TW200937445A - Mapping management methods and systems - Google Patents

Mapping management methods and systems Download PDF

Info

Publication number
TW200937445A
TW200937445A TW97107097A TW97107097A TW200937445A TW 200937445 A TW200937445 A TW 200937445A TW 97107097 A TW97107097 A TW 97107097A TW 97107097 A TW97107097 A TW 97107097A TW 200937445 A TW200937445 A TW 200937445A
Authority
TW
Taiwan
Prior art keywords
mapping
block
page
data
specific
Prior art date
Application number
TW97107097A
Other languages
Chinese (zh)
Other versions
TWI375229B (en
Inventor
Pei-Jun Jiang
Original Assignee
Via Tech Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Via Tech Inc filed Critical Via Tech Inc
Priority to TW97107097A priority Critical patent/TWI375229B/en
Publication of TW200937445A publication Critical patent/TW200937445A/en
Application granted granted Critical
Publication of TWI375229B publication Critical patent/TWI375229B/en

Links

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Mapping management methods and systems are provided. First, a first read command comprising mapping directory data, block mapping data and page mapping data is obtained. Then, a specific block mapping table is located from a plurality of block mapping tables according to the mapping directory data, and a first specific entry is located from the specific block mapping table according to the block mapping data, wherein the first specific entry comprises a mapping mode setting and block information. When the mapping mode setting is a page mapping mode, a specific page mapped block is located from a page mapped block table according to the block information, and a page mapping table is located based on the specific page mapped block. Thereafter, a second specific entry is located from the page mapping table according to the page mapping data, and a page of data is located from a storage unit according to the second specific entry.

Description

200937445 九、發明說明: 【發明所屬之技術領域】 本發明係有關於一種映射管理方法及系統,且特別有 關於一種可以結合區塊映射模式與頁面映射模式之映射管 理方法及系統。 【先前技術】 在對於儲存單元,如非揮發性記憶體進行儲存管理 時’資料通常係依據頁面映射(Page Mapping)、區塊映射200937445 IX. Description of the Invention: [Technical Field] The present invention relates to a mapping management method and system, and more particularly to a mapping management method and system that can combine a block mapping mode and a page mapping mode. [Prior Art] When storing and managing storage units such as non-volatile memory, the data is usually based on page mapping and block mapping.

(Block Mapping)、置換區塊(Repiacement Block)、日諸區 塊(Log Block)等模式進行儲存與定位。 在頁面映射模式中’儲存單元中的資料係以邏輯頁面 為單位進行儲存,且具有相應之頁面映射表。頁面映射表 中係以邏輯頁面為索引’且以實體頁面為表項(項目内 谷)。當採用頁面映射之儲存系統處理讀寫命令時,則可以 直接查找、以及更新頁面映射表以完成相關資料頁面之定 二=塊中’儲存單元中的資料係以 中⑽輯區塊為索引,且以實二=項&= 塊映射之儲料統處理讀寫命令時 及更新區塊映射表以完成相關資料 ^以 _命令’必須先分配-個區塊,且=的;:對 此分配之區塊中。:外=:4中其他的頁面複製至 除了資料區塊之外可以擁J广式係指每-邏輯區塊 卜7以擁有一串以鏈結形式存在之置換區 VICO7-OO58IO0/0608-A4l455-TW/Draft-Final 200937445 塊。每當重寫某一頁面時, 此頁面為空的第-個置換區 纟鏈結上從頭開始找相應 此置換區塊中的此頁面位;\。如果找到,便將資料寫入 端再增加一個置換區境,並則在鍵結的尾 此f面位t。日— π & 將貝枓寫入新增置換區塊中的 部的邏輯區塊,=!::=r個、多個相連或全 納需要新寫入邏輯區塊的資\=的日違區塊’用以容 ❹ ❹ 要掃描-㈣區塊表,以確\^理讀寫命令時,則需 資料在曰誌、區塊中),或者;區塊為乾淨的(沒有 表。 &者找到相應邏輯區塊的頁面映射 近年來’隨著積體電路枯 區塊尺寸已經大幅提高。此;、發展’儲存單元的頁面/ N層(與非)快閃記憶體中每個^的==單元’如 順序寫入,不能隨機寫入。 J頁面必須依照 式必須對於同-區塊的個別頁面進行隨機c魏模 寫、以區塊為單位刪除、以及在;頁面為單位讀 因此,採用區塊映射的儲存系統處理的特點, 原有資料區塊進行-次刪除。當 必須野 入性能大大降低。其次,頁面 ^大後將導致寫 複雜與繁亂。當空閒區塊用盡時,^合:=分饰非常 區塊的過程將會非常複雜。再者,日諸區,間 4頁面映射表的儲存空間,因此僅有最近 為了節 採用頁面映射。使得每次對於儲存單元進行讀寫二科頁面 ”、寺,都必 VIC07-0058IOO/0608-A41455-TW/Draft-Fmal 200937445 需對於-特定表格進行掃描,以確定 ❹(Block Mapping), Replacement Block (Repiacement Block), and Day Block (Log Block) modes are stored and located. In the page mapping mode, the data in the storage unit is stored in units of logical pages and has a corresponding page mapping table. In the page mapping table, the logical page is indexed' and the entity page is used as the entry (inside the project). When the page mapping storage system is used to process the read and write commands, the page mapping table can be directly searched and updated to complete the related data page. The data in the storage unit is indexed by the middle (10) block. And the real two = item &= block mapping storage system processing read and write commands and update the block mapping table to complete the relevant information ^ to the _ command ' must first allocate - a block, and =; In the allocated block. : The other pages in the outer =: 4 are copied to the data block. In addition to the data block, the J-type can refer to each of the logical blocks to have a series of replacement areas in the form of links. VICO7-OO58IO0/0608-A4l455 -TW/Draft-Final 200937445 block. Whenever a page is rewritten, this page is empty on the first permutation area 纟 link to find the corresponding page bit in the permutation block from scratch;\. If found, the data is written to the end and a replacement area is added, and the f-plane t is at the end of the key. Day — π & Write the bell to the logical block of the new block, =!::=r, multiple connected or inclusive, need to newly write the logical block of the resource\= Block 'to use ❹ ❹ To scan - (4) block table, to ensure that when reading and writing commands, the data is in the ambition, block; or; the block is clean (no table. & The page mapping of the corresponding logical block has been found in recent years. With the size of the block of the integrated circuit has been greatly improved. This; development of the 'storage unit page / N layer (and non-) flash memory in each ^ The == unit's are written sequentially and cannot be randomly written. The J page must be random c-mode write, block-by-block, and page-read for individual pages of the same-block. Therefore, using the characteristics of the block mapping storage system processing, the original data block is deleted-timed. When the wild-input performance is greatly reduced. Secondly, the page ^ will cause complicated writing and chaos. When the free block is exhausted When, ^合:= The process of sub-decorating very blocks will be very complicated. In addition, the Japanese areas The storage space between the four pages of the mapping table, so only the page mapping is used recently for the section, so that each time the storage unit is read and written two pages", the temple must be VIC07-0058IOO/0608-A41455-TW/Draft- Fmal 200937445 Need to scan for a specific table to determine ❹

於頁面映射的區域中’從而降低讀寫時的:身料是否位 強調可以結合區塊映射與頁面映射 人〃'。對於一些 言,其類似於日諸區塊模式,每次對於儲=映射模式而 時’亦必需對於-特定表格進行掃描,元進行讀寫 料是否位於頁面映射的區域中。由於— 疋欲讀寫之資 表格進行掃描’因此必須限制頁面:::,須對於特定 成特定表格的尺寸過大,造成掃°°域,以避免造 影響系統性能。然而,若頁面映二=時間過長、 射的區域痕容易被寫滿,使得處 ^過小’則頁面映 須即時完成一次或多次資料合、寫入命令之前,必 塊,亦嚴重影響系統的寫入性At 以獲取新的空閒區 u* 月 b 〇 【發明内容】 有鑑於此 _个私a杈供映射管理 可以結合區塊映射模式與頁面映射模式=:,月 表格,以維持系統的讀寫性能。 ·.,、/員知描特定 本發明實施例之映射管理 -讀取命令,其中第一 c百先’取得至少-第 料、-區塊映射資料、與-頁V:二括::射:: 射目錄資料由複數區塊映射表 、接者’依據映 :當設定與,資 由-頁面映射區塊表中定位式時,依據區塊資訊 荷疋頁面映射區塊,且透過 VIC07-0058I00/0608-A4l455-TW/Draft-Final 8 200937445 特疋頁面映射區塊定位至—頁面映射表 映射資料由頁面映射表中定位-第二特定:後’依據頁面 二特定項目由—儲存單元定位—資料頁面。、目’且依據第 本發明實施例之映射管理系統包括1 理模組。館存單元係依據-頁面映射模式與一處 式分別以至少-頁面與至少^區塊儲存資料射模 ❹ ❹ 传至少:第一讀取命令’其中第-讀取命令至少:C取 射目錄貝料、一區塊映射資料、與一頁:映 模組依據映射目錄㈣由複數區塊映、^枓。處理 塊=定射議特定心射== 與-區塊資訊。當映射模式設定為頁;:一映射模式設定 區塊資訊由-頁面映射區掩表中定3 映射區塊,且透過特定頁面映射區塊定位一寺疋頁面 表。處理模組依據頁面映射資料由頁面頁面映射 二特定項目’且依據第二狀項目 \表中定位一第 頁面。 田儲存早元定位一資料 本發明上述方法可以透過程式竭 中。當程式碼被機H載人執行時:收錄於實體媒體 發明之裝置。 f ^變成用以實行本 為使本發明之上述目的、特徵和 下文特舉實施例,並配合所附圖示,I犯更明顯易懂, 【實施方式】 料說明如下。 第1圖顯示依據本發明實施例之映 ~ 、t管理系統。In the area of the page map, which reduces the number of reads and writes: whether the body is bit-emphasis can be combined with block mapping and page mapping. For some words, it is similar to the day block mode, and each time it is stored for the mapping mode, it is necessary to scan the specific table, whether the material is read or written in the area of the page mapping. Since the 表格 read and write form is scanned, it is necessary to limit the page::: to a size that is too large for a particular form to cause a sweep of the domain to avoid affecting system performance. However, if the page is reflected twice, the time is too long, and the area of the shot is easily filled, so that the page is too small, then the page image must be completed one or more times before the data is combined and written, and the block will also seriously affect the system. Writeability At to obtain a new free area u* month b 〇 [Abstract] In view of this, a private map management can be combined with the block mapping mode and the page mapping mode =:, monthly table to maintain the system Read and write performance. ·, / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / :: The shooting directory data consists of a complex block mapping table and a receiver's mapping: when setting the location and resource-page mapping block table, the block is mapped according to the block information page, and through VIC07- 0058I00/0608-A4l455-TW/Draft-Final 8 200937445 Special page mapping block positioning to - page mapping table mapping data is located in the page mapping table - second specific: after 'by page two specific items by - storage unit positioning - Information page. The mapping management system according to the embodiment of the present invention includes a management module. The library unit is based on the -page mapping mode and the one-part mode to store data at least - page and at least ^ block respectively. At least: the first read command 'where the first read command at least: C take the directory Shell material, a block mapping data, and a page: the mapping module is based on the mapping directory (4) by the complex block mapping, ^ 枓. Processing block = fixed shot specific == and - block information. When the mapping mode is set to page;: a mapping mode setting block information is defined by the -page mapping area masking block 3, and a temple page table is located through a specific page mapping block. The processing module maps the second specific item by the page page according to the page mapping data and locates a page according to the second item \ table. Field storage early meta-location data The above method of the present invention can be exhausted through the program. When the code is carried by the person H, it is included in the device of the physical media invention. The above-mentioned objects, features, and specific embodiments of the present invention are set forth in the accompanying drawings, and the accompanying drawings are more clearly understood, and the description is as follows. Figure 1 shows a management system in accordance with an embodiment of the present invention.

VlC07-0058I00/0608-A41455-TW/Diaft-Fmal 200937445 依據本發明實施例之映 元110、-記憶體U0與一處理S系# 100包括一館存單 以是-非揮發性記憶體4處里模組13G。儲存單元110可 】1 〇令的資料可以以、NA ND快閃記憶體。儲存單元 t與,單位進行錯存。第2丄::,以 邏輯示意圖。其中,儲在 圃”肩丁儲存早7L 11〇之VlC07-0058I00/0608-A41455-TW/Diaft-Fmal 200937445 The mirror 110, the memory U0 and the processing S system #100 according to an embodiment of the present invention include a library deposit list in a non-volatile memory 4 Module 13G. The storage unit 110 can be used as the data of the NA ND flash memory. The storage unit t and the unit are stored in error. Section 2::, in a schematic diagram. Among them, the storage in the 圃" shoulders stored early 7L 11〇

與-頁面映射區域! :1:0包括-區塊映射區域1U ❹ Ο 祕117比丄 巧2 Q塊映射區域111盥頁而 輯區塊所組成。當區塊映射區域 „區塊,即成為擁有頁面映==映射 區塊。當頁面映射區域112中的區塊滿 面映射 切換等程序重新成為區塊映 映射區塊。舉例來說,#經過合 之- 區域112中-區塊的每-頁面都依序儲存資料::映射 塊將會變成區塊映射區域ηι中之區 才則此區 :存J元中依據頁面映射模式儲存資料的區的:二 :,頁面映射區域112的大小係依據記憶㈣的大f: 決定。其+ ’記憶體12G越大,則f面 小所 2大。值得說明的是,在-實施例中,頁面映射區二 的可以達到-絲度,且幾乎可以全將主機中 私序,如Wmdows,Linux等作業系統經常存取 =全部涵蓋。如此一來’熱資料便可—直以頁面映:: 式健存’從而避免了因為反復寫入熱資料而引起之人 併。另外,當比較少存取之資料(冷:#料)寫人後,由於^目 VIC07-〇〇58I〇〇/〇6〇8-A41455-TW/Draft-Final 200937445 應重寫的次數較少,亦可以很快地經過人 οχ Λ, ^ 1 1 1 〇讦重新回到區塊 映射&域111巾。注意的是,儲存單元11G亦 映 射目錄、複數區塊映射表(未顯示)、複數頁面夫 ,、以及複數頁面映_未顯示),其相關格 膂形將於後說明。處理模組130可以由上And - page mapping area! :1:0 includes -block mapping area 1U ❹ 秘 117 117 117 2 2 Q block mapping area 111 盥 page and the block is composed. When the block mapping area „block, that is, the owning page mapping== mapping block. When the block mapping area in the page mapping area 112 is switched, the program becomes the block mapping mapping block again. For example, #合合- In area 112 - each page of the block stores the data in sequence: the mapping block will become the area in the block mapping area ηι, then this area: the area in the J element that stores the data according to the page mapping mode : 2: The size of the page mapping area 112 is determined according to the large f: of the memory (4). The + 'the larger the memory 12G is, the smaller the f surface is. 2 It is worth noting that, in the embodiment, the page mapping Zone 2 can reach - silk, and almost all of the host's private order, such as Wmdows, Linux and other operating systems often access = all covered. So the 'hot data can be - directly to the page:: Save 'and thus avoid people who are caused by repeated writing of hot data. In addition, when less access to the data (cold: #料) writes people, because ^^ VIC07-〇〇58I〇〇/〇6〇 8-A41455-TW/Draft-Final 200937445 should be rewritten less often, or Soon after the person οχ Λ, ^ 1 1 1 〇讦 back to the block mapping & field 111 towel. Note that the storage unit 11G also maps the directory, the complex block mapping table (not shown), the plural page , and the plural page image _ not shown), the relevant grid shape will be described later. The processing module 130 can be

等接收讀取命令’將讀取命令分解為以頁‘為單 -子讀取命令。處理模組130依據區塊映射表早 區塊表、與頁φ映射表將子命令轉換為相應儲存單元⑽ ==且執行讀取命令,以由儲存單-元讀取資 舆頁面映 映射目錄、區塊映射表、頁面映射區塊表 射表之内容將說明於後。 映射目錄係映射使用者資料的根目錄。映射目錄包含 多個項目,其中—映射目錄編號可以作為映射目錄中個別 項目之索引,且每一項目可以指向相應一既定大小,如 256MB使用者資料之包含一部分區塊映射表的一頁面。第 3圖顯示依據本發明實施例之映射目錄項目的格式。如圖 所示,映射目錄項目300包括一,bInCache,攔位31〇、 一 ’CachedBmtNumber,攔位 320、與一,BmtPageL〇cati〇n, 攔位330。’blnCache’攔位310係記錄相應之區塊映射表頁 面是否在記憶體120中。舉例來說,若區塊映射表頁面 經在記憶體120中,則’blnCache,欄位310中記錄,1,.— 區塊映射表頁面不在記憶體120中,則,blnCache,攔位31〇 中記錄’0’。對於已經在記憶體12〇中的區塊映射表頁 VIC07-0O58I0O/O608-A41455-TW/Draft-Final 200937445 面’ ’CachedBmtNumber’欄位320記錄相應此區塊映射表 頁面在記憶體120中的位址。對於不在記憶體12〇中的區 塊映射表頁面’則’CachedBmtNumber,攔位 320 為’Invalid’。’BmtPageLocation’攔位330係記錄區塊映射表 頁面在儲存單元110上的實體儲存位址。 區塊映射表包含多個項目’其中一邏輯區塊編號可以 作為區塊映射表中個別項目之索引,且每一項目記錄相應 此邏輯區塊之資料的存儲位置資訊。第4圖顯示依據本發 ® 明貫施例之區塊映射表項目的格式。如圖所示,區塊映射 表項目 400 包括一’bPageMapped,攔位 410、與一,Blocklnfo, 攔位420。如前所述,儲存單元no中的資料可以是以區 塊映射模式或頁面映射模式進行儲存。因 此,’bPageMapped’攔位410係用以記錄一邏輯區塊所採用 的映射模式。舉例來說。若區塊是採用區塊映射模式, 貝1j’bPageMapped’攔位410中記錄,〇,;若區塊是採用頁面映 射模式,則’bPageMapped’欄位 410 中記錄,Γ。,m〇ckInf〇, ©攔位420則係記錄詳細之映射資訊。其中,對於採用區塊 映射模式之邏輯區塊,’Blocklnfo,攔位420中纪錄相廐眘 料區塊的實體位置,如實體區塊編號。對於採用頁面映射 模式之邏輯區塊,’Blocklnfo,攔位42〇中記錄一頁面映射 區塊編號,用以表示此邏輯區塊為第幾個頁面映射區塊。 頁面映射區塊表包含多個項目,其中頁面映射區塊編 號可以作為頁面映射區塊表中個別項目之索引,且每一項 目記錄相應此賴區塊之頁面映射表在記憶體⑽中的起 VIC07-0058IOO/0608-A41455-TW/Draft-Final 12 200937445 始位址。第5圖顯示依據本發明實施例之頁面映射區塊表 項目的格式。如圖所示,頁面映射區塊表項目500包括 —’PageMappedBlockNum’ 攔 位 510 、 與 一 ’PageMappingTablelnfo’ 攔 位 520 。 其 中’ ’PageMappedBlockNum’欄位510係用以記錄頁面映射 區塊編號’且’ PageMappingTablelnfo’攔位520係用以記錄 相應此邏輯區塊之頁面映射表在記憶體120中的起始位 址0 ❹ 頁面映射表包含多個項目,其中一頁面位移(Offset)可 以作為頁面映射表中個別項目之索引,且每一項目記錄相 應此邏輯區塊中邏輯頁面的實體儲存位置,如裝置編號、 實體區塊編號與實體頁面位移等。第6圖顯示依據本發明 實施例之頁面映射表項目的格式。如圖所示,頁面映射表 項目 600 包括一’PageOffset’攔位 610、與一,pageInf〇,欄位 620。其中’ ’PageOffset’欄位610係用以記錄頁面位移, 且Pagelnfo搁位620係用以§己錄相應此邏輯區塊中邏輯 ❹頁面的實體儲存位置。注意的是,在一些實施例中,頁面 映射表中的項目數目等於一個邏輯區塊中所包含的頁面數 量。 、 第7圖顯示依據本發明實施例之映射管理方法 首先,如步驟S702,由上層, 如應用程式等接收一讀 取命令,且如步驟S704, 頁面為單位之至少一第一 可以包括一映射目錄資料 將接收之魏命令分解為以邏輯 讀取命令。其中,第—讀取命令 、一區塊映射資料、—頁面映射 VICO7-0058IO0/0608-A41455-TW/Draft-Final 13 200937445 資料、一裳置資料、一、 —)〇 , B ^ 如圖所示,讀取命令δ 月貝施j之哨取命令的格式。 一’PO’攔位、,DN’襴位、t -,阶攔位、 位記錄映射目錄資料,、 欄。其中,,MDN,欄 ❹ ❾ 攔位記錄頁面映射資〇’攔位記錄區塊映射資料,,P0, 攔位記錄磁區資料:二•攔位記錄裝置資料 ,且’SO’ 映射目錄資料(映射目S7G6,依據第〜讀取命令中之 射表中定位-特定區塊映由映射目錄中之複數區塊映 區塊映射表是否已經載g °如步驟S7Q8 ’判斷此特定 定區塊映射表是否已;記憶體中。提醒的是,判斷特 中相應映射目錄資料之人於記憶體中可以依據映射目錄 斷。若特定區塊映射表X ★目^中’131110^116’攔位的資料來判 步驟S7U),由已經載 2體中(步驟S708的否),如 選區塊映射表’且如步驟^體中之區塊映射表中選擇-候 憶體中以取代候選區料,將特定區塊映射表載入記 施例中,當記憶體空間=的是,在-些實 映射表之操作,且候選區塊二=:選擇候選區塊 直接,區塊映射表载入:=仍具有空間,則可以 +特疋區塊映射表已經在記憶體中(步驟s观的是), 口 ’驟S714 ’依據第一讀取命令中之區塊映射資料(邏輯 區„)由特定區塊映射表中定位一第一特定項目,其中 第一特定項目包括一映射模式設定(,bPageMapped,欄位之The receive read command 'decomposes the read command into a single-sub-read command with page '. The processing module 130 converts the subcommand into a corresponding storage unit (10) == according to the block mapping table early block table and the page φ mapping table, and executes a read command to read the resource page mapping directory from the storage unit. The contents of the block mapping table and the page mapping block table will be described later. The mapping directory maps the root directory of user data. The mapping directory contains multiple items, where the mapping directory number can be used as an index for individual items in the mapping directory, and each item can point to a corresponding size, such as a page of a 256MB user profile that contains a portion of the block mapping table. Figure 3 shows the format of a mapping directory entry in accordance with an embodiment of the present invention. As shown in the figure, the mapping directory item 300 includes a bInCache, a block 31〇, a 'CachedBmtNumber, a block 320, and a BmtPageL〇cati〇n, a block 330. The 'blnCache' block 310 records whether the corresponding block map page is in the memory 120. For example, if the block mapping table page is in the memory 120, then 'blnCache, the record in the field 310, 1, the. block mapping table page is not in the memory 120, then the blnCache, the block 31〇 Record '0' in the middle. For the block mapping table page VIC07-0O58I0O/O608-A41455-TW/Draft-Final 200937445 "CachedBmtNumber" field 320 already in the memory 12〇, the corresponding block mapping table page is recorded in the memory 120. Address. For the block mapping table page 'not 'CachedBmtNumber' in the memory 12〇, the block 320 is 'Invalid'. The 'BmtPageLocation' block 330 is the physical storage address of the record block mapping table page on the storage unit 110. The block mapping table contains a plurality of items'. One of the logical block numbers can be indexed as an individual item in the block mapping table, and each item records the storage location information of the data corresponding to the logical block. Figure 4 shows the format of the block mapping table entry in accordance with this example. As shown, block mapping table item 400 includes a 'bPageMapped, block 410, and one, Blocklnfo, block 420. As previously mentioned, the data in the storage unit no can be stored in a block mapping mode or a page mapping mode. Therefore, the 'bPageMapped' block 410 is used to record the mapping mode employed by a logical block. for example. If the block is in the block mapping mode, the record is recorded in the block 1j'bPageMapped' block 410, and if the block is in the page mapping mode, the 'bPageMapped' field 410 is recorded, Γ. , m〇ckInf〇, ©Block 420 is to record detailed mapping information. Wherein, for the logical block using the block mapping mode, 'Blocklnfo, block 420 records the physical location of the block of caution, such as the physical block number. For the logical block using the page mapping mode, 'Blocklnfo, a page mapping block number is recorded in the block 42〇 to indicate that the logical block is the first page mapping block. The page mapping block table includes a plurality of items, wherein the page mapping block number can be used as an index of individual items in the page mapping block table, and each item record corresponds to the page mapping table of the blocking block in the memory (10) VIC07-0058IOO/0608-A41455-TW/Draft-Final 12 200937445 Starting address. Figure 5 shows the format of a page mapping block table entry in accordance with an embodiment of the present invention. As shown, page mapping block table entry 500 includes - 'PageMappedBlockNum' block 510 and a 'PageMappingTableInfo' stop 520. The 'PageMappedBlockNum' field 510 is used to record the page mapping block number 'and' PageMappingTableInfo' 520 is used to record the starting address of the page mapping table corresponding to the logical block in the memory 120. The page mapping table includes a plurality of items, wherein a page offset (Offset) can be used as an index of individual items in the page mapping table, and each item records a physical storage location corresponding to the logical page in the logical block, such as a device number, a physical area. Block number and physical page displacement, etc. Figure 6 shows the format of a page map entry in accordance with an embodiment of the present invention. As shown, page mapping table item 600 includes a 'PageOffset' block 610, and a page inf, field 620. The ' ‘PageOffset' field 610 is used to record the page displacement, and the Pagelnfo 620 is used to record the physical storage location of the logical ❹ page in the corresponding logical block. Note that in some embodiments, the number of items in the page map is equal to the number of pages contained in one logical block. FIG. 7 shows a mapping management method according to an embodiment of the present invention. First, in step S702, a read command is received by an upper layer, such as an application, and, in step S704, at least one of the first units of the page may include a mapping. The catalog material decomposes the received Wei command into a logical read command. Among them, the first - read command, a block mapping data, - page mapping VICO7-0058IO0 / 0608-A41455-TW / Draft-Final 13 200937445 data, a skirt data, a, -) 〇, B ^ Show, read the command δ yubei Shi j's whistle fetch command format. A 'PO' block, DN' bit, t-, block, bit record mapping directory data, column. Among them, MDN, column ❹ 拦 block record page mapping information 'block record block mapping data, P0, block record record area data: two · block record device data, and 'SO' map directory data ( Mapping the target S7G6, according to the positioning in the shot table in the first read command - the specific block mapping in the mapping directory, the complex block mapping block mapping table has been loaded g ° as step S7Q8 'determine this specific fixed block mapping Whether the table is already in the memory; the reminder is that the person who judges the corresponding mapping directory data in the memory can be broken according to the mapping directory in the memory. If the specific block mapping table X ★ ^ ^ ^131110^116' blocked The data is judged to be in the step S7U), and the candidate block material is replaced by the candidate block in the block type mapping table in the block body, and the candidate block material is replaced in the block mapping table in the step body. The specific block mapping table is loaded into the recording example, when the memory space = is, in the operation of the real mapping table, and the candidate block 2 =: selecting the candidate block directly, the block mapping table is loaded: = still have space, then you can + special block mapping table already In the memory (step s is), the port 'S714' locates a first specific item from the specific block mapping table according to the block mapping data (logical area „) in the first read command, where the first Specific items include a mapping mode setting (, bPageMapped, field

VIC07-0〇58I〇〇/〇6〇8-A41455-TW/Draft-FinaI 200937445 資料)與一區塊資訊(,BlockInf〇,攔位之次' S716’依據映射模式設定判斷相應區=料)/如步驟 頁面映射模^。若映射模式純塊 :射模式是否為VIC07-0〇58I〇〇/〇6〇8-A41455-TW/Draft-FinaI 200937445 data) and a block information (, BlockInf〇, the second block of the 'S716' according to the mapping mode setting to determine the corresponding area = material) / As step page mapping mode ^. If the mapping mode is pure block: Is the shot mode

的否),如步驟S718,判斷第一特定項、式時(步驟S716 資訊是否為無效。在—些實施例中,2了所錢的區塊 無效係判斷區塊資訊S否為-特定值,如區塊資訊是否為 無效(步驟S718的是),如步驟S72〇,直·。1。若區塊資訊 如〇。若區塊資訊並非無效(步驟S718回應—既定值, 依據第-特定項目中之區塊資訊由儲存^ ^步驟S722’ 塊,且依據第一讀取命令與區塊資:兀疋位―資料區 第二讀取命令。之後,如步驟以生相應:儲存單元之 以由儲存單元讀取此資料區塊。 行第一%取命令, 若映射模式為頁面映射模式時 驟S724,依據第_特定項目中之=驟次奶6的是),如步 編號)由一頁面映射區塊表中定位一貝訊(頁面映射區塊 如步驟S726,透過特定頁面映射區^頁面映—射區塊,且 表。如步驟S728,依據第一讀取命八 頁面映射 面位移)由頁面映射表中定位—第^頁面映射資料(頁 特定項目記錄此邏輯區塊中邏輯頁面:u二第二 :驟_ ’依據第二特定項目由儲存單元定 存單元之第三讀取命令。之後 驟目貝,生相應儲 取命令,以由鍅左--y 驟S732 ’執行第三讀 取h心儲存早㈣取此資料心 如前所述,第一讀取命令更可 :::思的疋, 匕括裝置資料與/或磁區資 VIC07-0058I00/0608-A41455-TW/Draft-Final 15 200937445 料。當第一讀取命令包括裝置 以特定項目與裝置資料與;二區資資 兀疋位出資料頁面。 艰匕#料由儲存早 有鑑於此,透過本案之映射 區塊映射模式與頁面映射模去及糸統可以結合 情形下,維持系統的讀寫性在‘·,、須掃描特定表格的 e ❾ =可=载電=儲,,其*二 “人且執η此機器變成用以參與 =月之裝置。本發明之方法與裝置也可以以程式碼型態 透,-些傳送媒體,如電、钱電繞、錢、或是任何傳輸 里心進行傳送’其中’當程式碼被機器,如電腦接收、載 入且執行時,此機器變成用以參與本發明之裝置。各在一 般用途處理器實作時,程式碼結合處理器提供一操;類似 於應用特定邏輯電路之獨特裝置。 雖然本發明已以較佳實施例揭露如上,然其並非用以 限定本發明,任何熟悉此項技藝者,在不脫離本發明之精 神和範圍内,當可做些許更動與潤飾,因此本發明之保護 範圍當視後附之申請專利範圍所界定者為準。 【圖式簡單說明】 第1圖為一示意圖係顯示依據本發明實施例之映射技 理系統。 第2圖為一示意圖係顯示依據本發明實施例之儲存單 VIC07-0058!00/0608-A41455-TW/Drafl-Final 16 200937445 元。 第3圖為一示意圖係顯示依據本發明實施例之映射目 錄項目的格式。 第4圖為一示意圖係顯示依據本發明實施例之區塊映 射表項目的格式。 第5圖為一示意圖係顯示依據本發明實施例之頁面映 射區塊表項目的格式。 第6圖為一示意圖係顯示依據本發明實施例之頁面映 〇 射表項目的格式。 第7圖為一流程圖係顯示依據本發明實施例之映射管 理方法。 第8圖為一示意圖係顯示依據本發明實施例之讀取命 令的格式。 【主要元件符號說明】 100〜映射管理系統; 110〜儲存單元; ❹ 111〜區塊映射區域; 112~頁面映射區域; 120〜記憶體; 130〜處理模組; 300〜映射目錄項目; 310〜’bInCachejWi; 320〜’CachedBmtNumber’欄位; 330〜’BmtPageLocation’欄位; VIC07-0058IOO/0608-A4145 5-TW/Draft-Final 17 200937445 40 0〜區塊映射表項目, 420〜’Blocklnfo’攔位; 5 00〜頁面映射區塊表項目, 510〜’卩&861^[&卩卩6(161〇〇1<;>111111’欄位; 520〜’PageMappingTablelnfo’欄位; 600〜頁面映射表項目; 610 〜’PageOffset’欄位; ❹ 620〜’Pagelnfo’欄位; S702、S704、…、S732〜步驟; 800〜讀取命令; MDN〜映射目錄資料棚位, Β Ο〜區塊映射資料棚位, PO〜頁面映射貢料棚位, DN〜裝置資料攔位; SO〜磁區貧料搁位。 ❹ YICO7-O058IOO/O6O8-A41455-TW/Draft-FinalIf no, in step S718, the first specific item is determined (the information in step S716 is invalid. In some embodiments, the block invalidity of the money is judged whether the block information S is - a specific value If the block information is invalid (YES in step S718), as in step S72〇, straight·1. If the block information is as follows. If the block information is not invalid (step S718 response-established value, according to the first-specific The block information in the project is stored in the ^^ step S722' block, and according to the first read command and the block resource: the position - the second read command of the data area. After that, the steps are as follows: the storage unit The data block is read by the storage unit. The first % fetch command is executed, if the mapping mode is the page mapping mode, step S724, according to the _specific item = the sudden milk 6 is), such as the step number) Positioning a page in the page mapping block table (the page mapping block is mapped to the block by the specific page mapping area, and the table is mapped through the specific page mapping area, and the table is displayed according to the first reading order page. Displacement) is located in the page mapping table - the second page mapping (The page specific item records the logical page in this logical block: u 2nd: _ 'The third read command is stored by the storage unit according to the second specific item. After that, the corresponding storage command is generated. According to the left--y step S732 'perform the third reading h heart storage early (four) take this information heart as described above, the first read command can be::: thinking, including device information and / or Magnetic zone resources VIC07-0058I00/0608-A41455-TW/Draft-Final 15 200937445. When the first read command includes the device with specific items and device data; the second district capitalizes the information page. Since the storage block mapping mode and the page mapping mode and the system can be combined, the read/write performance of the system can be maintained in the case of '·, and the specific table must be scanned. The power supply=storage, the second and the second, and the device becomes a device for participating in the month. The method and device of the present invention can also be transmitted in a coded form, and some of the transmission media, such as electricity and money. Transfer, money, or any transmission to the heart of the 'where' code When a machine, such as a computer receives, loads, and executes, the machine becomes a device for participating in the present invention. Each of the programs is implemented in conjunction with the processor when the general purpose processor is implemented; similar to the uniqueness of the application specific logic circuit The present invention has been described above by way of a preferred embodiment, and is not intended to limit the invention, and any modifications and refinements may be made without departing from the spirit and scope of the invention. The scope of the present invention is defined by the scope of the appended claims. [FIG. 1] FIG. 1 is a schematic diagram showing a mapping technique system according to an embodiment of the present invention. FIG. 2 is a schematic diagram. A memory list VIC07-0058!00/0608-A41455-TW/Drafl-Final 16 200937445 according to an embodiment of the present invention is shown. Figure 3 is a schematic diagram showing the format of a mapping directory entry in accordance with an embodiment of the present invention. Figure 4 is a schematic diagram showing the format of a block map entry in accordance with an embodiment of the present invention. Figure 5 is a diagram showing the format of a page mapping block table entry in accordance with an embodiment of the present invention. Figure 6 is a diagram showing the format of a page mapping item in accordance with an embodiment of the present invention. Figure 7 is a flow chart showing a mapping management method in accordance with an embodiment of the present invention. Figure 8 is a diagram showing the format of a read command in accordance with an embodiment of the present invention. [Main component symbol description] 100~ mapping management system; 110~ storage unit; ❹ 111~block mapping area; 112~page mapping area; 120~memory; 130~ processing module; 300~ mapping directory item; 'bInCachejWi; 320~'CachedBmtNumber' field; 330~'BmtPageLocation' field; VIC07-0058IOO/0608-A4145 5-TW/Draft-Final 17 200937445 40 0~block mapping table item, 420~'Blocklnfo' Bit; 5 00~ page mapping block table item, 510~'卩&861^[&卩卩6(161〇〇1<;>111111' field; 520~'PageMappingTablelnfo' field; 600~ Page mapping table item; 610 ~ 'PageOffset' field; ❹ 620~ 'Pagelnfo' field; S702, S704, ..., S732~ steps; 800~ read command; MDN~ mapping directory data shed, Β Ο~ area Block mapping data shed, PO ~ page mapping tribute shed, DN ~ device data block; SO ~ magnetic zone lean shelf. ❹ YICO7-O058IOO/O6O8-A41455-TW/Draft-Final

Claims (1)

200937445 十、申請專利範菌: 二種映射管理方法,包 取传至少-第-讀取命令,其=驟: 二括-映射目錄資料、一區塊映射=第:讀取命令至少 枓’ 町貝科、與一頁面映射資 依據該映射目錄資料 區塊映射表,· 、品塊映射表中定位一特定 ❹ 依據該區塊映射資料由該 一特定項目,其中該第—特定二表中定位-第 一區塊資訊; 、匕括—映射模式設定與 富該映射模式今 資訊由一苜; 、'、頁面映射模式時,依撼兮厂认 貝訊由頁面映射區塊表〗依據該區塊 透過該特定頁面映射區塊映射區塊,且 依據該頁面映射資料由該頁面映射=位以: 頁面。*该第一特疋項目由一儲存單元定位-資: ❹ 2」如申請專利範圍第丄項所述之映射管 田摘射模式設定為—區塊映射模柄,们更包 塊資訊由該健存單元定位一資料區塊。 、據该區 3·如”專利第2項所述之映射管 當該映射模式設定為該區塊映射模式時, ^其中 列步驟: Λ 、更包括下 判斷該區塊資訊是否為一特定值;以及 若該區塊資訊為該特定值,則直接回應— i-B* 〇 VIC07-0058I00/0608-A41455-TW/Draft-Final 19 200937445 4. 如申請專利範圍第3項所述之映射管理方法,其中 該特定值為-1,且該既定值為0。 5. 如申請專利範圍第1項所述之映射管理方法,更包 括下列步驟: 依據該映射目錄資料判斷該特定區塊映射表是否已經 載入於一記憶體中;以及 若該特定區塊映射表並未載入於該記憶體中,將該特 定區塊映射表載入該記憶體中。 ❿ 6.如申請專利範圍第5項所述之映射管理方法,更包 括下列步驟: 由已經載入該記憶體中之該等區塊映射表中選擇一候 選區塊映射表;以及 將該特定區塊映射表載入該記憶體中以取代該候選區 塊映射表。 7. 如申請專利範圍第6項所述之映射管理方法,其中 該候選區塊映射表係依據一最久未使用(LRU)演算法所選 ❹擇。 8. 如申請專利範圍第1項所述之映射管理方法,更包 括下列步驟: 依據該第二特定項目將該第一讀取命令轉換為相應該 儲存單元之至少一第二讀取命令;以及 執行該第二讀取命令,以由該儲存單元讀取該資料頁 面。 9. 如申請專利範圍第1項所述之映射管理方法,其中 VIC07-0058I00/0608-A41455-TW/Draft-Final 20 200937445 :第-讀取命令更包括 去更依據該第二特定項目 ^或一磁區資料,且該方 鍺存單元定位該資料頁面:、〜置資料或該磁區資料由該 ,二0’!°’請專利範圍第】項所过〜 "儲存早元包括一非:之映射管理方法 U.如申請專利C體。 ❹ :::存單元中依據該頁面映射之映射管理方法,其中 依據該記憶體的大小决定。、.果式錯存資科的區塊數量係 12.-種映射管理系統 . 一儲存單元,依據—百 . ❹ 分別以广頁面與至少與-區塊映射模式 處理模組,取得至少c料;以及 二Tv至少包括—映射目錄’取命7 ’其中該第一 二頁面映射資料,依據該映射目區塊映射資科、與 尹定位-特定區塊映射表=料由複數區魏映射表 區塊映射表t定位_第_^5亥區塊映射資料由該特定 包括-映射模式設定與—區=,其中該第-特定項目 該頁面映射模式時,依據該區塊該映射模式設定為 _疋位一特定頁面映射區塊 二頁面映射區魂表 定位至一頁面映射表,依據=該f定頁面映射區塊 表中定位一第二特定項目,以贫、射資料由該頁面映射 存單元定位一資料頁面。 、该第二特定項目由該儲 13.如申請專利範圍第12項所述之 "該映射模式設定為該區塊映射模式時,。::組; VIC07-0〇58!0〇/〇608-A41455-TW/Drafl-Final 21 200937445 直接區塊資訊由該錯存單 14.如申請專利範圍第13 =身枓區塊。 尹當該映射模式設定為該射管理系統,其 判斷該區塊資訊是否為―,映射叙式時,該處理模組更 定值十’直接回應—既定值^值’且當該區塊資訊為該特 15. 如申請專利範園第 令該特定值為心且該既定值為〇/射管理系統,其 16. 如申請專利範園 t該處理模叙更依據該 、次=之映射管理系統,其 表是否已經戴入於1_中==卿定區塊映射 ΐ載入於該記崎’將該特定區_射:載::: 17.如申請專利範衝 t該處理模組更由已經戴入該記理系統,其 中選擇-候選區塊映射表,且將 該相塊映射表 記憶體中以取代該候選區塊映射表"。、塊映射表載入該 18.如申請專利範圍 中該候選區塊映射表係'戶 映射管理系統’其 選擇。 蘇Μ未使用(LRU)演算法所 中該處理模組更依據該第二特定項 糸統,其 換為相應該儲存單元之至 子該苐一讀取命令轉 二讀取命令,以由該儲存單元讀:該;:J面且執行該第 20·如申請專利範圍第12項所述之映Γ管理系統,其 YIC07-0058I00/0608-A41455^TW/Draft-Final 22 200937445 中該第一讀取命令更包括一裝置資料或一磁區資料,且該 處理模組更依據該第二特定項目與該裝置資料或該磁區資 料由該儲存單元定位該資料頁面。 21. 如申請專利範圍第12項所述之映射管理系統,其 中該儲存單元包括一非揮發性記憶體。 22. 如申請專利範圍第12項所述之映射管理系統,其 中該儲存早元中依據該頁面映射模式儲存資料的區塊數罝 係依據該記憶體的大小決定。 ❹ VIC07-0058IOO/0608-A41455-TW/Drafl-Final 23200937445 X. Applying for patents: Two kinds of mapping management methods, including the at least-first-read command, which includes: step: two-mapped directory data, one block mapping = first: read command at least 枓' machi Beco, and a page mapping resource are located according to the mapping directory data block mapping table, and a specific parameter is located in the bar mapping table. According to the block mapping data, the specific item is located in the first specific table. - First block information; 匕 — 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射 映射The block maps the block by the specific page mapping block, and according to the page mapping data, the page maps the bits to: page. * The first special item is located by a storage unit - :: 」 2", as described in the scope of the patent application, the mapping field is set to the block mapping module, and the block information is The storage unit locates a data block. According to the area 3, such as the mapping tube described in the second item of the patent, when the mapping mode is set to the block mapping mode, ^ the column step: 、, further includes determining whether the block information is a specific value. And if the block information is the specific value, then directly respond - iB* 〇VIC07-0058I00/0608-A41455-TW/Draft-Final 19 200937445 4. The mapping management method as described in claim 3, The specific value is -1, and the predetermined value is 0. 5. The mapping management method according to claim 1, further comprising the following steps: determining, according to the mapping directory data, whether the specific block mapping table has been Loading in a memory; and if the specific block mapping table is not loaded in the memory, loading the specific block mapping table into the memory. ❿ 6. As claimed in the fifth item The mapping management method further includes the steps of: selecting a candidate block mapping table from the block mapping tables already loaded into the memory; and loading the specific block mapping table into the memory To replace this The block mapping table is as follows: 7. The mapping management method according to claim 6, wherein the candidate block mapping table is selected according to a longest unused (LRU) algorithm. The mapping management method of item 1, further comprising the steps of: converting the first read command to at least one second read command corresponding to the storage unit according to the second specific item; and performing the second read a command to read the material page by the storage unit. 9. The mapping management method according to claim 1, wherein VIC07-0058I00/0608-A41455-TW/Draft-Final 20 200937445: read-read The command further includes: according to the second specific item ^ or a magnetic area data, and the party storage unit locates the data page:, ~ set data or the magnetic area data by the, 2 0 '! ° 'Please patent scope The first item has passed the "storage early element including a non-: mapping management method U. For example, the patent C body. ❹::: The storage unit according to the map mapping management method, according to the size of the memory Decision The number of blocks in the case of the fruit type is stored in the 12.-type mapping management system. A storage unit, based on - hundred. 处理 The module is processed in a wide page and at least with the block mapping mode, at least c material; and the second Tv at least includes - mapping directory 'received 7', wherein the first two pages of mapping data, according to the mapping of the destination block mapping of the subject, and Yin positioning - specific block mapping table = material by the complex area Wei The mapping table block mapping table t positioning_the _^5 hex block mapping data is set by the specific including-mapping mode and the -region=, wherein the first-specific item is in the page mapping mode, according to the mapping mode of the block Set to _疋 a specific page mapping block 2 page mapping area soul table to a page mapping table, according to = the f page mapping block table to locate a second specific item, to poor, shoot data from the page The mapping unit locates a data page. The second specific item is stored by the store 13. If the mapping mode is set to the block mapping mode as described in claim 12 of the patent application scope. ::Group; VIC07-0〇58!0〇/〇608-A41455-TW/Drafl-Final 21 200937445 Direct block information from the wrong deposit order 14. If the patent application scope is 13th = body block. Yin Dang, the mapping mode is set to the shooting management system, which determines whether the block information is “,” and when the mapping is described, the processing module is more fixed to ten′′ direct response—established value ^ value' and when the block information For the special 15. If the specific value of the application for the patent garden is the heart and the established value is the 〇/射管理系统, 16. If the patent application is applied, the processing model is based on the mapping management of the second and the second. System, whether its table has been worn in 1_ ===定定块块 mappingΐLoaded in the memory of the specific area _射:::: 17. If applying for patent Fan Chong t the processing module Further, the candidate system is already inserted into the recording system, wherein the candidate block mapping table is selected, and the phase block is mapped in the table memory to replace the candidate block mapping table. The block mapping table loads the 18. The candidate block mapping table is selected by the 'home mapping management system' as in the patent application. The processing module of the Susie Unused (LRU) algorithm is further based on the second specific item system, and is replaced by the corresponding one of the storage unit to the second read command to the second read command. The storage unit reads: the; J face and executes the 20th. The mapping management system described in claim 12 of the patent application, the first of which is YIC07-0058I00/0608-A41455^TW/Draft-Final 22 200937445 The read command further includes a device data or a magnetic zone data, and the processing module further locates the data page by the storage unit according to the second specific item and the device data or the magnetic zone data. 21. The mapping management system of claim 12, wherein the storage unit comprises a non-volatile memory. 22. The mapping management system of claim 12, wherein the number of blocks in the storage element stored in the page mapping mode according to the page mapping mode is determined according to the size of the memory. ❹ VIC07-0058IOO/0608-A41455-TW/Drafl-Final 23
TW97107097A 2008-02-29 2008-02-29 Mapping management methods and systems TWI375229B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW97107097A TWI375229B (en) 2008-02-29 2008-02-29 Mapping management methods and systems

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW97107097A TWI375229B (en) 2008-02-29 2008-02-29 Mapping management methods and systems

Publications (2)

Publication Number Publication Date
TW200937445A true TW200937445A (en) 2009-09-01
TWI375229B TWI375229B (en) 2012-10-21

Family

ID=44867067

Family Applications (1)

Application Number Title Priority Date Filing Date
TW97107097A TWI375229B (en) 2008-02-29 2008-02-29 Mapping management methods and systems

Country Status (1)

Country Link
TW (1) TWI375229B (en)

Also Published As

Publication number Publication date
TWI375229B (en) 2012-10-21

Similar Documents

Publication Publication Date Title
CN104978281B (en) Data integrity management in data storage device
KR100526190B1 (en) Remapping method for flash memory
CN107168885A (en) The control method of storage system, information processing system and nonvolatile memory
CN107168884A (en) The control method of storage system, information processing system and nonvolatile memory
CN107168639A (en) The control method of storage system, information processing system and nonvolatile memory
TWI645404B (en) Data storage device and control method for non-volatile memory
CN107168640A (en) The control method of storage system, information processing system and nonvolatile memory
US20070016719A1 (en) Memory device including nonvolatile memory and memory controller
US20080162793A1 (en) Management method for reducing utilization rate of random access memory (ram) used in flash memory
CN101187898A (en) Apparatus and method of managing nonvolatile memory
CN109766341A (en) A kind of method, apparatus that establishing Hash mapping, storage medium
KR20100017094A (en) A method for efficient storage of metadata in flash memory
US20110153979A1 (en) Modified b+ tree to store nand memory indirection maps
TW200907995A (en) Method and system of defect management for storage medium
CN109407985B (en) Data management method and related device
CN108287665A (en) Flash memory device and access control method
KR100924021B1 (en) Flash memory mapping management method
JP2018160060A (en) Memory system
CN106547472A (en) Storage array management method and device
CN115904255B (en) Data request method, device, equipment and storage medium
CN106020726B (en) Method, equipment and the storage device of metadata is written
CN116340205B (en) Access request processing method and cache device
JP5373382B2 (en) Memory system and computer system
CN116661690A (en) Method, device, computer equipment and storage medium for recording memory state
CN108614671A (en) Key-data access method based on NameSpace and solid storage device