TWM524503U - Electronic device - Google Patents

Electronic device Download PDF

Info

Publication number
TWM524503U
TWM524503U TW105201300U TW105201300U TWM524503U TW M524503 U TWM524503 U TW M524503U TW 105201300 U TW105201300 U TW 105201300U TW 105201300 U TW105201300 U TW 105201300U TW M524503 U TWM524503 U TW M524503U
Authority
TW
Taiwan
Prior art keywords
logical block
data
electronic device
acceleration
acceleration zone
Prior art date
Application number
TW105201300U
Other languages
Chinese (zh)
Inventor
傅子瑜
Original Assignee
宏碁股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 宏碁股份有限公司 filed Critical 宏碁股份有限公司
Priority to TW105201300U priority Critical patent/TWM524503U/en
Publication of TWM524503U publication Critical patent/TWM524503U/en

Links

Landscapes

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

Abstract

An electronic device comprising: a CPU; and a controller, for: when receiving a command of the CPU, determining whether a plurality of logical block addresses are randomly distributed and whether capacity of data which the logical block address correspond is less than a first specific value according to the command; determining that a plurality of physical addresses corresponding to the logical block addresses are in an acceleration region and accessing the data from the acceleration region when the logical block addresses are randomly distributed and the capacity of the data is less than the first specific value; and determining that the physical address are in an non-acceleration region, and accessing the data from the non-acceleration region when the logical block address are not randomly distributed or the capacity of the data is not less than the first specific value.

Description

電子裝置 Electronic device

本新型係有關於一種硬碟的存取技術,特別係有關於一種固態硬碟(SSD)之存取技術其對應之電子裝置。 The present invention relates to a hard disk access technology, and more particularly to an electronic device corresponding to a solid state hard disk (SSD) access technology.

Intel提出了革命性的3D XPoint記憶體,其存取速度遠比NAND快閃記憶體還快(甚至能超越DRAM)、且將較於NAND快閃記憶體還沒有寫入抹除次數的限制。唯一美中不足的點是3D XPoint記憶體的良率、跟單位面積可儲存的資料量目前仍無法超越NAND快閃記憶體。況且,過往的計算機架構與作業系統都是建立在揮發性的記憶體與中央處理單元做溝通,目前的架構要修改為中央處理單元與非揮發性的記憶體溝通仍有待業界各方的努力。 Intel has proposed a revolutionary 3D XPoint memory that is much faster than NAND flash memory (even beyond DRAM) and will not have the limit of write erase times compared to NAND flash memory. The only fly in the ointment is that the yield of 3D XPoint memory and the amount of data that can be stored per unit area are still beyond the NAND flash memory. Moreover, the past computer architecture and operating system are based on the communication between the volatile memory and the central processing unit. The current architecture has to be modified to communicate with the central processing unit and non-volatile memory.

本新型認為目前此種記憶體的最好使用方式是當固態硬碟的Cache來使用,固態硬碟若改用3D XPoint記憶體當作控制器的Cache。本新型甚至可以儲存常用的資料在3D XPoint記憶體內,因為NAND快閃記憶體之小資料量的存取時間仍是無法與3D XPoint記憶體相比,因此我們可以將隨機小資料放入3D XPoint記憶體,如此一來,整體固態硬碟的存取速度就能再提升七倍(根據intel目前宣稱的數據)。 The new model believes that the best way to use this type of memory is to use the Cache of the solid state hard disk. If the solid state hard disk uses 3D XPoint memory as the Cache of the controller. The new model can even store commonly used data in 3D XPoint memory, because the access time of small data volume of NAND flash memory is still not comparable with 3D XPoint memory, so we can put random small data into 3D XPoint. Memory, as a result, the overall SSD access speed can be increased by a factor of seven (according to Intel's current claims).

在本新型將3D XPoint記憶體作為固態硬碟的Cache時,那些資料要放入3D XPoint記憶體就成為一個新的問題。本新型當然可以直接效仿過往SSHD(用NAND快閃記憶體幫硬碟(HDD)加速的架構),此技術主要是由存取頻率來決定哪些資料放進加速區。然而,此種方式只能確保很頻繁使用的資料被加速,但有些資料系統可能一天只存取一次,這些資料就會無法被加速到,因為它們存取的頻率實在太低。 When the new 3D XPoint memory is used as the Cache of the solid state hard disk, the data to be put into the 3D XPoint memory becomes a new problem. This new model can of course directly emulate the previous SSHD (the architecture accelerated by NAND flash memory for hard disk (HDD)), which is mainly based on the access frequency to determine which data is placed in the acceleration zone. However, this approach only ensures that frequently used data is accelerated, but some data systems may only be accessed once a day, and these data cannot be accelerated because they are accessed too slowly.

有鑑於此,本新型在此提出一種利用新儲存記憶體3D XPoint的電子裝置,上述電子裝置具有一時間標籤,使得電子裝置可將隨機小資料放入3D XPoint新儲存記憶體中,以讓固態硬碟將近期存取過的資料保留在加速區中進行存取,進而使這些隨機小資料的存取速度提高數倍。 In view of this, the present invention proposes an electronic device using a new memory 3D XPoint, the electronic device having a time stamp, so that the electronic device can put random small data into the 3D XPoint new storage memory to make the solid state The hard disk keeps the recently accessed data in the acceleration zone for access, thereby increasing the access speed of these random small data several times.

本新型提供一種電子裝置,包括:一中央處理單元;以及一控制器,用以:當接收來自上述中央處理單元之一指令時,根據上述指令判斷複數個邏輯區塊位址是否為一隨機分布且上述邏輯區塊位址所對應之資料之容量是否小於一第一特定數值;當上述邏輯區塊位址為上述隨機分布且上述資料之容量小於上述第一特定數值,判斷為對應於上述邏輯區塊位址之複數個實體位址在一加速區中,上述資料於上述加速區中進行存取;以及當上述邏輯區塊位址不為上述隨機分布或上述資料之容量不小於上述第一特定數值,判斷為對應於上述邏輯區塊位址之上述實體位址在一非加速區中,上述資料於上述非加速區中進行存取。 The present invention provides an electronic device comprising: a central processing unit; and a controller for determining whether a plurality of logical block addresses are randomly distributed according to the instruction when receiving an instruction from the central processing unit And whether the capacity of the data corresponding to the logical block address is less than a first specific value; when the logical block address is randomly distributed and the capacity of the data is less than the first specific value, it is determined to correspond to the logic a plurality of physical addresses of the block address in an acceleration zone, wherein the data is accessed in the acceleration zone; and when the logical block address is not the random distribution or the capacity of the data is not less than the first The specific value is determined to be that the physical address corresponding to the logical block address is in a non-acceleration area, and the data is accessed in the non-acceleration area.

為讓本新型之上述和其他目的、特徵、和優點能 更明顯易懂,下文特舉出較佳實施例,並配合所附圖式,作詳細說明如下: To enable the above and other objects, features, and advantages of the present invention It is more obvious and easy to understand. The preferred embodiments are described below, and the detailed description is as follows with the accompanying drawings:

100‧‧‧電子裝置 100‧‧‧Electronic devices

102‧‧‧中央處理單元 102‧‧‧Central Processing Unit

104‧‧‧進階主機控制器介面 104‧‧‧Advanced Host Controller Interface

106‧‧‧固態硬碟 106‧‧‧ Solid State Drive

108‧‧‧控制器 108‧‧‧ Controller

110‧‧‧記憶體 110‧‧‧ memory

112‧‧‧加速區 112‧‧‧ acceleration zone

114‧‧‧非加速區 114‧‧‧Non-acceleration zone

LBA1、LBA2、LBA3、LBA4、LBA7、LBA10、LBA23、LBA37、LBA49‧‧‧邏輯區塊位址 LBA1, LBA2, LBA3, LBA4, LBA7, LBA10, LBA23, LBA37, LBA49‧‧‧ logical block addresses

Y1、Y2、Y3、Y4、Y5、Y6、Y7、Y8、Y9、Y10‧‧‧實體位址 Y1, Y2, Y3, Y4, Y5, Y6, Y7, Y8, Y9, Y10‧‧‧ physical addresses

S10、S20、S30、S40‧‧‧步驟 S10, S20, S30, S40‧‧‧ steps

第1圖顯示為本新型之一電子裝置之主要架構圖;第2圖顯示邏輯區塊位址及實體位址之一第一對應關係表;第3A圖顯示邏輯區塊位址、實體位址以及時間標籤之一第二對應關係表;第3B圖顯示邏輯區塊位址、實體位址以及時間標籤之一第二對應關係表;第3C圖顯示邏輯區塊位址、實體位址以及時間標籤之一第二對應關係表;第4圖顯示依據本新型之電子裝置之操作方法之一流程圖。 Figure 1 shows the main architecture of one of the electronic devices of the present invention; Figure 2 shows the first correspondence table of logical block addresses and physical addresses; Figure 3A shows the logical block addresses and physical addresses. And a second correspondence table of time labels; FIG. 3B shows a second correspondence table of logical block addresses, physical addresses, and time tags; FIG. 3C shows logical block addresses, physical addresses, and times One of the labels is a second correspondence table; and FIG. 4 is a flow chart showing a method of operating the electronic device according to the present invention.

以下將詳細討論本新型各種實施例之製造及使用方法。然而值得注意的是,本新型所提供之許多可行的新型概念可實施在各種特定範圍中。這些特定實施例僅用於舉例說明本新型之製造及使用方法,但非用於限定本新型之範圍。 The methods of making and using the various embodiments of the present invention are discussed in detail below. It is to be noted, however, that many of the possible novel concepts provided by the present invention can be implemented in a variety of specific ranges. These specific examples are merely illustrative of the methods of manufacture and use of the present invention, but are not intended to limit the scope of the present invention.

第1圖顯示為本新型之一電子裝置100之主要架構圖。如第1圖中所示,電子裝置100包括一中央處理單元102、一進階主機控制器介面(AHCI)104以及一固態硬碟106。固態硬碟106包括一控制器108、一記憶體110、一加速區112以及一非加速區114。中央處理單元102是電子裝置100(例如一電腦)的主要裝置之一,功能主要是解釋電腦指令以及處理電腦軟體中的 資料。中央處理單元102用以透過進階主機控制器介面104傳送一指令至控制器108。 FIG. 1 shows a main architectural diagram of an electronic device 100 of the present invention. As shown in FIG. 1, the electronic device 100 includes a central processing unit 102, an advanced host controller interface (AHCI) 104, and a solid state drive 106. The solid state drive 106 includes a controller 108, a memory 110, an acceleration zone 112, and a non-acceleration zone 114. The central processing unit 102 is one of the main devices of the electronic device 100 (for example, a computer), and the function mainly explains computer instructions and processing in the computer software. data. The central processing unit 102 is configured to transmit an instruction to the controller 108 via the advanced host controller interface 104.

進階主機控制器介面104用以規範出中央處理單元102與儲存裝置(例如:固態硬碟106)間傳輸資料或指令之方式。在本新型的一實施例中,進階主機控制器介面104具有將來自中央處理單元102之一SATA指令傳送至控制器108的功能。在本新型的另一實施例中,進階主機控制器介面104具有將來自中央處理單元102之一PCIe指令傳送至控制器108的功能。控制器108用以當接收來自中央處理單元102之指令時,判斷與指令中之複數個邏輯區塊位址所對應之複數個實體位址是否在加速區112中,其餘有關控制器108的詳細動作容後再述。記憶體110用以儲存一第一對應關係表以及一第二對應關係表。有關第一對應關係表以及第二對應關係表之詳細內容於第2圖以及第3圖所述。在本新型的某些實施例中,記憶體110亦可在加速區112或非加速區114中,本新型則不以此為限。在本新型之一實施例中,加速區112可為一3D XPoint記憶體,非加速區114可為一NAND快閃記憶體,本新型則不以此為限。 The advanced host controller interface 104 is used to regulate the manner in which data or instructions are transferred between the central processing unit 102 and a storage device (eg, solid state drive 106). In an embodiment of the present invention, the advanced host controller interface 104 has the function of transmitting SATA commands from the central processing unit 102 to the controller 108. In another embodiment of the present invention, the advanced host controller interface 104 has the function of transmitting PCIe instructions from the central processing unit 102 to the controller 108. The controller 108 is configured to, when receiving an instruction from the central processing unit 102, determine whether a plurality of physical addresses corresponding to the plurality of logical block addresses in the instruction are in the acceleration area 112, and the details of the remaining related controllers 108 The action will be described later. The memory 110 is configured to store a first correspondence table and a second correspondence table. The details of the first correspondence table and the second correspondence table are described in FIG. 2 and FIG. In some embodiments of the present invention, the memory 110 may also be in the acceleration zone 112 or the non-acceleration zone 114, and the present invention is not limited thereto. In one embodiment of the present invention, the acceleration region 112 can be a 3D XPoint memory, and the non-acceleration region 114 can be a NAND flash memory. The present invention is not limited thereto.

第2圖顯示邏輯區塊位址及實體位址之一第一對應關係表。第一對應關係表係為包括在非加速區114中邏輯區塊位址與實體位址。如第2圖所示,每一個邏輯區塊位址接對應一個實體位址。在本新型之一實施例中,邏輯區塊位址LBA1對應於實體位址Y1,邏輯區塊位址LBA10對應於實體位址Y2,邏輯區塊位址LBA23對應於實體位址Y3,邏輯區塊位址LBA37對應於實體位址Y4,邏輯區塊位址LBA49對應於實體位址Y5。 Figure 2 shows a first correspondence table of logical block addresses and physical addresses. The first correspondence table is a logical block address and a physical address included in the non-acceleration area 114. As shown in Figure 2, each logical block address is connected to a physical address. In an embodiment of the present invention, the logical block address LBA1 corresponds to the physical address Y1, the logical block address LBA10 corresponds to the physical address Y2, and the logical block address LBA23 corresponds to the physical address Y3, the logical area The block address LBA37 corresponds to the physical address Y4, and the logical block address LBA49 corresponds to the physical address Y5.

第3A圖顯示邏輯區塊位址、實體位址、時間標籤(time stamp)、本次存取時間以及上次存取時間之一第二對應關係表。在本新型之一實施例中,邏輯區塊位址LBA1對應於實體位址Y6,邏輯區塊位址LBA10對應於實體位址Y7,邏輯區塊位址LBA23對應於實體位址Y8,邏輯區塊位址LBA37對應於實體位址Y9,邏輯區塊位址LBA49對應於實體位址Y10,時間標籤具有一第二特定數值(例如:10),本次存取時間為13:05,上次存取時間為13:00。本次存取時間與上次存取時間之時間差距為一存取間隔。當存取間隔越大時,時間標籤之數值越小。在本新型之一實施例中,當資料於加速區112每被存取一次,本次存取時間與上次存取時間皆會被更新。舉例而言,當資料於13:06時於加速區112被存取,本次存取時間由13:05改為13:06,上次存取時間由13:00改為13:05,以便控制器108進行關於時間標籤之數值的運算。在本新型之一實施例中,當實體位址在加速區112中,控制器108將時間標籤之數值設定為一第二特定數值(例如:10)。在本新型之另一實施例中,如第3B圖所示,當於加速區112中進行存取之資料每存取間隔(例如:30分鐘)內沒被存取,控制器108將時間標籤之數值減一。舉例而言,上次存取時間為13:05,本次存取時間為13:35,上次存取時間與本次存取時間之時間間隔大於或等於30分鐘,控制器108將時間標籤之數值減為9。除此之外,當時間標籤之數值為零時,資料從加速區112中移至非加速區114進行存取。在本新型之另一實施例中,如第3C圖所示,當於加速區112中進行存取之資料於複數個存取間隔(例如:10*30分鐘=5小時)內沒被 存取。舉例而言,一旦本次存取時間在18:05以後(意即本次存取時間與上次存取時間之時間間隔大於或等於10*30分鐘=5小時),控制器108便將時間標籤之數值減至零。如此一來,資料便從加速區112中移至非加速區114進行存取。也就是說,邏輯區塊位址與實體位址之對應關係會改變。值得注意的是,在本新型之一實施例中,若資料在上述複數個存取間隔內之任意時間點被存取,控制器108將時間標籤之數值設定回第二特定數值(例如:10)。 Figure 3A shows a second correspondence table of logical block addresses, physical addresses, time stamps, current access times, and last access time. In an embodiment of the present invention, the logical block address LBA1 corresponds to the physical address Y6, the logical block address LBA10 corresponds to the physical address Y7, and the logical block address LBA23 corresponds to the physical address Y8, the logical area The block address LBA37 corresponds to the physical address Y9, the logical block address LBA49 corresponds to the physical address Y10, the time tag has a second specific value (for example: 10), and the current access time is 13:05, last time The access time is 13:00. The time difference between this access time and the last access time is an access interval. When the access interval is larger, the value of the time stamp is smaller. In one embodiment of the present invention, each time the data is accessed in the acceleration zone 112, both the current access time and the last access time are updated. For example, when the data is accessed in the acceleration zone 112 at 13:06, the access time is changed from 13:05 to 13:06, and the last access time is changed from 13:00 to 13:05. Controller 108 performs an operation on the value of the time stamp. In one embodiment of the present invention, when the physical address is in the acceleration zone 112, the controller 108 sets the value of the time stamp to a second specific value (eg, 10). In another embodiment of the present invention, as shown in FIG. 3B, when the data accessed in the acceleration zone 112 is not accessed every access interval (eg, 30 minutes), the controller 108 sets the time stamp. The value is decremented by one. For example, the last access time is 13:05, the current access time is 13:35, and the time interval between the last access time and the current access time is greater than or equal to 30 minutes, and the controller 108 sets the time stamp. The value is reduced to 9. In addition, when the value of the time stamp is zero, the data is moved from the acceleration zone 112 to the non-acceleration zone 114 for access. In another embodiment of the present invention, as shown in FIG. 3C, the data accessed in the acceleration zone 112 is not received in a plurality of access intervals (eg, 10*30 minutes=5 hours). access. For example, once the access time is after 18:05 (meaning that the time interval between the current access time and the last access time is greater than or equal to 10*30 minutes = 5 hours), the controller 108 will time. The value of the label is reduced to zero. As a result, the data is moved from the acceleration zone 112 to the non-acceleration zone 114 for access. That is to say, the correspondence between the logical block address and the physical address changes. It should be noted that in one embodiment of the present invention, if the data is accessed at any point in time within the plurality of access intervals, the controller 108 sets the value of the time stamp back to a second specific value (eg, 10). ).

第4圖顯示依據本新型之電子裝置100之操作方法之一流程圖。首先,步驟S10中,當控制器108接收來自中央處理單元102之一指令時,根據上述指令判斷複數個邏輯區塊位址是否為隨機分布且邏輯區塊位址所對應之資料之容量是否小於一第一特定數值。也就是說,控制器108可根據指令所包括之內容以及邏輯區塊位址是否為隨機分布,以決定資料於加速區112或非加速區114中進行存取。當邏輯區塊位址不為隨機分布或邏輯區塊位址所對應之資料之容量不小於一第一特定數值時,進入步驟S20。在步驟S20中,控制器108判斷出對應於邏輯區塊位址之實體位址在非加速區114中,故於非加速區114中存取對應的資料,且邏輯區塊位址與實體位址係根據第一對應關係表進行對應(見第2圖)。在本新型之一實施例中,以邏輯區塊位址是否為一連續分布作為判斷邏輯區塊位址是否為隨機分布的標準。舉例而言,邏輯區塊位址係為連續分布,就不是隨機分布。在本新型之一實施例中,複數個邏輯區塊位址LBA1、LBA2、LBA3、LBA4、LBA5屬於一種連續分布。在 本新型之另一實施例中,邏輯區塊位址所對應之資料之容量為100kb,第一特定數值為64kb。也就是說,邏輯區塊位址所對應之資料之容量不小於一第一特定數值。當邏輯區塊位址為隨機分布且邏輯區塊位址所對應之資料之容量小於第一特定數值時,進入步驟S30。在步驟S30中,控制器108判斷出對應於邏輯區塊位址之實體位址在加速區112中,故於加速區112中存取對應的資料,並且邏輯區塊位址、實體位址、時間標籤、本次存取時間以及上次存取時間係根據第二對應關係表進行對應(見第3A-3C圖)。在本新型之一實施例中,邏輯區塊位置LBA1、LBA2、LBA3、LBA4、LBA7不屬於一種連續分布(意即隨機分布),且邏輯區塊位址所對應之資料之容量為30kb,第一特定數值為64kb。也就是說,邏輯區塊位址所對應之資料之容量小於第一特定數值。接著,進入步驟S40。於步驟S40中,控制器108根據時間標籤之數值是否為零判斷是否要回非加速區114中存取資料。舉例而言,資料在存取間隔內沒被存取,時間標籤之數值會遞減。當時間標籤之數值遞減至零,則回至步驟S20。舉例而言,若在存取間隔(例如:30分鐘、1小時,諸如此類的時間)內資料沒有進行存取,控制器108將時間標籤之數值減一,若再經過存取間隔(例如:30分鐘、1小時,諸如此類的時間)內資料沒有進行存取,控制器108再將時間標籤之數值減一,以此類推。若經過複數個特定時間,資料皆沒有進行存取,數值會降至零。值得注意的是,一旦資料進行存取,時間標籤之數值回至第二特定數值。在本新型之一實施例中,第二特定數值為10。若時間標籤之數值沒有遞減至零,則 繼續步驟S30。 Figure 4 is a flow chart showing one of the methods of operation of the electronic device 100 in accordance with the present invention. First, in step S10, when the controller 108 receives an instruction from the central processing unit 102, it is determined according to the above instruction whether the plurality of logical block addresses are randomly distributed and the capacity of the data corresponding to the logical block address is less than A first specific value. That is, the controller 108 can determine whether the data is accessed in the acceleration zone 112 or the non-acceleration zone 114 according to whether the content included in the instruction and the logical block address are randomly distributed. When the logical block address is not randomly distributed or the capacity of the data corresponding to the logical block address is not less than a first specific value, the process proceeds to step S20. In step S20, the controller 108 determines that the physical address corresponding to the logical block address is in the non-acceleration area 114, so the corresponding data is accessed in the non-acceleration area 114, and the logical block address and the physical bit are accessed. The address is mapped according to the first correspondence table (see Figure 2). In an embodiment of the present invention, whether the logical block address is a continuous distribution is used as a criterion for judging whether the logical block address is a random distribution. For example, logical block addresses are continuously distributed and are not randomly distributed. In an embodiment of the present invention, the plurality of logical block addresses LBA1, LBA2, LBA3, LBA4, and LBA5 belong to a continuous distribution. in In another embodiment of the present invention, the data corresponding to the logical block address has a capacity of 100 kb, and the first specific value is 64 kb. That is to say, the capacity of the data corresponding to the logical block address is not less than a first specific value. When the logical block address is randomly distributed and the capacity of the data corresponding to the logical block address is smaller than the first specific value, the process proceeds to step S30. In step S30, the controller 108 determines that the physical address corresponding to the logical block address is in the acceleration area 112, so the corresponding data is accessed in the acceleration area 112, and the logical block address, the physical address, The time stamp, the current access time, and the last access time are corresponding according to the second correspondence table (see Figures 3A-3C). In an embodiment of the present invention, the logical block locations LBA1, LBA2, LBA3, LBA4, and LBA7 do not belong to a continuous distribution (ie, randomly distributed), and the capacity of the data corresponding to the logical block address is 30 kb. A specific value is 64 kb. That is to say, the capacity of the data corresponding to the logical block address is smaller than the first specific value. Next, the process proceeds to step S40. In step S40, the controller 108 determines whether to return to the non-acceleration area 114 to access the data according to whether the value of the time stamp is zero. For example, if the data is not accessed during the access interval, the value of the time stamp will be decremented. When the value of the time stamp is decremented to zero, it returns to step S20. For example, if the data is not accessed during an access interval (eg, 30 minutes, 1 hour, or the like), the controller 108 decrements the value of the time stamp by one, and then passes the access interval (eg, 30). The data is not accessed within minutes, hours, and the like, and the controller 108 decrements the value of the time stamp by one, and so on. If the data is not accessed after a certain number of specific times, the value will drop to zero. It is worth noting that once the data is accessed, the value of the time stamp is returned to the second specific value. In one embodiment of the present invention, the second specific value is 10. If the value of the time stamp is not decremented to zero, then Proceed to step S30.

因此,本新型所揭露之電子裝置以及其應用方法,可使得固態硬碟之部分資料在3D XPoint記憶體(即加速區112)中進行存取,而不是所有資料都在NAND快閃記憶體(即非加速區114)中進行存取,進而使得NAND快閃記憶體可以獲得更大的預留空間(Over Provisioning),使得NAND快閃記憶體更有效率地去處理NAND快閃記憶體的耗損平均(Wear Leveling)、垃圾回收(Garbage Collection)、壞塊管理(Bad Block Management)等工作,甚至能利用3D XPoint記憶體來減少固態硬碟的寫入放大因子,讓固態硬碟壽命更長,同時也使得固態硬碟在處理特定資料時的處理速度大幅地獲得提升。 Therefore, the electronic device disclosed in the present invention and the application method thereof can make part of the data of the solid state hard disk access in the 3D XPoint memory (ie, the acceleration area 112), instead of all the data in the NAND flash memory ( That is, access is made in the non-acceleration area 114), so that the NAND flash memory can obtain more Over Provisioning, so that the NAND flash memory can more efficiently process the loss of the NAND flash memory. Work such as Wear Leveling, Garbage Collection, Bad Block Management, etc., can even use 3D XPoint memory to reduce the write amplification factor of SSDs, and make SSDs last longer. At the same time, the processing speed of the solid state hard disk in processing specific data is greatly improved.

雖然本新型已以數個較佳實施例揭露如上,然其並非用以限定本新型,任何所屬技術領域中具有通常知識者,在不脫離本新型之精神和範圍內,當可作任意之更動與潤飾,因此本新型之保護範圍當視後附之申請專利範圍所界定者為準。 Although the present invention has been described above in terms of several preferred embodiments, it is not intended to limit the present invention, and any one of ordinary skill in the art can make any changes without departing from the spirit and scope of the present invention. And the scope of protection of this new type is subject to the definition of the scope of the patent application.

100‧‧‧電子裝置 100‧‧‧Electronic devices

102‧‧‧中央處理單元 102‧‧‧Central Processing Unit

104‧‧‧進階主機控制器介面 104‧‧‧Advanced Host Controller Interface

106‧‧‧固態硬碟 106‧‧‧ Solid State Drive

108‧‧‧控制器 108‧‧‧ Controller

110‧‧‧記憶體 110‧‧‧ memory

112‧‧‧加速區 112‧‧‧ acceleration zone

114‧‧‧非加速區 114‧‧‧Non-acceleration zone

Claims (10)

一種電子裝置,包括:一中央處理單元;以及一控制器,用以:當接收來自上述中央處理單元之一指令時,根據上述指令判斷複數個邏輯區塊位址是否為一隨機分布且上述邏輯區塊位址所對應之資料的容量是否小於一第一特定數值;當上述邏輯區塊位址為上述隨機分布且上述資料之容量小於上述第一特定數值,判斷為對應於上述邏輯區塊位址之複數個實體位址在一加速區中,上述資料於上述加速區中進行存取;以及當上述邏輯區塊位址不為上述隨機分布或上述資料之容量不小於上述第一特定數值,判斷為對應於上述邏輯區塊位址之上述實體位址在一非加速區中,上述資料於上述非加速區中進行存取。 An electronic device comprising: a central processing unit; and a controller for: determining, when the instruction from one of the central processing units is received, determining, according to the instruction, whether the plurality of logical block addresses are a random distribution and the logic Whether the capacity of the data corresponding to the block address is less than a first specific value; when the logical block address is randomly distributed and the capacity of the data is less than the first specific value, it is determined to correspond to the logical block bit The plurality of physical addresses of the address are in an acceleration zone, and the data is accessed in the acceleration zone; and when the logical block address is not the random distribution or the capacity of the data is not less than the first specific value, It is determined that the physical address corresponding to the logical block address is in a non-acceleration area, and the data is accessed in the non-acceleration area. 如申請專利範圍第1項所述之電子裝置,其中上述電子裝置更包括一記憶體,用以儲存一第一對應關係表以及一第二對應關係表。 The electronic device of claim 1, wherein the electronic device further comprises a memory for storing a first correspondence table and a second correspondence table. 如申請專利範圍第2項所述之電子裝置,其中上述第一對應關係表包括在上述非加速區中之上述邏輯區塊位址與上述實體位址,以及上述第二對應關係表包括在上述加速區中之上述邏輯區塊位址與上述實體位址,上述第二對應關係表更包括一時間標籤,上述時間標籤係根據一本次存取時間以及一上次存取時間所產生之一存取間隔所決定。 The electronic device of claim 2, wherein the first correspondence table includes the logical block address and the physical address in the non-acceleration area, and the second correspondence table is included in the foregoing The logical block address in the acceleration zone and the physical address, the second correspondence table further includes a time tag, and the time tag is generated according to a current access time and a last access time. The access interval is determined. 如申請專利範圍第3項所述之電子裝置,其中上述時間標籤具有一數值,當上述實體位址在上述加速區中,上述控制器將上述數值設定為一第二特定數值。 The electronic device of claim 3, wherein the time stamp has a value, and when the physical address is in the acceleration zone, the controller sets the value to a second specific value. 如申請專利範圍第4項所述之電子裝置,其中當於上述加速區中進行存取之上述資料在上述存取間隔內沒進行存取,上述控制器將上述第二特定數值減一。 The electronic device of claim 4, wherein the controller accesses the data in the acceleration zone without accessing the access interval, and the controller decrements the second specific value by one. 如申請專利範圍第4項所述之電子裝置,其中當上述數值為零,上述資料從於上述加速區中移至上述非加速區進行存取。 The electronic device of claim 4, wherein when the value is zero, the data is moved from the acceleration zone to the non-acceleration zone for access. 如申請專利範圍第1項所述之電子裝置,其中上述指令為一SATA指令。 The electronic device of claim 1, wherein the instruction is a SATA command. 如申請專利範圍第1項所述之電子裝置,其中上述指令為一PCIe指令。 The electronic device of claim 1, wherein the instruction is a PCIe instruction. 如申請專利範圍第1項所述之電子裝置,其中上述加速區為一3D XPoint記憶體。 The electronic device of claim 1, wherein the acceleration zone is a 3D XPoint memory. 如申請專利範圍第1項所述之電子裝置,其中上述非加速區為一NAND快閃記憶體。 The electronic device of claim 1, wherein the non-acceleration zone is a NAND flash memory.
TW105201300U 2016-01-28 2016-01-28 Electronic device TWM524503U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW105201300U TWM524503U (en) 2016-01-28 2016-01-28 Electronic device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW105201300U TWM524503U (en) 2016-01-28 2016-01-28 Electronic device

Publications (1)

Publication Number Publication Date
TWM524503U true TWM524503U (en) 2016-06-21

Family

ID=56757872

Family Applications (1)

Application Number Title Priority Date Filing Date
TW105201300U TWM524503U (en) 2016-01-28 2016-01-28 Electronic device

Country Status (1)

Country Link
TW (1) TWM524503U (en)

Similar Documents

Publication Publication Date Title
JP7089830B2 (en) Devices, systems, and methods for write management of non-volatile memory data
US9928167B2 (en) Information processing system and nonvolatile storage unit
JP6414853B2 (en) Memory system and control method
US10114556B2 (en) Method and apparatus for improving read performance of a solid state drive
JP6414852B2 (en) Memory system and control method
US9645924B2 (en) Garbage collection scaling
JP2017021805A (en) Interface providing method capable of utilizing data attribute reference data arrangement in nonvolatile memory device and computer device
US11270734B2 (en) Method for accessing shingled magnetic recording SMR disk, and server
US10915791B2 (en) Storing and retrieving training data for models in a data center
TWI746510B (en) Technologies for managing immutable data on a data storage device
US11675709B2 (en) Reading sequential data from memory using a pivot table
US11620066B2 (en) Storage device with expandible logical address space and operating method thereof
US10997080B1 (en) Method and system for address table cache management based on correlation metric of first logical address and second logical address, wherein the correlation metric is incremented and decremented based on receive order of the first logical address and the second logical address
CN111796765B (en) Method for optimizing performance of data storage system
EP3506116A1 (en) Shared memory controller in a data center
TWI501081B (en) Wear leveling with marching strategy
US20230393753A1 (en) Wear leveling in solid state drives
TW201805811A (en) Memory module and method for controlling a programmable deduplication ratio thereof
CN105917303A (en) Controller, method for identifying data block stability and storage system
US9524236B1 (en) Systems and methods for performing memory management based on data access properties
CN111324549B (en) Memory and control method and device thereof
CN114610654A (en) Solid-state storage device and method for writing data into solid-state storage device
CN105138294B (en) A kind of hard disc apparatus and information processing method
TWM524503U (en) Electronic device
TW202311967A (en) Memory module, memory system including memory module, and method of operating the same

Legal Events

Date Code Title Description
MM4K Annulment or lapse of a utility model due to non-payment of fees