該處理器不需以連 續順序檢視資料。在其他實財，主列表可儲存其他資訊 以指示儲存檔案資料之區段位置。例如，在某些實例中， 主列表之檔案條目之各區段位置欄位包括一連續數目欄 位，其指不來自與區段位置條目（含有檔案資料之連續區塊） 聯合之區段磁軌中的連續區段數目。因此，資料儲存系統 處理器（如116)可由主列表判定包括檔案資料區塊之連續區 段群因此，母磁軌之此主列表僅可含有之一個區段位置 欄位’其位於磁軌之連續順序的區段群。 在某些實例中，由資訊裝置141接收之使用者資料可為非 揮發性地儲存於記憶體126中以便減少資料擷取時間及/或 有效地增加非1C式儲存媒體之容量。 圖7所示為以符號表現之實例中，其記憶體126利用以非 揮發性儲存使用者資料，以減少將檔案儲存於硬碟表面712 之擴取時間。在此實例中，記憶體126用以非揮發性地儲存 87246 -19- 200413907 檔案之資料，若該檔案完全地儲存於硬碟表面712，則將具 有一相對較長的再審時間。例如，使用者資料檔案最初由 以下三者所生成：儲存於硬碟表面712區段A之第一資料區 塊DATA A，儲存於硬碟表面712區段B之第二資料區塊DATA B及儲存於硬碟表面712區段C之第三資料區塊DATA C。因 資料區塊儲存於三個連續順序之區段A，B及C，故三資料 區塊DATA A，DATA B及DATA C組成的檔案之擷取時間相 對較短。區段區塊706，708及710分別表現儲存於區段A，B 及C之資料。· - 在檔案（由DATA A，DATA B及DATA C所組成）最初儲存於 磁碟表面712之後，資訊裝置141伴隨檔案之修正提供資料儲 存系統110。該修正包括一附加資料區塊DATA D (如區段區 塊714所示）。因缺少如區段A，B及C般位於磁碟表面712之 相同磁軌的可得區段，故儲存DATA D於磁碟表面712將大幅 增加擷取檔案之時間，因為該檔案之擷取可意味著將讀取 頭在二個不同磁碟之間移動至少兩次。然而，在圖7之實例 中，儲存區段區塊714於記憶體126中，其中區段區塊708之 尾節提出儲存DATA D之記憶體126之位置（LOC D)。區段區 塊714之尾部區段提出儲存資料區塊DATA C之磁碟表面712 區段C的位置（LOCC)。因DATA D儲存於1C式記憶體126中， 故擷取修正檔案（由DATA A，DATA B，DATA C及DATA D 所組成）所需之時間，其約與擷取原檔案所需之時間相同 (由 DATA A，DATA B及 DATA C所組成）。 可以其他方式利用記憶體126以非揮發性地儲存使用者 87246 -20- 200413907 資料，以增加資料擷取速度或增加媒體112之實際容量。例 如，若一使用者資料之檔案為一長度，其檔案之最終資料 區塊為幾個數元組，相對於需佔據媒體112之全部區段，可 儲存該數個位元組於記憶體126中。在一實例中，可設計處 理器116以媒體112之每一區段單價及媒體126每一位元單 價。若檔案結束區塊之位元數目夠小，而儲存資料於記憶 體126之成本，會較利用媒體112之全部區段之成本更少，接 著處理器116儲存最終區塊於記憶體126中。在另一實例中， 若僅修訂檔案·之中間區塊因其超過可儲存於區段之大小，-則超過區段大小之修訂資料區塊部分可儲存於記憶體126 中〇 在其他實例中，記憶體126可利用以儲存資料，其除此以 外儲存於判定為毁壞之區段内。在此範例為關於實作於低 價磁碟（RAID)系統之冗餘陣列的資料儲存系統。圖9為表示 實作RAID系統之部分硬碟系統。硬碟系統901包括二硬碟 904及905，其中儲存資訊於硬碟904及905的兩侧。圖8顯示 硬碟904及905之各側，其中硬碟表面807位於硬碟905之底 側，硬碟表面808位於硬碟905之頂側，硬碟表面809位於磁 碟904之底表面及硬碟表面810位於硬碟904之頂表面。 習用之RAID系統中，使用者資料儲存在位於多重硬碟表 面之區塊中。實作RAID系統之儲存系統的驅動線路同時驅 動各表面，故與各表面聯合之讀寫頭位置遍佈區段，因各 表面均具有相同之區段位置。各表面均具有相同區段位置 的各區段上之資料區塊為平行地讀取，以獲得檔案之資 87246 -21 - 200413907 料。用以於多重硬碟表面儲存資料之技術有時稱為資料分 段。 ，、刀 參考圖8及圖9’因磁碟904及9〇5同時移動至於各磁碟表面 之相同區段（如807) ’若有-區段毁壞（如當由在表面綱之χ 指出區段822為毀壞的）’則位於所有硬碟表面區段位置之 區段（如82卜822, 823及824)成為無法使用。因此，藉由平 行地操作之磁碟表面數目，在一 pr f /二罢、 在區段位置炙區段毁壞為倍 數增加的。 在-系統之.實財，其在多重硬碟表面上實作資料分段 (如同-RAID系統），若判定磁碟表面之—區段為毁壞的， 則待窝人至該毁壞區段之資料區塊會儲存於記憶體⑶ 中。當該區段位置沒有毁壞區段時，儲存其他資料區^在 其他磁碟表面之其他區段）具有相同之區段位置。例如，在 圖8,因區段822為毁壞的，故待窝入至該區段之资料备寫 入至記憶體m，其中資料分段之其他資料區塊寫入至：段 82卜823及824 (具有如區段822的相同區段位置之硬碟表面 8〇7 ’ ’及810之區段）。藉由利用記憶體ΐ2ό以儲存待寫入 至毀壞區段之資料，則毁壞區段無法使其他位於其他磁碟 表面之區段位置之區段成為無法使用。 在，、他貝例中，一非揮發性ic式記憶體可利用於資料儲 存^统，以增加儲存於資料儲存系統之非IC式媒體的資料 數里♦考圖10，於習用之硬碟表面（如1 〇 1 〇)中，其資料错 存方、區#又刀割中（如區段分割4)之區段上（如1〇29)，兑位於 環繞硬碟之集線器刪的磁軌（如磁軌18)上。區段—般位於 87246 -22- 200413907 界足區段分割之區段線之間（如1031及1032)。在某些實例 中’词服資料可儲存於區段線（如1031及1032)上。伺服資料 之範例包括磁軌號碼，區段號碼及磁軌内用以放置讀寫頭 113之類比資料。習用中，估量各區段之大小以保持資料之 標準大小區塊（如一範例之512位元）。在某些範例中，多重 區段可位於一區段分割之磁軌中。 因習用硬碟之區段位在磁軌中，各磁軌因集線器（如1〇2〇) 而具有不同半徑，因區段分割中之磁軌的磁碟表面空間少 於儲存資料標·準大小區塊所需之空間，部分磁碟可能為無 法使用。例如，在磁軌23及22之區段分割4 (位於區段線1〇31 及10324間）的磁碟表面數量其少於儲存資料之標準大小區 塊（如512)所需空間。區段分割4之磁軌23上的空間1〇12短少 了空間1016之數量，其藉位於磁軌23由區段線1932右側之虛 線所示’相關圖示在圖10中。由虛線直接至區段線1〇32右側 之所示部分表示磁碟表面空間，而直接位於區段線1〇32之 貝料之標準大小區塊。例 I ’而磁軌20之空間1024 左側之空間數量不足用以儲存一 如，部分1023表現磁碟空間之數 不足以儲存資料之標準大小區塊。因此，在習甩之 將無法利用這些空間，而造成於硬碟表 厂、、 又呆衣面乏儲存空間的浪 費。 圖11所示為根據本發明某些眘例，甘主 ^ J 其表示儲存於記憶體&lt; 126之部分資料。在本發明的某虺眘你丨士 木一⑤例中，硬碟表面之空 間’較儲存資料之標準大小區塊所；^ ; 、 兄W而 &lt; 可得儲存空間更 小，該硬碟表面空間可用以儲存資料々挪 -灯&lt; %準大小區塊的一 87246 -23- 200413907 部份，而其他部份之資料標 中。例如，資料之標準大 ；^讀存於記憶體126 空間购中（見圖1G)。資料之^的—部份财#磁軌Μ之 為在圖1G中空間1G16之大^'小區塊的剩餘部分（其 記憶體m亦儲存資料之標 =“al26&lt;位置⑴6。 aA甘^、 區塊的剩餘部分，該資料 的其他邵分儲存於硬碟表面1〇1〇 ，、十 蝴浐冒1 1 1 W纟六A 。例如，記憶體126之記憶 胆:位置1115儲存資料之擇谁+ ,v 心卞大小區塊的剩餘部分（其大小由 部为1015指出），該資料的其八 、他口15刀储存於位置1011 (見圖 10)。 在某些實例中，與表面1010上 ^ 川上&lt;各毁壞空間聯合的記憶 骨豆12 6位置保留以t 田以、、'口工間。例如，保留記憶體126之區段 1130’以儲存磁軌23全部區段的全部磁碟空間之毁壞部份。 在某些實财，可儲存資料區塊之第—部份於記憶體126及 儲存剩餘部分於磁碟表面1〇1〇。在一實例中，相對於由儲 存其他部分之硬碟空間使用所剩，可依據儲存該部分於記 te體126之成本，來判定是否要使用記憶體126來儲存資料之 標準大小區塊的一部份。 #頁取圖10及11實例之一資料區塊，處理器U6可擷取以下 兩者··來自磁碟空間（如1012)的資料部分及儲存於記憶體 126的資料部分（如位置U16)。參見關於圖14之論述。處理 器116可在提供其至資訊裝置141之前先將該部分結合。因資 料的剩餘部分儲存於1(^式記憶體（如126)中，故包括該剩餘 部分之區塊與全部儲存於硬碟區段之區塊，兩區塊之擷取 時間實質上為相同。因此，儲存位於一 1C式之部份資料標 87246 -24- 200413907 準大小區塊，非揮發性記憶體中可較有益地利用一硬碟表 面。 配置某些習用之硬碟系統，故區段可位於區段線（如刪） 之兩側。在這些系統中’硬碟系統包括程式編碼以由越過 區段線之區段摘取資料（如，該系、統包括編碼以移動伺服資 料’其儲存於區段之使用者資料）。利用記憶體126以儲存 部分使用者資料（如圖1 g及11所示）使得配置-系統，因為區 段區塊之剩餘部分儲料記憶體126中，故其中沒有區段越 過區：線二因.此’在此實例中將不需要上述之程式編碼。 在某些貫例中’為了加速儲存於資料儲存系統110之使用 者資料檔案的擷取時間，該檔案之部分可寫入至IC式記憶 體126中，用於非揮發性儲存與靜止資料之全部資料儲存於 非1C式媒體112。在若干這些實例中，依據儲存於㈣式媒 體之資料㈣取時間，來決定寫人賤式記憶體（如126)及 非1C式媒體(如112)的資料。在若干這些實例中，欲減少檔 案心所有掏取時間，如藉資訊裝置（如141)由資料儲存系統 (如110)擴取檔案。 入fIC式記憶體126中，故可減少資料的#1取時間。圖12為 來藉資料儲存系統11G之處理器116實作之流程圖。 在削，處理器116需執行呈現於圖12之方法，以回應自 圖12根據本發明而提出-寫入操作實例，其用以實作一 系^，其位於1以非揮發性記憶體巾非揮發性地儲存使用 =資料之部分構案’故可加速樓案之拍I取時間。在圖12之 貝例中冑入至非IC式媒體i 12之下—資料，其引導部份寫 87246 -25- 200413907 資訊裝置⑷接收之窝入存取請求。資料儲存系統㈣错 用者資料之檔案於區塊（其各區塊已儲存）中，在一余歹 中，則儲存在位於媒體112之磁碟表面的區段中。在— 中，自未使用之區段列表（未示）獲得用以儲存檔案之區 段，其儲存於記憶體126。在其他實例中，若欲寫入之檔: 為m反’則第-區段為先前修訂之第—區段儲存資料， 如主列表127所指出。在12〇3，區段號碼（N)設定為1及 1、2〇5,將資料之第—區i寫人至媒體112之區段ι (自 之區段列表獲·得之第一區段）。 在12〇7,處理器116自第一區段（N)判定下一區段（N+1)之 搜，時間。下_區段為由未使用之區段列表（於—新構^ 獲得之下一區段，或儲存先前修訂資料之下一區塊的下1 區段。搜尋時間為讀寫頭113由第一區段（N)終點進行至下一 區段（N+1)的所需時間。如先前區段之情況，了 一區ς為浐 於相同磁軌中下一連續區段（見圖7所示範例之磁碟表面7二 中’、區段A，Β及C位於相同磁軌之連續順序），其搜尋時間 比軚釔然而，若下一區段（N+1)位於與先前區段（N)不同 之磁軌，則其搜尋時間將比較長。 、口 在1210，處理益116判定儲存於記憶體126之資料大小。儲 存於Ate體126之資料大小為根據於區段間之搜尋時間，其 中之資料為先前所寫入（N)及下一區段（N+1)。若下—區段 :磁軌中先前區段之的下一連續區段，因為相對較短心 二時間’處理器116在121〇可判定出無需將資料儲存於記憶 把126中。然而若’搜尋時間比較長，處理器116將在121〇 87246 -26- 200413907 判定儲存於記憶體126之資料數量。 儲存於記憶體126之資料為儲存在下一區段之資料的引 領部份。如同下一區段之使用者資料之引領部份，如此的 資料包括下一區段之區段區塊檔頭。在1211中，引領部份 寫入於記憶體126中。在1213中，處理器116將一指示寫入主 列表127 (或記憶體126的其他部份），其資料之下一區塊的引 領部份位於記憶體126中。 U15中，資料寫入於下一區段（N+1)。在一實例中，只有 使用者資料（在資料區塊寫入至前一區段（N)後之資料）之下 一資料區塊的剩餘部份（未寫入至記憶體126的部份）寫入於 下一區段（N+1)。因此，具有一標準區段區塊大小512位元之 系統中，若將第一 62位元寫入記憶體126時，則512位元區塊 剩餘之450位元可寫入於下一區段（N+1)。在其他實例中，可 將檔案資料對下一標準區段區塊大小寫入於下一區段 (N+ 1)。例如，若62位元之檔案寫入於記憶體126，貝|J檔案之 下一區塊大小可寫入於下一區段（N+1)。在其他實例中，資 料之完整下一區塊大小寫入下一區段（N+1)(如，下一 512位 元之資料包括寫入至記憶體126之62位元之資料），該資料包 括寫入至記憶體126之資料引領部份。在此實例中，檔案之 完整備份儲存於媒體112中。在某些實例中，寫入記憶體126 之部份會較資料（如，512)標準大小區塊更大。再另一實例 中，若搜尋時間較預設閥值更大，則下一資料之標準大小 區塊（如，50位元，100位元，512位元或1024位元）儲存於記 憶體126中。在一實例中，其中儲存於下一區段區塊（N+1) 87246 -27- 200413907 之資料亦儲存於記憶體126中，則在一讀取存取期間，處現 詻116會跳越區段N+1之讀取且進行至區段N+2。 1219中，判定是否所有的㈣資料㈣存於資料臂存系 統110。S是的話，則處理器終止。若否的話，咖中〜會 增加至N+丨及處理器116會回到12〇7且重複該方法直到儲二 所有的樓案資料份。圖12之實財，在㈣之資料存取: 間，處理器116可將儲存於記憶體126之引領部份資料提供至 資訊裝置，而讀寫頭113移動到下一區段。 圖13根據本發明提出實作一系統之寫入操作的下一會 例，該系統用以儲存位於資料儲存系統之IC式非揮發性記 憶體中的使用者資料部份檔案，以便加速檔案的擷取時 間。除了在圖13之實例中，將寫入非IC式媒體112區段之资 料尾部部份寫入至1(：式記憶體126之外，圖13之實例與圖j2 之實例相同。在此方法中，當二區段間有一長搜尋時間， 項寫頭（如，1Π)可在完成讀取儲存於第一區段之資料前先 由一第一區段移動至下一區段。 1301中，處理器116反應由資訊裝置接收一寫入請求以儲 存一使用者資料檔案而開始執行圖13所提出之方法。13〇3 中’區段數目（N)設為1及1304中，檔案資料的第一區塊寫入 媒體112上之區段N(指定儲存檔案之媒體U2第一區段）。若 檔案尾部（EOF)未到達如1305中所判定處，則13〇7中，處理 為116判足指定在媒體112上儲存一檔案之下一區段（N+1)的 位置。1309中，處理器116計算由當時區段（N)(在此方法環 路期間之區段1)至下一區段（N+1)之搜尋時間。 87246 -28- 200413907 1310中，處ί里器116判定13〇4中寫入當時區段⑼之資 邵部份的：小(其儲存在記憶體126中)。尾部部份為待寫： 至^段^資料的最後部份。在某些實例中，尾部部份包括 下區#又《叙置。儲存於記憶體126之尾部部份大小基於 #時區段（Ν)與下—區段（Ν+1)間之搜尋時間。若下一區段 (Ν+1)為由田時區段（Ν)在磁軌中之下一連續區段，則處理 為116會在1310中會由於相對較短的搜尋時間而判定沒有资 料需要儲存於記憶體126。若然而，搜料間相對較長，^ 理器116會在131G中判定儲存在記憶體126中之資料數量。在 某些實例中，若搜尋時間符合特定閥值時，尾部部份之標 準大小會儲存於記憶體126中。在某些實例中，若到區段ν+ι &lt;搜尋時間夠長的話，體將寫入區段N之整個資料區塊寫入 記憶體112。在實例中，其中整個區段區塊餘存於記憶體⑶ 中’處理斋II6會控制讀寫頭ιη而由區段跳越至區段 N+1。 1311中，具有1310中所判定大小的尾部部份儲存於記憶體 126中丨3丨4中，一扣示為區段尾部部份寫入主列表丨27 ,。然 而，在某些實例中，資料尾部部份與一指示連續儲存於記 fe骨豆126中，泫指示為特定尾部部份所屬之區段區塊。因 此，處理器116可判定記憶體126中之哪個區段區塊尾部部份 會藉由讀取記憶體126中（其儲存尾部部份）之位置而儲存。 1315中，N增加1而其中處理器U6回到13〇4重複程序。 在圖13之貫例中’所有檔案資料寫入於媒體112之區段， 包括寫入記憶體126之資料。然而，在其他實例中，只有未 87246 -29- 200413907 寫入記憶體126之檔案資料會寫入 入媒體112。It is stored on the hard disk surface 212, which is located in three blocks of user data marked DATA A, DATA B, and DATA C, and is individually stored in the sections 215 ', 216, and 217 of the hard disk surface 212. Segment block 206 displays the material stored in segment 215 '£ 4 and block 208 displays the data stored in segment 216 and segment block 210 displays the data stored in segment 217. Each section block (206, 207, and 208) includes a header, a user data block stored in a section (such as section block 206 including block DATA A), and a block containing the next The end of the physical location of the file user / shell material block. For example, the end of section block 206 includes the physical location of storage area. Block DATA 6 of section 2-6. To view a file stored on the disk surface, the data storage processor (eg, 116) first needs to access the master list 205 in order to obtain the data stored in section 215 of the first user data block (DATA A). Physical location. The processor then accesses the data in section 215 to obtain user data at that location and at the physical location of section 216 (which stores the next user ^ data block (DATAB)). The physical location of section 2Π is located at the end of section block 208 (stored in section 216) = the mark at the end of section 1 file (E0F) is located in section block 21o (stored in section 7) &lt; 尾 #. The mark at the end of the case indicates the section block 210 including the final point of the case data. . ^ Use &lt; The hard drive system ’master list is stored on one (or more) hard drives. After accessing the case, the processor of the hard disk will access the main list on the disc to obtain the location of the first section of the project data. When the new capture was written in Hard Rock, the main column 矣 Yu, Tian R, and Diao Table produced a new entry. This new entry contains the image in the file π ^ m 罘 The physical location of the section within 5 blocks. Back to the morning, Figure 1 ’The main list 127 of the data storage system 110 is stored in a one-button 87246 -10- 200413907 Μ & &amp; β m’ Because the main list 127 needs to access the hard disk 112 to make the main list &lt; Access time is significantly reduced. Therefore, it is possible to shorten the access time of the file by reducing the access to the hard disk 112 at least once. In addition, since the main list 127 is non-volatilely stored in the memory 126, the main list can be automatically updated even if the emergency power is turned off. In addition, because the main list does not need to be viewed from the hard disk, t can be written back to the hard disk with the entry of the new building case, which will reduce the time to write a new file to the hard disk. In addition, in a hard disk system, the master list is stored in a section on the surface of the hard disk. If the section of the hard ring is destroyed ’, the entire data storage system will become unusable. In some non-memory-like memory (such as a hard disk system or a tape drive system), the probability of data being stored in non-IC memory will be greater than the probability of data being stored in 1C-type memory (such as MRAM memory). Bigger. Therefore, using these more reliable memories makes the data storage system more reliable. For example, in some examples, a backup of the master list can also be stored on the hard disk. In some such instances, a backup of the master list can be stored anywhere on any hard drive 112. And in these examples, the processor 116 may compare the backup master list to the original table to find the error. FIG. 3 is a flowchart of an example of data access according to the present invention. The flowchart of FIG. 3 shows that the processor 116 executes the codes stored in the program memory 124. In 302, the process 116 receives an access request from the information device 14m. In go #, processing state 116 determines the access as a read access or a write access. If the access is a read access, the processor 116 of the 307 will access the main list 127 in the memory 126 to find the location of the first sector storing the requested file data. At 310, the processor 116 retrieves files from the hard disk 112 section. If in 87246 -11-200413907 304, the access is determined to be written to the memory, then in 306, the processor is called to execute the access and in 308, the main list 127 is updated. In the example of FIG. 2, each entry of the main list 2G5 includes a physical location of a hard disk surface 212 located in the first section of the hard county surface 212, and the hard disk surface 212 stores a block of project data associated with the entry. However, in some examples of the present invention, the master list includes other information and / or additional information for each file entry. For example, consider Figure 5. The entry 52 of each file in the master list includes the location of each section (the block where the child file data is stored). Refer to the discussion of FIG. 5 below. In other examples, the processor 110 may be configured to store other kinds of system data in the memory 126. For example, the memory 126 may store the following list: a description of the damaged section of the media 112, a list of the sections can be obtained at the media 112, and the bad section and the replacement section (destroyed section) used to destroy the section Replacement list) and the backup list of sections and / or data on the protected media 112 ^. Memory Π6 can also be configured to store read / write channel parameters, such as head gain and servo parameters, such as II moment constant. The memory 126 can be stored in the media 112 according to the media = 2 and the method of storing data in the media 112, and other types of systems can be stored. The storage system data in the non-volatile 1C type memory can make a data storage system processor faster. Access to system data, so you can use the data storage system to reduce the overall access time for user data. The non-volatile memory 126 can also be used to store and receive information from the information device 141 and the user. Using memory 126 to stagger user data can improve the performance, speed, and / or efficiency of the data storage system. Fig. 4 is a flowchart of a method for transferring data access, in which a part of the memory 126 is used to cache user data. In the example of FIG. 4, the user data file received by the information device 141 87246 -12- 200413907 is stored in the memory 126. Therefore, if the file has a later request, it takes less time to provide the file stored in the memory 126 to the information device 141 than the time to obtain the data by accessing the medium 112. Referring to FIG. 4, the processor 116 receives an access request from the information device 141. If it is determined in 414 that the access request is a write access, in 430, the processor 116 will check whether the memory 126 is full. If the memory 126 is not full, then in 434, the processor 116 writes the received data to the memory 126. If the memory 126 is determined to be full in 430, then in 432, the processor 116 executes a cache clearing algorithm to write the file currently stored in the memory 126 to the medium 112, so that the memory 126 is Receive data to make room. At 434, the data received from the information device 141 is written to the memory 126. In one example, the cache removal algorithm means that the earliest written data in the mobile memory 126 is transferred to the media 112. In another example, the cache removal algorithm includes moving the least recently used data. When caching data, another cache clearing algorithm moves one file or multiple files of the same size. Those skilled in this art should understand that according to the teachings in the text, they can also implement other cache removal algorithms. If it is determined in 414 that the access is a read access, in 416, the processor .116 will check whether the requested data is cached in the memory 126. If the requested data is found in the memory 126, then in 422, the requested data is retrieved by the memory 126 and provided to the information device 141. If the requested data determined in 416 cannot be found in the memory 126, the processor receives the requested data from the media 112. In another example, the processor 116 may determine that a file stored in the media 112 is frequently accessed within a predetermined time, and the processor 116 may write the file to the memory 126. As a result, the time required to provide files to the information device 141 can be significantly reduced from 87246 to 13-13907. By caching user data in the 1C-type memory of the data storage system, the access time for providing data to the information device can be greatly reduced. In addition, by using non-volatile memory to cache data, the data stored in memory 126 can be retained even in the event of energy loss. Therefore, if there is a file backup stored in the memory 112 and the memory 126, the file backup in the memory 126 is a newer backup, thereby eliminating the need to determine which backup is a newer backup. In addition, the memory 126 is used as a data write cache, so that the data storage system 110 can perform a replacement data access more quickly after the data is written. In some conventional data storage systems, after the data is written, the data storage system cannot handle data access until the data written in the data is written to non-ic media (such as hard disks). However, if a part of the memory 126 is used as a buffer for data writing, after the data written in the first data can be written into the 1C-type memory 126 immediately, the data storage system can execute the second data storage immediately. take. Since the memory 126 is a non-volatile memory, user data can be maintained in the memory 126 even when the power is turned on from the data storage system. Therefore, in some instances, there is no urgent need to write the data stored in the memory ⑶ to the storage system 11G, and therefore, during the period when the information device i4i has no access, the data can be written to the media 112 . In other examples, the data storage system in the memory 126 may cache the file data received by the information device 141, wherein the processor 6 may write to the hard disk 112 in bursts. Therefore, because of the multiple files (the power required by the crying writer will be greater than the multiple orders of each bid-the writer needs = 87246 -14-200413907 for the evening), the power consumed by the hard disk system i 10 will be reduced. In the example, the implementation-the flow of the write buffer system in Figure 4 above: in 432 "cache clear algorithm, configured to determine that the memory 126 water has sufficient buffer space available for use Data write. Because, in 414, the first access is data write, and the processor ye can immediately read the data and write it to the memory 126. Right, implement the method of FIG. 4 in the data storage system, which includes Removable media (see 1512 in Figure 15) 'The data stored in the memory (see Figure is) can be written to the media in response to requests and requests from private media 5-12. With reference to Figures 5 and 6, the response comes from the information The data accessed by the media 1 by the read access request of the device M1 can also be buffered in the memory 126 to shorten the time required for read access. Referring to FIG. 5, the file training of user data can be read. The table is stored in a plurality of sections on the hard disk surface 501 (such as 53 丨). In the example of Figure $, the file 519 includes 6 blocks of data (521, 522, 523, 524, and 6) and each block is stored in a different section of the disk surface 501. For example, block 50 丨 is stored in the disk surface 501 Section 531. In some data storage systems, blocks of data can be stored in discontinuous sections on the surface of the readhead (such as 113) disks that require one of the data retention systems to move between multiple tracks. Go back and move out to retrieve data. For example, section 53m (the first block of data 521 where it stores files 519) is located on track 522 on the disk surface 501. Section 532 (the second block where it stores data 522 Block) is located on magnetic track 551. Section 533 (the third block that stores data 523) is located on magnetic track 551. In practice, to retrieve file data, a reading head (113) is required to move to magnetic track 552 to Retrieve the first block 521 stored in section 531, then move to track 551 to retrieve 87246 -15- 200413907 second data block 522 stored in section 532, and then move back to the track to retrieve Take the third data block 533 stored in section 533. In practice, during the read access period, the acquisition time of the conventional method can be Between the tracks—the movement of the reading head (such as 113) is delayed. The present invention implements non-sequential data reading to optimize the data acquisition time. In these examples, the data is provided to Before the information device 141, it can be retrieved in non-sequential order and buffered in the memory 126. Referring to FIG. 5, in these examples, each file entry of the main list 127 (such as 52 ... includes each section of the disk surface 501) (Such as 531), the disk surface 501 stores the file data block that is combined with the file entry. For example, the main list entry 52 includes the respective positions of the sections 531 '532' 533 '534, 535, and 536 LI, L2, L3, L4, L5 and L6. The location of each segment including the file data in the file entry of the master list allows the system to retrieve the data in a non-sequential order. As a result, the time required to retrieve the lean materials can be reduced. Fig. 6 is a flow chart of an implementation method for transferring a data access through a storage system for private materials. The data storage system implements non-sequential data acquisition. At 603, when the processor 116 receives a data access request for a file from the information device 141, and determines that the requested access is a read at 605, the processor 116 accesses the main list 127 entry associated with the requested file, In 607, the area where the file data is stored is retrieved. In this example, each entry of the main list 127 is similar to the entry 520 of Fig. 5 because it contains the location of each sector where the archive data associated with the entry is stored. In 609, the 'processor 116 classifies the positions (e.g., L1, L2) of the segment (e.g., 531, 532)' to determine the best path for retrieving the data in the segment. Judgment 87246 -16- 200413907 The rule for taking the best path in cattle is based on the operation of the fish storage system in some conventional hard-clam systems. For example, compared with the movement of the track, the difference between the track and the track Inter-segment-the way is to proceed to the next-:::: to take more time. Therefore, all the areas of the block's construction / H are stored in a plurality of tracks of a plurality of tracks, E p / ', In the hard disk system, the first track that enters the required track is another one, which can be judged as the best path. The one on the hard disk is Gong Jing. The fetching request is located at the factor of "separately determining the best way to send the requested information to the job". The purpose of the operation of the data storage system of a device (such as 141) is to minimize the time required for the information device to request data and the information requested from the information device to the information device. As shown in Figure 5 In the example S, 'If the read / write head 113 is located on the external magnetic actuator, the data storage system will first expand the data on the magnetic track 551 (such as in the section M2, similar to the shell material of 533)' and then retrieve The data on track 522 (such as the material in section 1 and 533). The specific order of data retrieval in a track will depend on the position of the transcription head when entering the track. In one example, The optimal segment expansion path can be in the order of segments 532, 534, 535, 531, 536, and 533. Referring to FIG. The data in the first section and the data in the memory 126 are buffered in 615. In the above retrieval path, the processor u6 first retrieves the data block 522 in the section 532, and writes the data block 522. Into the memory 126 (see FIG. 5). In 615, the processor 116 determines whether the data in the memory 126 includes the information and then sends it to the information The data block of device 141. Therefore, the current memory 126 87246 -17- 200413907 includes only the second data block 522, and because the first data block 521 cannot be transmitted to the information device 141, the processor 116 returns to 611 with # The data (block 524) from the next section (section 534) of the path is captured, and then proceeds to 613 to nest data block 524 to memory 126. Because the first data block 52m cannot be retrieved, so Until block 521 is retrieved by section 531, processor 116 does not return to 61. At this time, data block 52; 1,522,524, and 525 are stored in memory 126. Therefore, in 615, processor 116 It is determined that a data block (521) transmitted to the information device 141 is located in the memory 126. In 627, the data block 521 is transmitted to the information device 141. The processor 116 then returns to 615, which determines the next transmitted data The block (data block 522) is buffered in the memory 126, and therefore the shell material block 522 is transmitted in 627. At this time, because the data block to be transmitted next is not located in the file 126, it is processed. The processor H6 returns to 611 to expand the next data block 526 from section 536, and in 6Π, block 526 Into memory 丨 26. Since the data block 523 (the next block to be transmitted) is not located in memory 126 in 615, the processor 116 returns to 611 to retrieve block 523 from section 533 Since block 523 is now located in memory 126, processor 116 then sends block 523 to information device 141. Since the remaining blocks 524, 525, and 526 are also located in memory 126, processor 116 continues The procedure from 615 to 627, in which the remaining blocks are transmitted to the information device 141 in the order of mackerel: In 615, when there is no data to be transmitted, the processor 116 proceeds to 619, because All blocks, so the operation is terminated at 623. In other examples, Tian ’s phase storage system can use other kinds of non-sequential optimal acquisition methods. For example, the processor 116 may be designed with data blocks, and when the first data block is stored in the sector two = 87246 -18- 200413907, a plurality of sectors <all blocks located on the same track are punched. Process the lamo gland to this track, which has a section where the next standby &amp; shell material block is stored (the feed block, which is not located at the end of the first block of storage) &lt; ^ # Again track), and fetch; 5 绾 stored in multiple sections of the track &amp;. T J 不 田木 &lt; All data blocks. Therefore, in this mode, the read / write head can acquire data from a magnetic track during each expansion. And in other examples, it is also possible to buffer data located in another type = $ selffeature (such as DRAM or SRAM). Provide-master list, which includes all section locations containing building case data. The master list enables the data storage system of the example of Figure 6 to be implemented—a non-sequential summary. Because the processor knows the location of all files, the processor does not need to view the data in sequential order. In other real assets, the master list can store other information to indicate the location of the section where the file data is stored. For example, in some instances, each section location field of a file entry in the main list includes a continuous number of fields, which refers to a section magnetization that does not come from a section location entry (a contiguous block containing file data) The number of consecutive sections in the track. Therefore, the data storage system processor (such as 116) can determine the continuous segment group including the file data block from the master list. Therefore, this master list of the parent track can only contain one sector position field 'It is located on the track Group of consecutive sequential segments. In some examples, the user data received by the information device 141 may be non-volatilely stored in the memory 126 in order to reduce data retrieval time and / or effectively increase the capacity of the non-C storage medium. Fig. 7 shows an example represented by symbols. The memory 126 uses non-volatile storage of user data to reduce the expansion time of storing files on the hard disk surface 712. In this example, the memory 126 is used to store the data of the 87246-19-200413907 file non-volatilely. If the file is completely stored on the hard disk surface 712, it will have a relatively long re-review time. For example, the user data file is initially generated by the following three: the first data block DATA A stored on the hard disk surface 712 section A, the second data block DATA B stored on the hard disk surface 712 section B, and The third data block DATA C is stored on the hard disk surface 712 in section C. Because the data blocks are stored in three consecutive sections A, B, and C, the retrieval time of the files composed of the three data blocks DATA A, DATA B, and DATA C is relatively short. Segment blocks 706, 708, and 710 represent data stored in segments A, B, and C, respectively. -After the file (composed of DATA A, DATA B, and DATA C) is initially stored on the disk surface 712, the information device 141 provides the data storage system 110 with the modification of the file. The amendment includes an additional data block DATA D (shown in section block 714). Due to the lack of available sectors on the same track as the sectors A, B, and C on the disk surface 712, storing DATA D on the disk surface 712 will greatly increase the time to retrieve the file, because the file is retrieved This may mean moving the read head between two different disks at least twice. However, in the example of FIG. 7, the storage block 714 is stored in the memory 126, and the tail of the block 708 proposes the location (LOC D) of the memory 126 where DATA D is stored. The end section of the sector block 714 proposes the disk surface 712 where the data block DATA C is stored 712 the position (LOCC) of the sector C. Because DATA D is stored in 1C-type memory 126, the time required to retrieve the correction file (composed of DATA A, DATA B, DATA C, and DATA D) is about the same as the time required to retrieve the original file (Composed of DATA A, DATA B and DATA C). The memory 126 may be used to store user 87246 -20-200413907 data in a non-volatile manner in other ways to increase data retrieval speed or increase the actual capacity of the media 112. For example, if the file of a user data is a length, the final data block of the file is several bytes. Compared with the entire segment of the media 112, the number of bytes can be stored in the memory 126. in. In one example, the processor 116 may be designed at a unit price per section of the media 112 and a unit price per bit of the media 126. If the number of bits in the end block of the file is small enough, the cost of storing data in the memory 126 will be less than the cost of using all the sections of the media 112, and then the processor 116 stores the final block in the memory 126. In another example, if only the intermediate block of the file is modified because it exceeds the size that can be stored in the section,-the portion of the revised data block that exceeds the section size can be stored in the memory 126. In other examples The memory 126 can be used to store data, and in addition, it is stored in a section determined to be damaged. This example is a data storage system for a redundant array implemented in a low-priced disk (RAID) system. Fig. 9 shows a part of a hard disk system implementing a RAID system. The hard disk system 901 includes two hard disks 904 and 905, which store information on both sides of the hard disks 904 and 905. FIG. 8 shows each side of the hard disks 904 and 905. The hard disk surface 807 is located on the bottom side of the hard disk 905, the hard disk surface 808 is located on the top side of the hard disk 905, and the hard disk surface 809 is located on the bottom surface and hard disk 904. The dish surface 810 is located on the top surface of the hard disk 904. In a conventional RAID system, user data is stored in blocks located on multiple hard disk surfaces. The drive circuit of the storage system implementing the RAID system drives each surface at the same time, so the positions of the read / write heads associated with each surface are all over the sections, because each surface has the same section position. The data blocks on each section with the same section position on the surface are read in parallel to obtain the data of the file 87246 -21-200413907. The technology used to store data on multiple hard disk surfaces is sometimes called data segmentation. Refer to Fig. 8 and Fig. 9 'due to disks 904 and 905 moving to the same section (such as 807) on the surface of each disk at the same time' if there is-the section is destroyed (as indicated by χ in the surface class The segment 822 is damaged), then the segments (such as 82, 822, 823, and 824) located on all hard disk surface segment positions become unusable. Therefore, with the number of disk surfaces operating in parallel, the destruction of the sector at the sector position at a prf / 2 is increased in multiples. In-system of. Real money, it implements data segmentation on the surface of multiple hard disks (like the -RAID system). If it is determined that the -sector on the surface of the disk is damaged, then the people will go to the damaged sector. The data blocks are stored in the memory ⑶. When the sector position does not destroy the sector, other data areas (other sectors on the surface of other disks) have the same sector position. For example, in Fig. 8, because the segment 822 is damaged, the data to be nested in this segment is written to the memory m, and the other data blocks of the data segment are written to: segment 82, 823 and 824 (Sectors with the same segment position as the segment 822, 807 'and 810). By using the memory to store the data to be written to the destroyed sector, the destroyed sector cannot make other sectors on the surface of other disks unusable. In this case, a non-volatile ic memory can be used in the data storage system to increase the number of non-IC-type media stored in the data storage system. ♦ Consider Figure 10 on a conventional hard disk. On the surface (such as 010), the data is stored in the wrong place, and the section # is cut (such as section division 4) on the section (such as 1029), which is located on the magnetic disk deleted by the hub surrounding the hard disk. (Such as track 18). Sections-generally between 87246 -22- 200413907 between the segmented lines of the foot segment (such as 1031 and 1032). In some instances, 'syntax data may be stored on segment lines (such as 1031 and 1032). Examples of servo data include track numbers, sector numbers, and analog data used to place the read / write head 113 in the tracks. In practice, a standard size block (such as an example of 512 bits) is used to measure the size of each segment to hold data. In some examples, multiple sections may be located on a track divided by one section. Because the sectors of conventional hard disks are located in the tracks, each track has a different radius due to the hub (such as 1020), because the disk surface space of the track in the sector division is less than the standard size of the stored data Part of the disk space may be unusable. For example, the number of disk surfaces divided into 4 (located between section lines 1031 and 10324) in sections 23 and 22 is less than the space required for standard size blocks (such as 512) for storing data. The space 1012 on the track 23 of the segment division 4 is shorter than the space 1016, which is shown in FIG. 10 by the position of the track 23 indicated by the dashed line to the right of the section line 1932. The portion shown from the dashed line directly to the right side of the segment line 1032 represents the disk surface space, while the standard size block of the shell material located directly on the segment line 1032. Example I ', and the space on the left side of the space 1024 of the magnetic track 20 is not enough for storage. For example, the number of disk spaces in part 1023 is not enough to store standard size blocks of data. Therefore, it will not be possible to use these spaces in Xi Duo, which will cause waste to the hard disk watch factory and lack of storage space on the clothes. FIG. 11 shows some precautionary examples of the present invention. &lt; Part of 126. In a certain example of the present invention, the space on the surface of the hard disk is larger than the standard size block where the data is stored; &lt; The available storage space is smaller, and the hard disk surface space can be used to store data. &lt; 87246 -23- 200413907 part of the quasi-size block, and the data of the other parts are marked. For example, the standard of the data is large; ^ is stored in the memory 126 space purchase (see Figure 1G). The data of ^-part of the financial # magnetic track M is the remaining part of the large block of space 1G16 in Figure 1G (its memory m also stores data = "al26 &lt; Position ⑴6. aA Gan ^, the rest of the block, the rest of the data is stored on the hard disk surface 1010, and 10 浐 纟 1 1 1 纟 A A. For example, memory 126 of memory 126: who chooses to store the data at location + +15, v The remaining part of the heart sizing block (its size is indicated by the department 1015). 1011 (see Figure 10). In some instances, with surface 1010 ^ Kawakami &lt; Combined memory of each destroyed space Bone beans 12 6 positions are reserved with t'i, '' kou workshop. For example, sector 1130 'of memory 126 is reserved to store the corrupted portion of the entire disk space of all sectors of track 23. In some real money, the first part of the data block can be stored in the memory 126 and the remaining part can be stored on the disk surface 1010. In an example, compared to the remaining space used by the hard disk to store other parts, the cost of storing the part in the memory 126 can be used to determine whether to use the memory 126 to store one of the standard size blocks of data. Part. #Page takes one of the data blocks in Figures 10 and 11. The processor U6 can capture the following two ... The data part from disk space (such as 1012) and the data part stored in memory 126 (such as location U16) . See the discussion on FIG. 14. The processor 116 may combine this part before providing it to the information device 141. Because the remaining part of the data is stored in 1 (^ type memory (such as 126), the block including the remaining part and all the blocks stored in the hard disk section, the retrieval time of the two blocks is substantially the same Therefore, the storage is located in a 1C-type part of the data standard 87246 -24- 200413907 quasi-sized block, non-volatile memory can be more beneficial to use a hard disk surface. Configure some conventional hard disk system, so the area Segments can be located on both sides of the segment line (eg, delete). In these systems, the 'hard disk system includes program coding to extract data from the segment that crosses the segment line (eg, the system includes encoding to move servo data 'It is stored in the user data of the segment.] The memory 126 is used to store part of the user data (as shown in Figures 1g and 11) to make the configuration-system, because the remaining part of the segment block stores the memory 126. Therefore, there is no section crossing area: line two because of this. In this example, the above-mentioned program coding will not be needed. In some examples, in order to speed up the retrieval of user data files stored in the data storage system 110 Take time, the part of the file The data can be written into the IC-type memory 126, and all data used for non-volatile storage and stationary data is stored in the non-1C-type media 112. In several of these examples, based on the data retrieval time of the stored media, To decide to write data in low-level memory (such as 126) and non-C media (such as 112). In some of these cases, to reduce the time required for the file to be retrieved, such as using an information device (such as 141) to store data The system (such as 110) expands the file. It is stored in the fIC-type memory 126, so the # 1 fetch time of the data can be reduced. Figure 12 is a flowchart of the implementation of the processor 116 of the data storage system 11G. The device 116 needs to execute the method shown in FIG. 12 in response to the example of the write-write operation according to the present invention from FIG. 12, which is used to implement a system ^, which is located at a non-volatile memory with a non-volatile memory towel. Storage and use = part of the construction of the data 'so it can speed up the acquisition time of the building case. In the example in Figure 12, it is entered under the non-IC media i 12-data, the leading part of it is 87246 -25- 200413907 Incoming access request received by information device. Data storage system The file of the misused user's data is stored in the block (its blocks have been stored), and in a period of time, it is stored in the section located on the disk surface of the media 112. In-, since the unused section The list (not shown) obtains the sections used to store the files, which are stored in the memory 126. In other examples, if the file to be written: is m ', then the-section is the previously revised section- Store the data, as indicated by the main list 127. At 1203, the section number (N) is set to 1 and 1,205, and the first section of the data is written to the section 112 of the media (from The first segment is obtained from the segment list.) At 1207, the processor 116 determines the search time for the next segment (N + 1) from the first segment (N). The next _ section is a list of unused sections (from —new construction ^ to obtain the next section, or to store the next section of the next section of the previously revised data. The search time is read / write head 113 by the first The time required for the end of one zone (N) to reach the next zone (N + 1). As in the previous zone, one zone is the next consecutive zone in the same track (see Figure 7) In the example shown in the example of the disk surface 72, the sections A, B, and C are located in the same order of the tracks), the search time is longer than that of yttrium. However, if the next section (N + 1) The segment (N) has a different track, so its search time will be longer. When the mouth is at 1210, the processing benefit 116 determines the size of the data stored in the memory 126. The size of the data stored in the Ate body 126 is based on the interval between sectors. Search time, where the data is the previously written (N) and the next sector (N + 1). If the next-sector: the next consecutive sector of the previous sector in the track, because it is relatively short-centered At two times, the processor 116 can determine that it is not necessary to store the data in the memory handle 126. However, if the search time is longer, the processor 116 will The number of data stored in the memory 126 is determined at 121〇87246 -26- 200413907. The data stored in the memory 126 is the leading part of the data stored in the next section. It is the same as the leading part of the user data in the next section. This information includes the section header of the next section. In 1211, the leading section is written in the memory 126. In 1213, the processor 116 writes an instruction to the main list 127 (or The other part of the memory 126), the leading part of the next block of data is located in the memory 126. In U15, the data is written in the next section (N + 1). In one example, it is only used The remaining part of the data (the part after the data block is written to the previous section (N)) (the part that is not written to the memory 126) is written to the next section (N + 1). Therefore, in a system with a standard sector block size of 512 bits, if the first 62 bits are written to the memory 126, the remaining 450 bits of the 512-bit block can be written Enter the next sector (N + 1). In other examples, the file data can be written to the next standard sector block size The next section (N + 1). For example, if a 62-bit file is written in memory 126, the next block size of the | J file can be written in the next section (N + 1). In other In the example, the complete next block size of the data is written into the next block (N + 1) (for example, the next 512-bit data includes the 62-bit data written to the memory 126), the data includes The data leading portion written to the memory 126. In this example, a complete backup of the file is stored in the media 112. In some instances, the portion written to the memory 126 will be more standard than the data (eg, 512) The size block is larger. In another example, if the search time is greater than the preset threshold, the standard size block of the next data (for example, 50-bit, 100-bit, 512-bit, or 1024-bit ) Is stored in the memory 126. In an example, the data stored in the next block (N + 1) 87246 -27- 200413907 is also stored in the memory 126. During a read access, the processing volume 116 will skip The reading of section N + 1 proceeds to section N + 2. In 1219, it is determined whether all the data are stored in the data arm storage system 110. If S, the processor terminates. If not, the number of coffee beans will increase to N + 丨 and the processor 116 will return to 1207 and repeat this method until all the building case data of the second are stored. In the real money of FIG. 12, during the data access: the processor 116 can provide the leading part of the data stored in the memory 126 to the information device, and the read / write head 113 moves to the next section. FIG. 13 shows a next example of implementing a write operation of a system according to the present invention, which is used to store part of user data files in IC-type non-volatile memory of a data storage system in order to speed up the file Retrieval time. The example in FIG. 13 is the same as the example in FIG. J2, except that the tail portion of the data written to the 112 section of the non-IC media is written to 1 (:-type memory 126) in the example of FIG. 13. In the middle, when there is a long search time between the two sections, the item write head (for example, 1Π) can be moved from the first section to the next section before reading the data stored in the first section. 1301 中The processor 116 responds to the information device receiving a write request to store a user data file and starts executing the method proposed in FIG. 13. The number of sections (N) in 1303 is set to 1 and 1304, and the file data The first block of the file is written into the sector N on the media 112 (the first sector of the media U2 designated to store the file). If the end of the file (EOF) does not reach the place determined as in 1305, in 1307, the processing is 116 determines the location where the next segment (N + 1) of a file is stored on the media 112. In 1309, the processor 116 calculates the segment (N) at the time (sector 1 during this method loop) Search time to the next segment (N + 1). 87246 -28- 200413907 In 1310, the processor 116 determines that the current segment is written in 1304. The part of the asset: small (it is stored in the memory 126). The trailing part is to be written: to the last part of the paragraph ^ data. In some examples, the trailing part includes the lower area # 又 《 Description. The size of the tail portion stored in the memory 126 is based on the search time between # 时段 (Ν) and 下 — 段 (N + 1). If the next segment (N + 1) is the Yuta time segment ( Ν) In the next continuous section of the track, processing as 116 will determine that there is no data to be stored in the memory 126 due to the relatively short search time in 1310. However, if the search room is relatively long, ^ The processor 116 will determine the amount of data stored in the memory 126 in 131G. In some examples, if the search time meets a certain threshold, the standard size of the tail part will be stored in the memory 126. In some examples, if the segment ν + ι &lt; If the search time is long enough, the entire data block written in the sector N is written into the memory 112. In the example, the entire sector block is left in the memory ⑶. Processing II6 will control the read-write head and skip from sector to sector N + 1. In 1311, the tail portion having the size determined in 1310 is stored in the memory 126, 3, 4, and a buckle is shown as the tail portion of the section is written in the main list 27. However, in some examples, the trailing portion of the data and an indication are stored consecutively in the memory 126, and the indication is the sector block to which the particular trailing portion belongs. Therefore, the processor 116 can determine which section of the memory 126 has a block tail portion to be stored by reading the position in the memory 126 (its storage tail portion). In 1315, N is incremented by one and processor U6 returns to 1304 to repeat the procedure. In the example of FIG. 13, all the file data is written in the section of the media 112, including the data written in the memory 126. However, in other examples, only the file data that was not written in memory 126 87246 -29- 200413907 will be written to media 112.

前，可先移動讀寫頭113， 來自媒體112區段之資料讀取結合及提 • 141。因為資料區段區塊之尾部部份儲 在完成儲我區塊區段之區段資料讀取 113 ’因而減少在儲存系統11〇中擷取一 檔案所需時間。 一在其他貫例中，檔案之開始部份會儲存於記憶體126中及 檔案剩餘部份會儲存於媒體112中。因此，當一檔案請求發 出後，處理益可提供檔案開始部份至資訊裝置141，而移動 讀寫頭以擷取資料剩餘部份。在一範例中，一檔案開始部 份之預設大小（如，100位元，1000位元或2〇資料區塊）可儲 存於記憶體126中。會反應一讀取請求，依據資料儲存系統 由媒體112擷取資料所需時間而決定開始部份之大小。. 圖14顯示儲存於資料儲存系統11〇中之資料讀取存取操 作實例。伴隨修正使用圖14中提出之方法，以擷取儲存於 媒體112及1己憶體126兩者中之使用者資料，如範例中所示及 關於圖4，7，8，9，10，11，12及13之文中所述及這些圖之 相關文件所述，如同其他實例在文中未明確描述一般。 1400中，處理器116反應由資訊裝置141接收一檔案之讀取 請求而開始執行圖14提出之方法。1403中，處理器116判定 87246 -30- 200413907 檔案資料是否儲存於記憶體126。在某些實例中，一指示為 檔案資料是否儲存於記憶體126中，在主列表127中之檔案條 目發現該指示’。在其他實例中，記憶體126包括快取於記憶 體126之檔案快取列表，反應一檔案請求而藉處理器存取該 記憶體126。若1403中，無資料判定為儲存於記憶體126中， 則會由媒體112擷取檔案。在某些實例中，沒有任何檔案資 料儲存於記憶體126中之一般指示。在這些實例中，檔案資 料是否儲存於記憶體126中之指示可藉由區段基準而判定 位於一區段上·。例如，察看圖7之實例。 若1403為是的話，處理器116判定所有檔案資料是否儲存 於記憶體126中。若完整檔案儲存於記憶體126中，則在14〇6 中’處理器116由記憶體126擷取檔案。例如，圖4之實例， 在記憶體126中快取所有檔案資料。然而，因那些實例皆不 餘存完整檔案於記憶體126中，故可省略區塊1404及1406。 若1404為否的話，處理器116在1407中會判定使用者資料 之第一區塊是否儲存於記憶體126中。若為是的話，1411中， 處理器116由記憶體126擷取資料區塊部份及接著1412中.，由 媒體112區段擷取資料區塊其他部份（若有的話）。1413中， 處理器將由記憶體擷取之部份與由媒體112擷取之部份結 合。若結合資料區塊不含有檔案終止指示（EOF)，則處理器 116回到1407直到獲得所有資料為止。若1407為否的話，處 理器116獲得來自媒體112中區段之資料。 熟習此技藝者應了解，基於文中所教示，圖14所提出之 貝料相1取方法會依使用者資料儲存於媒體112及記憶體126 87246 -31 - 200413907 、方弋、而L正。例如，參照圖7之實例，依區段區塊（如， )尾、卩之^置貪訊而作出資料區塊為存取記憶體126或媒 體112之決定。力同1Λ 、a 在圖10及11惑貫例中及在圖8及9之實例中， 依：骨豆112中硬碟表面上之實際區段位置而決定是否存取 :己^拉126胃10及11之實例中，由媒體12#頁取資料區塊之 4伤私式，包括比較區段位置與一區段位置表（未示），其 包括記憶體126 (如，位置1116)中之毁壞空間及位址區段位 置（如1〇12)，孩記憶體126為儲存這些區段區塊資料剩餘 4伤之處。圖· 8及9之實例中，處理器1 “存取一毁壞區段 表’其·包括儲存各毀壞區段之資料區塊的記憶體126中之位 址二該表儲存於記憶體126,程式記憶體124,或其他可由處 理器116所存取之記憶體（未示）。 圖15為根據本發明之抽取式媒體資料儲存系統之實例方 鬼圖抽取式媒體資料儲存系統141〇由抽取式媒體HU擷取 (及在某氪貝例中為寫入）使用者資料，該抽取式媒體Μ U 置=資料儲存系統及提供使用者資料至資訊裝置1541 (及 在某，實例中，由資訊裝置1541擷取使用者資料以儲存）。 ，只例中，資料儲存系統1510為一光碟（CD)機，為只可 項取或可讀取/寫人。在其他實例中，資料儲存系統⑸何 為下列之一：DVD播放機，手提CD播放機，光碟機，磁帶 機，抽取式軟碟機，具有抽取式硬碟之系統，或其他任一 把夠接收抽取式媒體之儲存系統。一般抽取式媒體之設計 為可由抽取式媒體資料儲存系統移除之任何媒體。在所示 實例中，抽取式媒體設計為可由資料儲存系統151〇移動， 87246 -32- 200413907 該資料儲存系統1510包括讀寫頭及轉軸馬達1514,其中 一其他抽取式媒體可置入其位置。然而，在某些實例中， 抽取式媒體包括（且亦可移除）其他電路（圖ls所示），如讀寫 頭或驅動電路（如，轉軸馬達）。 固15所示之貝例，貝料儲存系統J 5 J 〇包括控制電路15工工 以控制儲存系統1510之操作。控制電路15ιι包括一處理器 1516及一程式記憶體1524以儲存由處理器MM執行之編 碼控制電路1511亦包括一非揮發性IC式記憶體1526，以儲 存關於抽取式·媒體（如，1512)之運算資料。在某些實例中， 記憶^ 1S26以MRAM記憶體實作。在一實例中，控制電路 1511實作資一電路板（未示）上，其位於資料儲存系統之 外罩（未π )内。在某些實例中，記憶體1524及記憶體Μ%在 同一記憶體陣列中實作。 、在圖15所示之實例中，記憶體1526包括一表格1527，其關 於各抽^取式媒體儲存兩類型之運算資料，系統資料（儲存於 系’充〃料攔位1532)及利用資料（儲存於利用資料欄位 = 33)。系統資料1532之範例包括主列表，毁壞區段列表， 禁用區段列表，毁壞區段取代列表，備份保護特徵，或其 他相似類型〈特定用於抽取式媒體（如，1512)類型之系統資 料。在其他實例中，配置記憶體1526以儲存抽取式 系統資料或利用資料之其一。 、'利：資料為關於資料儲存中之媒體使用之資訊。利用資 料包括，如，存取媒體之次數，存取儲存於媒體上 、各榣木之’入數，存取媒體上特定檔案之次數，媒體之最 87246 -33 - 200413907 後存取位置’ H系統15_存取任何媒體之次數，要求伊 案〈使料，關於-檑案資料之錯誤檢測與糾正及安全: 料。例如，在卿播收機中，—類型之利用資料為—磁碟 《最後位置，在使用者截止系統前會播放該磁碟。 在圖丄5之實例中，因預先置入資料像存系統⑽之數個抽 取式媒體而配置表格1527以儲存***資料（搁幻议中）及 利用資料（欄位1533中）。因此，若—料之抽取式媒體置入 資料儲存系統W0，處理器⑸6可由表格助存取關於肝 之操作資料。.例如，若資料儲存系統151〇為一卿播放機， 表格助可用以儲存最後播放位置，其藉由該播放機播放 最後20個DVD。在此—系統κ用者可在影片之特定胃占 停止播放機及移動媒體。將來，使用者可再置入媒體，其 中處理益1516可存取欄位1533之利用資料以判定影片停止 ，最後位置。在此-系統中，DVD播放機可提供使用者由 最後停止位置重新播放之選擇。 表格1527之各條目（如，1542)包括一山欄位1531，其儲存 與條目内資料相聯之抽取式媒體的識別指示。各條目亦包 括一系統資料欄位1532 (如，以A表示之條目1542),其中儲 存與抽取式媒體相聯之系統資料及一利用資料攔位（如，以 Μ表不之條目1542)，其中儲存與抽取式媒體之利用資料。 處理态1516利用ID攔位1531以判定一媒體是否已預先置入 系統1510。若已置入媒體，處理器1516由與其ID相聯之條目 擷取抽取式媒體之其運算資料。 資料儲存系統1510包括往抽取式媒體1512的資料存取之 87246 -34- 存取電路。在圖I5之實例中，存取電路包括一讀寫頭1513， 用以由抽取式媒體15丨2讀取資料及用以寫入（於可寫入至抽 取式媒體之系統中）資料至抽取式媒體1512。耦合讀寫頦 1513以由媒體1512藉讀取信號線及放大器1518提供擷取資 料至處理器1516及藉寫入信號線及放大器152〇提供擷取資 料至寫入媒體1512之擷取資料。處理器1516藉一頭控信號線 至驅動器而控制讀寫頭1513之移動。處理器1516亦可藉馬達 控制線及放大器1522所傳送之信號控制轉軸馬達1514之旋 轉。. . 圖16為藉資料儲存系統151〇利用一抽取式媒體之方法實 例之流程圖。1603中，反應置入一抽取式媒體（如，1512 ) &lt;指示，處理器1516在1605中由抽取式媒體擷取一識別指示 (ID)。在一實例中，；[D對特定抽取式媒體為全世界獨一無二 的。例如，若系統15 1 〇為一硬碟機，各抽取式硬碟包括一 獨特之序列號碼，其對媒體群為獨特的，如一批號。在其 他實例中，識別資訊對儲存於硬碟之使用者資訊亦為獨特 的。例如在一實例中，ID可為一影片名稱或音樂^〇名稱或 音樂CD之UPC碼。在其他實例中，出可為藉由使用者添加 之號碼。 1607中，處理益1516將來自置入抽取式媒體之仍與儲存於 攔位15 3 i之Z D相比較以判定搁位】5 2 7是否、包括肖置入抽取 式媒體相聯之運算資料。若在攔位1527中發現置人抽取式 媒體之ID，則1609中，處理器1516擷取儲存於條目之運算資 料。若在欄位1527中未發現出，則處理器1516由1512中之置 87246 -35- 200413907 入抽取式媒體擷取任何運算資料，如主列表，毁壞區段及 禁用檔案。在一實例中，可在抽取式媒體之啟動區段上發 現此一資訊。 1612中，處理器1516存取欄位1527以判定欄位1527中是否 具有足夠的空間以儲存置入抽取式媒體之運算資料。在一 實例中’判斷欄位1527之大小以保持抽取式媒體之預設數 目之運算資料（如，100)，該抽取式媒體置入資料儲存系統 1510。若抽取式媒體之預設數目的運算資料已儲存於記憶 1526之欄位1527中，則1613中，處理器1516選擇攔位1527中 之表格條目以覆寫ID及置入抽取式媒體之運算資料。 在Λ例中，處理益1516選擇最後使用的抽取式媒體之 表格條目。在另_實例巾，處理器咖覆窝最近使用的抽 取式媒體之表格條目。在更—實例中，可基於各種標準選 擇指定覆寫之表格條目，該標準包括置人頻率及抽取式媒 體的置^順序兩者。在1613之後或若1612為是的話，1614中 (處理器1516會儲存記憶體1526之欄位切中的置人媒體之 ID及運算資料。 j 1615中，使用儲存於攔位1527之運算資料以操作資料儲 存系統1510。力_ ^ , 六 、 貝例中’運算資料包括一由處理器1516 存取之主列表，戈炎令 .^ 〃局儲存於置入抽取式媒體中之檔案資料 :二看關於圖1之資料儲存系統m實例之主列表的應 -2二4淪。亦如關於資料儲存系統110所討論的，系統 貝枓包括毀壞區段丨 ^ ^八 j表，示用區段列表，毀壞區段取代列 表’備份保護特徵，+甘^ ^ 或其他任何藉處理器1516使用以存取 87246 -36- 200413907 使用者資料於抽取式媒體之資料。熟習於此技藝者應了 解，基於文中所教示，可依據使用者資料儲存於抽取式媒 體之方法決定儲存於欄位1527中之類型系統資料u如， 在CD儲存歌曲時，主列表包括哪些磁軌儲存某些歌曲。 在抽取式媒體資料儲存系統之非揮發性圯式記憶體中的 儲存系統資料之優點為，使資料儲存系統可更快速地由抽 取式媒體㈣資料。再者’因為記憶體為非揮發性，多重 抽取式媒體之運算資料可儲存於攔位助中，且在電源中 斷操作後依然.可取得該運算資料。因此，此一系統^具有 -優點，即為可縮短存取資料所需時間，因當處理器⑽ 識別抽取式媒體之肋寺，無需負載抽取式媒體之系統資 料，處理器1516即可開始由抽取式媒體存取使用者資料。 另-優點來自於包括記憶體1526之主列表，即為可減少 在資料寫人之抽取式媒體中之存取次數。—般，部份抽取 式媒體為可寫入的，每一抽取式媒體之寫入存取需要更新 抽取式媒體上之主列表。然而在資料料系統咖之某些 實例中，因主列表可儲存在記憶體1526之中，儲存在肝 郎上之主列表沒有必要在每次窝人時即更新。在—實例 中，只有當抽取式媒體1512由系統151〇移動時可更新抽 媒體1512之主列表。 間’處理器1516產生 k 1527中。例如，處 目（如，1542)的利用 ’執行檔案）儲存於 1616中，在資料儲存系統151〇操作期 運算資料及存該資料於記憶體丨526之攔 理器1516可儲存於抽取式媒體之表格條 欄位1533’會存取檔案（如，歌曲，影片 87246 -37- 200413907 媒體1512上之次數。在一實例中，可依求提供此資訊至資 訊裝置1S41。在另一實例中，運算資料可用以快取至記情 體1526中，由媒體1512之最常快取檔案用以縮短存取時間以 提供該檔案至資訊裝置1541。於1616產生的其他類型之利用 資料包括利用檔案之次數。 當在1617判定媒體1512為由資料儲存系統151〇移動時，處 理器1516儲存關於欄位1527中抽取式媒體1512之使用者終 止的運异資料。在一貫例中’主列表儲存系統丨5丨〇包括一 感應器（未示）_，以當由資料儲存系統1510移動置入抽取式媒 體（如，1512)可感應得到。使用運算資料終止之範例包括 最後存取之檔案，最後存取之時間，檔案存取之事實及置 入抽取式媒體之事實，使用系統之時間，存取資料之使用 者。 热習於此技藝者應了解，基於文中所教示，可以上述關 於記憶體126相同方式利用記憶體1526。例如，參照圖4，可 利用記憶體1526以快取最近存取或最常存取之檔案，其儲 存於抽取式媒體1512。因此，因不需要存取媒體1512以擴取 良好使用之檔案，故可大幅縮短檔案之存取時間。此外， 右寫入至媒體1512，則至媒體1512之寫人可在記憶體⑸时 快取二因而允許一隨後之資料存取。在此一實例中，若有 要求為移動媒體，則還未位於媒體1512中之記憶體1526 中的所有資料在移動前可寫入至媒體1512。 、,、、再者，關於抽取資料儲存系統1510之描述特徵亦可以資 料儲存系統1 10之實例眘 w ^ 只川貝作。例如，配置資料儲存系統110 87246 •38- 200413907 以儲存利用資料於記憶體126中。例如，配置資料儲存系統 110以儲存儲存於媒體丨12的檔案之存取及/或修正次數。 热習此技勢者應了解其於文中教示之顯示及描述本發明 的明確貝例，在本發明及其較宏觀之觀點内可做改變及改 良，且因此，如同本明之精神及範圍内，亦可在後附之申 叫專利範轉内做此一改變及改良。 【圖式簡單說明】 熱習此技藝者藉由參考附圖會可更了解本發明及更清楚 本發明眾多的·目的，特徵及優點。 圖1為根據本發明資料儲存系統之實例方塊圖。 圖2所不為以符號表示之主列表之習用實作。 圖3所TF為根據本發明執行資料存取之資料儲存Before that, the head 113 can be moved, and the data from the 112 section of the media can be read and combined. Because the tail part of the data section block is stored in the section 113 of the storage block section, the time required to retrieve a file in the storage system 11 is reduced. In other examples, the beginning of the file is stored in memory 126 and the remainder of the file is stored in media 112. Therefore, when a file request is issued, the processing benefit can provide the beginning of the file to the information device 141, and move the read / write head to retrieve the remainder of the data. In one example, a preset size (for example, 100 bits, 1000 bits, or 20 data blocks) of the beginning of a file may be stored in the memory 126. It will respond to a read request and determine the size of the starting part based on the time required for the data storage system to retrieve data from the media 112. Fig. 14 shows an example of data read access operation stored in the data storage system 110. The method proposed in FIG. 14 is used with modification to retrieve user data stored in both media 112 and 1 memory 126, as shown in the example and with reference to FIGS. 4, 7, 8, 9, 10, 11 , 12 and 13 are described in the text and related documents of these figures, as other examples are not explicitly described in the text. In 1400, the processor 116 starts executing the method proposed in FIG. 14 in response to receiving a file read request from the information device 141. In 1403, the processor 116 determines whether 87246-30-30200413907 file data is stored in the memory 126. In some examples, an indication is whether the file data is stored in the memory 126, and the indication is found in the file entry in the master list 127 '. In other examples, the memory 126 includes a file cache list cached in the memory 126, which is accessed by the processor in response to a file request. If no data is determined to be stored in the memory 126 in 1403, the file is retrieved by the media 112. In some instances, there is no general indication that file data is stored in memory 126. In these examples, the indication of whether or not the archive data is stored in the memory 126 can be determined on a sector by the sector reference. For example, look at the example of FIG. 7. If yes at 1403, the processor 116 determines whether all file data is stored in the memory 126. If the complete file is stored in the memory 126, the processor 116 retrieves the file from the memory 126 in 1406. For example, in the example of FIG. 4, all file data is cached in the memory 126. However, since those examples do not have a complete file remaining in memory 126, blocks 1404 and 1406 may be omitted. If 1404 is no, the processor 116 determines in 1407 whether the first block of user data is stored in the memory 126. If so, the processor 116 in 1411 retrieves the data block portion from the memory 126 and then in 1412. The media 112 section retrieves the rest of the data block (if any). In 1413, the processor combines a portion retrieved from the memory with a portion retrieved from the medium 112. If the combined data block does not contain an EOF, the processor 116 returns to 1407 until all data is obtained. If 1407 is no, the processor 116 obtains data from the sectors in the media 112. Those skilled in this art should understand that based on the teachings in this article, the method for extracting shellfish phase 1 proposed in Figure 14 will be stored in media 112 and memory 126 87246 -31-200413907, Fang Yi, and L Zheng according to user data. For example, referring to the example of FIG. 7, a decision is made to access the memory 126 or the media 112 according to the location of the block block (e.g., the end of the block). The force is the same as 1Λ and a. In the examples of FIGS. 10 and 11 and in the examples of FIGS. 8 and 9, it is determined by the actual segment position on the hard disk surface in Bone Bean 112. In the example of 10 and 11, the 4th private type of data block is fetched from page 12 # of the media, including comparing the section position with a section position table (not shown), which includes memory 126 (eg, position 1116). Destroyed space and address segment location (such as 1012), the child memory 126 is the remaining 4 wounds for storing the segment block data. Figures 8 and 9. In the example, the processor 1 "accesses a corrupted segment table 'which includes the address in memory 126 that stores the data blocks of each corrupted segment. The table is stored in memory 126. Program memory 124, or other memory (not shown) that can be accessed by processor 116. Figure 15 is an example of a removable media data storage system according to the present invention. The media HU captures (and writes in some examples) user data. The extractable media M U is set to a data storage system and provides user data to the information device 1541 (and in an example, by The information device 1541 retrieves user data for storage). In the example, the data storage system 1510 is a compact disc (CD) player, which is only accessible or readable / writeable. In other examples, the data storage system What is one of the following: DVD player, portable CD player, optical disc drive, tape drive, removable floppy disk drive, system with removable hard disk, or any other storage system capable of receiving removable media. Design of general removable media Any media removed by the removable media data storage system. In the example shown, the removable media is designed to be removable by the data storage system 1510, 87246 -32- 200413907 The data storage system 1510 includes a read / write head and a spindle motor 1514 One of the other removable media can be placed in its position. However, in some instances, the removable media includes (and can also be removed) other circuits (shown in Figure ls), such as a read / write head or drive circuit (such as , Shaft motor). The shell example shown in Figure 15, the shell material storage system J 5 J 〇 includes a control circuit 15 workers to control the operation of the storage system 1510. The control circuit 15 includes a processor 1516 and a program memory 1524 to The encoding control circuit 1511 stored by the processor MM also includes a non-volatile IC-type memory 1526 to store operational data on removable media (eg, 1512). In some examples, the memory ^ 1S26 is MRAM Memory implementation. In one example, the control circuit 1511 is implemented on a circuit board (not shown), which is located in the outer cover (not π) of the data storage system. In some examples, the memory 1524 and memory M% are implemented in the same memory array. In the example shown in FIG. 15, the memory 1526 includes a table 1527, which stores two types of operation data and system data on each extractable media. (Stored in the system's filling station 1532) and utilization data (stored in the utilization data field = 33). Examples of system data 1532 include the main list, the destroyed sector list, the disabled sector list, and the destroyed sector replacement list. , Backup protection features, or other similar types of system data specific to the type of removable media (eg, 1512). In other examples, the memory 1526 is configured to store one of the removable system data or the utilization data. , 'Profit: Data is information about the use of media in data storage. The use of data includes, for example, the number of accesses to the media, the number of accesses stored on the media, the number of accesses to each cypress, the number of accesses to specific files on the media, and the maximum access position of the media after 87246 -33-200413907. H System 15_ The number of times any media was accessed, requested the Iraqi case <Materials, error detection and correction of-case data, and security: materials. For example, in the broadcast receiver, the type of utilization data is-disk "last position, the disk will be played before the user cuts off the system. In the example in Figure 丄 5, the form 1527 is configured to store system data (under discussion) and utilization data (in column 1533) because several removable media of the data image storage system 预先 are pre-installed. Therefore, if the extractable media is inserted into the data storage system W0, the processor ⑸6 can access the operation data about the liver through the form. For example, if the data storage system 1510 is a player, the form helper can be used to store the last playback position, which is used to play the last 20 DVDs by the player. Here—System users can stop the player and mobile media at specific locations in the movie. In the future, the user can re-enter the media, in which the processing data 1516 can access the utilization data of field 1533 to determine the video stop and the last position. In this system, the DVD player can provide the user with the option to restart playback from the last stopped position. Each entry (e.g., 1542) in Form 1527 includes a mountain field 1531 that stores identification instructions for removable media associated with the data in the entry. Each entry also includes a system data field 1532 (eg, entry 1542 indicated by A), which stores system data associated with removable media and a use data block (eg, entry 1542 indicated by M), Which stores and extracts the use of media. The processing state 1516 uses the ID block 1531 to determine whether a media has been previously placed in the system 1510. If the media has been inserted, the processor 1516 retrieves its operation data of the removable media from the entry associated with its ID. The data storage system 1510 includes 87246 -34- access circuits for data access to the removable media 1512. In the example of Figure I5, the access circuit includes a read / write head 1513 for reading data from the removable media 15 and 2 and for writing (in a system that can be written to the removable media) data to extraction 1512 style media. Coupling read / write 颏 1513 provides the retrieval data to the processor 1516 by the read signal line and amplifier 1518 from the media 1512 and the retrieved data from the write medium 1512 by the write signal line and amplifier 1520. The processor 1516 controls the movement of the read / write head 1513 by a head control signal line to the driver. The processor 1516 can also control the rotation of the shaft motor 1514 by a signal transmitted from the motor control line and the amplifier 1522. Fig. 16 is a flowchart of an example of a method for using a removable media by the data storage system 1510. In 1603, a removable media (e.g., 1512) is placed in response to an instruction, and the processor 1516 in 1605 retrieves an identification instruction (ID) from the removable media. In one example, [D is unique worldwide for a particular extractable media. For example, if the system 15 10 is a hard disk drive, each removable hard disk includes a unique serial number, which is unique to the media group, such as a batch number. In other instances, identification information is also unique to user information stored on the hard disk. For example, in an example, the ID may be a movie name or a music title or a UPC code of a music CD. In other examples, the number may be a number added by the user. In 1607, the processing benefit 1516 compares the data from the inserted removable media with the ZD stored in the stop 15 3 i to determine the shelving] 5 2 7 Whether or not it includes the operation data associated with the inserted removable media. If the ID of the removable media is found in block 1527, then in 1609, the processor 1516 retrieves the operation data stored in the entry. If it is not found in the field 1527, the processor 1516 retrieves any operational data from the removable media in the location 15246 87246 -35- 200413907, such as the main list, destroy the sector and disable the file. In one example, this information can be found on the boot section of removable media. In 1612, the processor 1516 accesses the field 1527 to determine whether there is sufficient space in the field 1527 to store the operation data placed in the removable media. In one example, the size of the field 1527 is judged to maintain a preset number of operational data (e.g., 100) of the removable media, which is inserted into the data storage system 1510. If the preset number of calculation data of the removable media has been stored in the field 1527 of the memory 1526, then in 1613, the processor 1516 chooses to block the table entry in 1527 to overwrite the ID and the operation data of the removable media. . In the Λ example, the process 1516 selects the table entry of the last removable media used. In another example, the processor has a table entry for the most recently used removable media. In a further example, the form entry for the specified override can be selected based on various criteria, including both the frequency of placement and the order of placement of the removable media. After 1613 or if 1612 is true, in 1614 (the processor 1516 will store the ID and operation data of the human media cut in the field of memory 1526. j In 1615, the operation data stored in block 1527 is used to operate Data storage system 1510. Force_ ^, VI. The calculation data in the example includes a master list accessed by the processor 1516, Ge Yanling. ^ The file data stored in the removable media by the bureau: Second look The master list of the instance of the data storage system m in FIG. 1 should be -2. 2. Also as discussed with regard to the data storage system 110, the system includes a destroyed section table, and a list of sections is shown. Destroyed section replaces the list 'backup protection feature, + gan ^ ^ or any other borrowed by processor 1516 to access 87226 -36- 200413907 user data in removable media data. Those skilled in this art should understand that based on The teaching in the text can determine the type of system data stored in field 1527 according to the method of storing user data in removable media. For example, when a CD stores songs, which tracks are included in the main list to store certain songs. The advantage of storing system data in non-volatile memory of media data storage system is that the data storage system can more quickly extract data from removable media. Furthermore, because the memory is non-volatile, multiple extraction The operation data of the media can be stored in the stopper, and it can still be obtained after the power is interrupted. Therefore, this system ^ has the advantage of shortening the time required to access the data because it should be processed. The device identifies the ribbed temples of the removable media. Without loading the system data of the removable media, the processor 1516 can start to access the user data from the removable media. Another advantage comes from the master list including the memory 1526, which is It can reduce the number of accesses in the removable media of the data writer.—Generally, some removable media are writable, and the write access of each removable media needs to update the master list on the removable media. However, in some instances of the data system, since the master list can be stored in the memory 1526, the master list stored on the Ganlang does not need to be updated each time the people are nested. —In the example, the master list of the extracted media 1512 can be updated only when the removable media 1512 is moved by the system 1510. The 'processor 1516 generates k 1527. For example, the use of the purpose (eg, 1542)' to execute the file) Stored in 1616, the data is calculated during the 1510 operation period of the data storage system, and the data is stored in the memory. 526, the processor 1516 can be stored in the form field 1533 'of the removable media, and will access the file (eg, song , Video 87246 -37- 200413907 Media 1512. In one example, this information can be provided to the information device 1S41 on demand. In another example, the operation data can be cached to the memory 1526 by the media The most commonly cached file of 1512 is used to shorten the access time to provide the file to the information device 1541. Other types of utilization data generated in 1616 include the number of times archives were used. When it is determined in 1617 that the medium 1512 is moved by the data storage system 1510, the processor 1516 stores the termination data for the user of the removable medium 1512 in the field 1527. In the conventional example, the 'master list storage system 5' 5 includes a sensor (not shown) so that it can be sensed when the data storage system 1510 moves into the removable media (eg, 1512). Examples of the use of computing data termination include the file last accessed, the time of last access, the fact that the file was accessed and the fact that the removable media was inserted, the time the system was used, and the user who accessed the data. Those skilled in the art should understand that based on the teachings in the text, the memory 1526 can be used in the same manner as the memory 126 described above. For example, referring to FIG. 4, memory 1526 may be utilized to cache recently accessed or most frequently accessed files, which are stored in removable media 1512. Therefore, since it is not necessary to access the medium 1512 to expand a well-used file, the access time of the file can be greatly shortened. In addition, the right write to the media 1512, then the writer to the media 1512 can cache two at the time of memory, thus allowing a subsequent data access. In this example, if there is a requirement for mobile media, all data in the memory 1526 that is not yet in the media 1512 can be written to the media 1512 before moving. ,,,, and again, the description of the extracted data storage system 1510 can also be used as an example of the data storage system 1 10 carefully. For example, the data storage system 110 87246 • 38- 200413907 is configured to store utilization data in the memory 126. For example, the data storage system 110 is configured to store the number of accesses and / or corrections of files stored in the media 12. Those who are eager to learn this skill should understand the clear examples of the teachings shown and described in the text, which can be changed and improved within the present invention and its broader perspective, and therefore, as within the spirit and scope of the present invention, This change and improvement can also be made in the attached patent application. [Brief description of the drawings] Those skilled in the art can better understand the present invention and understand it more clearly by referring to the accompanying drawings. The numerous objects, features, and advantages of the present invention. FIG. 1 is a block diagram of an example of a data storage system according to the present invention. Figure 2 is not a conventional implementation of the master list represented by symbols. The TF in Figure 3 is the data storage for performing data access according to the present invention.

一資訊裝置儲存使用者資料。 圖5為根據本發明由一媒# 媒體之最佳化區段擷取系統之實 例符號圖。 一媒體之最佳化區段之實例流程圖。An information device stores user data. Fig. 5 is a symbolic diagram of an example of an optimized segment extraction system from a media # media according to the present invention. An example flowchart of an optimization section of a media.

87246 圖6為根據本發明由一 圖7為以符號表示根據 -39- 20041390787246 Fig. 6 shows a symbol according to the present invention.

圖l 〇為根據本發明 之硬碟表面上之區段佈局之 η例俯視 圖η為根據本發 、、、 性ic式媒體内之使η、&amp;於貞訊儲存系統之非揮發 胃資料邵分㈣之錯存。 圖12為根據本發明 11 4- /、枓舄入存取之實例流程圖。 固為根據本發明之資料寫、 ® 14Λ^« * ^ 仔取艾另一實例流程圖。 口為根據本發明之資料讀 〜 ffl 15Λ ϋ iit 、 〈貫例流程圖。 5為根據本發明 塊圖。 、木把具枓儲存系統之實例方 圖16為根據本發明藉由資料儲 之運算資料之實例流程圖。除^來利用抽取式壯 之同樣參考符號表示同樣項目。指明’否則不同圖中 圖式代表符號說明】 110 資料儲存系統 111 控制系統 112 媒體 113 寫入頭 114 轉軸馬達 116 處理器 118 放大器 119 放大器 120 放大器 122 放大器 124 程式記憶體 87246 -40. 200413907 126 記憶體 127 主列表 128 匯流排 131 頭驅動器 141 資訊裝置 205 主列表 206 區段區塊 208 區段區塊 210 區段_區塊 212 硬碟表面 215 ' 區段 216 區段 217 區段 302 流程圖區塊 304 判定菱形框 306 流程圖區塊 308 流程圖區塊 414 判定菱形框 416 判定菱形框 422 流程圖區塊 430 判定菱形框 432 流程圖區塊 434 流程圖區塊 501 硬碟表面 -41 87246 200413907 519 檔案 520 主列表輸入 521 資料區塊 522 資料區塊 523 資料區塊 524 資料區塊 525 資料區塊 526 區塊 531 區段. 532 區段 533 區段 534 區段 535 區段 536 區段 551 磁軌 552 磁軌 603 流程圖區塊 605 判定菱形框 607 流程圖區塊 609 流程圖區塊 613 流程圖區塊 619 判定菱形框 623 停止 627 流程圖區塊 87246 200413907 705 主列表 706 區段區塊 708 區段區塊 710 區段區塊 712 硬碟表面 714 區段區塊 807 硬碟表面 808 硬碟表面 809 硬碟.表面 810 硬碟表面 821 區段 822 區段 823 區段 824 區段 901 硬碟系統 904 硬碟 905 硬碟 1010 硬碟表面 1011 位置 1012 空間 1015 部分 1016 空間 1020 集線器 1023 部分FIG. 10 is a top view of an example of a section layout on the surface of a hard disk according to the present invention. Η is a non-volatile stomach data in a storage system according to the present invention. Mistakes of tillering. Fig. 12 is a flowchart of an example of a 4- /, entry access according to the present invention. In accordance with the information of the present invention, ® 14Λ ^ «* ^ is another example flowchart. The port is read according to the material of the present invention ~ ffl 15Λ ϋ iit 〈Continuous flow chart. 5 is a block diagram according to the present invention. Example of a wooden handle storage system Fig. 16 is a flowchart of an example of computing data by data storage according to the present invention. Divide ^ to use the same reference symbols for extractables to denote the same items. Indicate 'otherwise the representation of the symbols in different figures] 110 data storage system 111 control system 112 media 113 write head 114 spindle motor 116 processor 118 amplifier 119 amplifier 120 amplifier 122 amplifier 124 program memory 87246 -40. 200413907 126 memory Body 127 Master list 128 Bus 131 Head drive 141 Information device 205 Master list 206 Section block 208 Section block 210 Section_block 212 Hard disk surface 215 'Section 216 Section 217 Section 302 Flow chart area Block 304 decision diamond 306 flowchart block 308 flowchart block 414 decision diamond box 416 decision diamond box 422 flowchart block 430 decision diamond box 432 flowchart block 434 flowchart block 501 hard disk surface -41 87246 200413907 519 file 520 master list input 521 data block 522 data block 523 data block 524 data block 525 data block 526 block 531 section. 532 section 533 section 534 section 535 section 536 section 551 section magnetic Orbit 552 magnetic track 603 flowchart block 605 decision diamond box 607 flowchart block 609 flow Figure block 613 flowchart block 619 decision diamond box 623 stop 627 flowchart block 87246 200413907 705 main list 706 block block 708 block block 710 block block 712 hard disk surface 714 block block 807 hard Disk surface 808 Hard disk surface 809 Hard disk. Surface 810 Hard disk surface 821 Segment 822 Segment 823 Segment 824 Segment 901 Hard disk system 904 Hard disk 905 Hard disk 1010 Hard disk surface 1011 Position 1012 Space 1015 Part 1016 Space 1020 Hub 1023 part

87246 -43- 200413907 1024 空間 1029 區段 1031 區段線 1032 區段線 1115 記憶體位置 1116 位置 1130 區塊 1201 開始 1203 流程_圖區塊 1205 流程圖區塊 1207 流程圖區塊 1211 流程圖區塊 1213 流程圖區塊 1215 流程圖區塊 1219 判定菱形框 1221 流程圖區塊 1227 停止 1301 開始 1303 流程圖區塊 1304 流程圖區塊 1305 判定菱形框 1309 流程圖區塊 1310 流程圖區塊 1311 流程圖區塊 87246 200413907 1314 流程圖區塊 1315 流程圖區塊 1400 開始 1403 判定菱形框 1404 區塊 1406 流程圖區塊 1407 判定菱形框 1411 流程圖區塊 1412 流程_圖區塊 1413 流程圖區塊 1414 判定菱形框 1415 停止 1510 資料儲存系統 1511 控制電路 1514 轉軸馬達 1516 處理器 1518 放大器 1519 放大器 1520 放大器 1522 放大器 1524 程式記憶體 1526 記憶體 1527 表 1530 驅動器 87246 200413907 1531 ID欄位 1532 系統資料欄位 1533 欄位 1541 資訊裝置 1542 輸入 1603 判定菱形框 1605 流程圖區塊 1607 判定菱形框 1609 流程肩區塊 1612 判定菱形框 1613 流程圖區塊 1614 流程圖區塊 1615 流程圖區塊 1616 流程圖區塊 1617 判定菱形框 -46- 8724687246 -43- 200413907 1024 space 1029 section 1031 section line 1032 section line 1115 memory position 1116 position 1130 block 1201 start 1203 flow_Figure block 1205 flow block 1207 flow block 1211 flow block 1213 flowchart block 1215 flowchart block 1219 decision diamond 1221 flowchart block 1227 stop 1301 start 1303 flowchart block 1304 flowchart block 1305 decision diamond box 1309 flowchart block 1310 flowchart block 1311 flowchart Block 87246 200413907 1314 Flow chart block 1315 Flow chart block 1400 Start 1403 Decision diamond box 1404 Block 1406 Flow chart block 1407 Decision diamond box 1411 Flow chart block 1412 Flow_Figure block 1413 Flow chart block 1414 Decision Diamond frame 1415 Stop 1510 Data storage system 1511 Control circuit 1514 Spindle motor 1516 Processor 1518 Amplifier 1519 Amplifier 1520 Amplifier 1522 Amplifier 1524 Program memory 1526 Memory 1527 Table 1530 Driver 87246 200413907 1531 ID field 1532 System data field 1533 Field 1541 Information devices 1542 Enter 1603 decision diamond box 1605 flowchart block 1607 decision diamond box 1609 flow shoulder block 1612 decision diamond box 1613 flowchart block 1614 flowchart block 1615 flowchart block 1616 flowchart block 1617 decision diamond box -46- 87246

Claims (1)

拾、申請專利範圍·· L 一種硬碟系統，包括·· 一硬碟’配置以儲存使用者資料，其中可配置一主列 表以識別至少一資料區塊之實質位置，該資料與使用者 資料檔案相聯；及 一非揮發性積體電路式記憶體，配置非揮發性積體電 路式A彳思體以儲存與硬碟相聯之主列表。 戈申W專利範圍第i項之硬碟系統，其中非揮發性積 路式記憶體更包括： A、 磁電阻式隨機存取記憶體（MRAM)。 3· —種儲存系統，包括： &quot;非積體電路式媒體，配置以儲存儲存系統之使用者 貝料，其中可配置系統資料以識別至少一資料區塊之會 質位置，該資料與儲存使用者資料之檔案相聯；及 一非揮發性積體電路式記憶體，配置非揮發性積體電 路式記憶體以儲存與非積體電路式媒體相聯之系 寺牛° 、，、 4. 一種系統，包括： 一處理器；及 一記憶體陣列，可耦合至用以非揮發性儲存程式資料 及使用者資料之處理器，其中程式資料及使用者資料可 以長式電壓運作’其具有不大於5.5伏特之強度。 5. —種抽取式媒體儲存系統，包括： 存取電路，用以當抽取式媒體耦合至抽取式媒體儲存 系統時，從抽取式媒體讀取使用者資料；及 87246 200413907 一非積體電路式記憶體，配置非積體電路式記憶體以 儲存與抽取式媒體相聯之資料，該媒體耦合至抽取式媒 體儲存系統。 6. 如申請專利範圍第5項之抽取式媒體儲存系統，其中非揮 發性積體電路式記憶體包括一磁電阻式隨機存取記憶體 (MRAM)。 7. —種將資訊儲存於抽取媒體儲存系統之方法，包括： 從一抽取式媒體讀取使用者資料，該抽取式媒體耦合 至抽取式媒體儲存系統；及 儲存與抽取式媒體相聯之運算資料，該抽取式媒體位 於抽取式媒體儲存系統之非揮發式積體電路式記憶體 内。 8. —種抽取式媒體儲存系統，包括： 抽取式媒體存取電路，當抽取式媒體耦合至抽取式媒 體儲存系統以存取抽取式媒體上之使用者資料時，配置 抽取式媒體存取電路以從抽取式媒體讀取使用者·資訊； 及 一非揮發性積體電路式記憶體，非揮發性積體電路式 記憶體儲存與抽取式媒體相聯之運算資料，該抽取式媒 體已耦合至系統。 9. 一種資料儲存系統，包括·· 一輸出，配置以從一資訊來源接收使用者資料； 一非積體電路式儲存媒體，配置以儲存使用者資料； 及 87246 200413907 一積體電路式非揮發性記憶體（NVM)，配置非揮發性 記憶體亦以儲存使用者資料。 10. —種資料儲存系統，包括·· 一非積體電路式儲存媒體，配置以儲存使用者資料； 一處理器，耦合至非積體電路式媒體；及 一積體電路式非揮發性記憶體（NVM)，其耦合至處理 器，配置非揮發性記憶體亦以儲存使用者資料。 11. 一種資料儲存方法，包括： 從資訊來源接收使用者資料； 儲存部份之使用者資料於一非積體電路式儲存媒體； 及― 儲存其他部份之使用者資料於一積體電路式非揮發性 記憶體（NVM)。 87246Scope of patent application ... L A hard disk system, including a hard disk configuration configured to store user data, which can be configured with a master list to identify the physical location of at least one data block, the data and user data File association; and a non-volatile integrated circuit-type memory configured with a non-volatile integrated circuit-type A memory to store a master list associated with a hard disk. Goshen W's patent for the hard disk system in item i, in which non-volatile memory includes: A. Magnetoresistive random access memory (MRAM). 3. · A storage system, including: &quot; Non-integrated circuit media, configured to store user materials of the storage system, where system data can be configured to identify the meeting position of at least one data block, the data and storage User data file association; and a non-volatile integrated circuit-type memory configured with non-volatile integrated circuit-type memory to store non-integrated circuit-based media A system comprising: a processor; and a memory array coupled to a processor for non-volatile storage of program data and user data, wherein the program data and user data can operate at a long voltage. No greater than 5.5 volts. 5. A removable media storage system including: an access circuit for reading user data from the removable media when the removable media is coupled to the removable media storage system; and 87246 200413907 a non-integral circuit Memory, configured with non-integrated circuit memory to store data associated with removable media, which is coupled to a removable media storage system. 6. For example, the removable media storage system under the scope of patent application No. 5, wherein the non-volatile integrated circuit memory includes a magnetoresistive random access memory (MRAM). 7. —A method for storing information in a removable media storage system, including: reading user data from a removable media, the removable media being coupled to the removable media storage system; and operations associated with storing and extracting media Data, the removable media is located in the non-volatile integrated circuit memory of the removable media storage system. 8. —A removable media storage system, including: a removable media access circuit, which is configured when the removable media is coupled to the removable media storage system to access user data on the removable media. To read user information from removable media; and a non-volatile integrated circuit memory, which stores operational data associated with removable media, which is coupled To the system. 9. A data storage system comprising: an output configured to receive user data from an information source; a non-integrated circuit-based storage medium configured to store user data; and 87246 200413907 an integrated circuit-type non-volatile Non-volatile memory (NVM). Non-volatile memory is also configured to store user data. 10. A data storage system comprising: a non-integrated circuit-type storage medium configured to store user data; a processor coupled to the non-integrated circuit-type media; and a integrated circuit-type non-volatile memory A body (NVM), which is coupled to a processor and is configured with non-volatile memory to store user data. 11. A data storage method comprising: receiving user data from an information source; storing part of the user data in a non-integrated circuit storage medium; and ― storing other part of the user data in an integrated circuit type Non-volatile memory (NVM). 87246