WO2016187801A1 - 一种用于储存介质的数据管理方法及固态硬盘与控制芯片 - Google Patents

一种用于储存介质的数据管理方法及固态硬盘与控制芯片 Download PDF

Info

Publication number
WO2016187801A1
WO2016187801A1 PCT/CN2015/079794 CN2015079794W WO2016187801A1 WO 2016187801 A1 WO2016187801 A1 WO 2016187801A1 CN 2015079794 W CN2015079794 W CN 2015079794W WO 2016187801 A1 WO2016187801 A1 WO 2016187801A1
Authority
WO
WIPO (PCT)
Prior art keywords
host
solid state
hard disk
state hard
read
Prior art date
Application number
PCT/CN2015/079794
Other languages
English (en)
French (fr)
Inventor
崔开平
Original Assignee
崔开平
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 崔开平 filed Critical 崔开平
Priority to PCT/CN2015/079794 priority Critical patent/WO2016187801A1/zh
Publication of WO2016187801A1 publication Critical patent/WO2016187801A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers

Definitions

  • the present invention relates to a data management method for a storage device, a storage device and a control chip thereof, and more particularly to a data management method for a storage medium and a solid state hard disk and a control chip thereof.
  • the existing hard disks can be roughly divided into two types: Hard Disk Drive and Solid State Drive.
  • the biggest difference between the two is that the magnetic hard disk uses the disk as the data storage medium, while the solid state hard disk uses the flash memory as the storage medium. .
  • the data management method between the host computer and the solid state hard disk is that the host outputs the data to be read and written to the solid state hard disk, and the solid state hard disk must pass the special data processing program according to the access rule of the flash memory, and then the flash memory.
  • Data reading and writing, and its data processing program is similar to the existing USB portable memory, including: Data management/Wear leveling/FTL/ECC/Data security/Garbage collection, etc.
  • the data management method of the architecture is that the host outputs the data to be read and written to the solid state hard disk, and the solid state hard disk must pass the special data processing program according to the access rule of the flash memory, and then the flash memory.
  • Data reading and writing, and its data processing program is similar to the existing USB portable memory, including: Data management/Wear leveling/FTL/ECC/Data security/Garbage collection, etc.
  • the data management method of the architecture is that the host outputs the data to be read and written to the solid state hard disk, and the solid state hard disk must pass the special
  • Initialization setting that is, the host obtains various main information of the solid state hard disk, such as model, hard disk capacity, space division and the like;
  • the host outputs a file access instruction, that is, an instruction issued by the host to write/read the plurality of files to the solid state hard disk;
  • the solid state hard disk executes a data processing program, and the data processing program is executed by the control chip according to the access rule of the flash memory by the file to be written/read by the host, that is, the program processing including the above data management/Wear leveling/FTL/ECC/Data security is performed.
  • Write/read files to complete the steps of file compression, address matching, and cutting;
  • the hardware architecture of the existing solid state hard disk control chip 10 must include at least: A memory access controller 11 (DMA controller), a central program processor 12 (CPU), a dynamic random access memory controller 13 (DRAM controller), a buffer memory 14 (Local Buffer Memory), a data encryption and decryption engine 15 (Data Security Engine), The debug engine 16 (ECC Engine), the read queue management module 17 (Read Queue Management), and the write buffer management module 18 (Write Buffer Management); and such a complicated hardware architecture, there are the following drawbacks:
  • DMA controller memory access controller 11
  • CPU central program processor 12
  • DRAM controller dynamic random access memory controller 13
  • buffer memory 14 Lical Buffer Memory
  • Data Security Engine Data Security Engine
  • ECC Engine The debug engine 16
  • the read queue management module 17 Read Queue Management
  • the write buffer management module 18 Write Buffer Management
  • the control chip has a large amount of computation and is easy to drag and drop data reading and writing speed.
  • control chip is responsible for executing many and complicated programs, the development cost is high, and the development time required is long.
  • the design of the control chip needs to be different according to the design of each flash memory. With the architecture, the developers of control chips often need to cooperate with the flash memory design.
  • development time required for the current control chip from the beginning of the project to the time of market launch, usually takes more than 2 years, and the product cycle is short. For SSDs, too long development time will be quite unfavorable.
  • the present invention provides a data management method for the solid state hard disk architecture and the solid state hard disk and the control chip.
  • the data management method for the storage medium includes the following steps:
  • the corresponding file is accessed in a set of flash memory.
  • the data management method for the solid state hard disk architecture of the present invention is to upload a data processing program to be processed by the original control chip to the host when the solid state hard disk is connected to the host, and the host executes a data processing program originally executed by the control chip.
  • the file to be accessed is processed and then accessed by the SSD, so that the following effects can be achieved:
  • control chip only needs to perform the operations of writing and reading, which can save many data processing steps, reduce the computational burden, and improve the data access speed.
  • the present invention changes the data processing program originally executed by hardware into software and is executed by the host, thereby shortening the development time and improving the development time. Manufacturer competitiveness.
  • the SSD includes a control chip and a set of flash memory, and the control chip is connected to the set of flash memory and includes:
  • host interface for connecting to a host
  • ROM boot ROM memory
  • a read/write control unit is connected to the host control interface and the set of flash memories. After receiving a file access command output by the host, the corresponding file is taken for the set of flash memories.
  • control chip of the solid state hard disk includes:
  • host interface for connecting to a host
  • ROM boot ROM memory
  • a flashing interface for connecting to a set of flash memories
  • a read/write control unit is connected to the host control interface and the set of flash interfaces. After receiving a file access command output by the host, accessing the corresponding file to the set of flash memories through the set of flash interfaces.
  • Figure 1 Flowchart for a prior data management method for a solid state drive architecture.
  • Figure 2 A block diagram of the hardware architecture of an existing solid state drive.
  • FIG. 3 is a flow chart of a data management method for a solid state hard disk architecture of the present invention.
  • Figure 4 is a detailed flow chart of one of the steps of Figure 3.
  • FIG. 5 is another detailed flow chart of one of the steps of Figure 3.
  • FIG. 6 is a flow chart showing an implementation of the solid state hard disk connected to the host according to the present invention.
  • FIG. 7 is a flow chart showing another implementation of the SSD of the present invention after being connected to the host.
  • Figure 8 is a block diagram showing the hardware architecture of the solid state hard disk of the present invention.
  • control chip 11, direct memory access controller; 12, central program processor; 13, dynamic random access memory controller 14, buffer memory; 15, data encryption and decryption engine; 16, debugging Engine 17, read queue management module 18, write buffer management module 20, host; 30, solid state drive 40, control chip; 41, host control interface 42, boot boot read-only memory module 43 flash control interface; 44, read Queue management module; 45, write buffer management module; 50, flash memory.
  • the data management method for the storage medium architecture of the present invention includes the following steps.
  • the storage medium is implemented by a control chip like a solid state hard disk composed of a flash memory, and the control chip is used for corresponding connection and control.
  • the data accesses a flash memory and communicates with a host to perform a corresponding control and access procedure.
  • the steps of the method include:
  • the data processing program includes data management/over-provision/average spoofing storage block technology corresponding to the flash access rule (Wear leveling) / / Flash translation layer (FTL) / error checking and correction code (ECC) / data security (Data security) / garbage collection (Garbage collection) and other data processing steps, the details will be detailed later;
  • the chip Connecting a host, the chip is connected between the flash memory and the host, and when the host is turned on, connecting with the host and transmitting communication;
  • Uploading the data processing program to the host so that the host can process the file to be accessed as a file access instruction that can be directly written or directly read, and can include file compression, encryption and decryption in a better processing manner.
  • the file is cut into a flash-compatible cluster, access address matching, optimized data writing, confirmation of data write/read is correct, error checking and correction of read data is performed, Properly handle garbage collection, properly maintain reserved space, etc.
  • the corresponding file is accessed in a set of flash memories.
  • the control chip can directly access the data in the corresponding address of the flash memory.
  • the data access method of the embodiment is performed by a control chip integrated in a solid-state hard disk to perform calculation steps such as compression, encryption and decryption, data cutting, etc., to a hardware execution calculation (for example, a central processing unit) of the host, and the control chip is transparent.
  • a control chip integrated in a solid-state hard disk to perform calculation steps such as compression, encryption and decryption, data cutting, etc., to a hardware execution calculation (for example, a central processing unit) of the host, and the control chip is transparent.
  • the firmware or software control it is responsible for the access communication with the connected flash memory, and through the above steps, the majority of the calculation program is transferred to the host to run, so that the control function required by the solid state hard disk itself can be greatly reduced, the control chip Just need to include the base ditch with the flash and the host Through interface and storage space, this can greatly reduce hardware development time, reduce the price and make it possible to adapt to the flash memory of a new different hardware architecture or process.
  • the step of uploading the data processing program to the host may be implemented by the control chip performing an initialization process when the SSD is connected to the host for the first time.
  • the program consists of the following steps:
  • the data processing program is uploaded from the read only memory to the host to complete the initialization process;
  • the boot boot is performed by a flash memory, and the data processing program is uploaded from the flash memory to the host to complete the boot initialization procedure.
  • the purpose of using the read-only memory or the flash memory to boot is determined because the solid-state hard disk is connected to the host for the first time and the booting process is executed, and then the read-only memory is selected to complete the first booting process to make the solid-state hard disk. After the format is completed and the boot disk is specified, the subsequent boot can be replaced by the flash memory's specific address instead of the read-only memory boot mode used for the initial boot.
  • a configuration management program is further executed. Please refer to FIG. 5 further.
  • the configuration management program is further described below, which includes the following steps:
  • the solid state hard disk is set as a stand-alone device, and the configuration management program is completed;
  • SSD storage layer is tiered, all SSDs are configured in the same file transfer layer comparison table (FTL table), and the average storage shard block technology can be implemented across the SSDs to extend the life of the SSD and improve The speed of data reading and writing to complete the configuration management program;
  • FTL table file transfer layer comparison table
  • cluster configuration If the cluster configuration is not enabled, configure the multiple drive IDs to the multiple solid state drives and complete the configuration management program.
  • the host can access the SSD of the present invention. Please refer to FIG. 6 further.
  • the host When a host writes a file to the SSD of the present invention, the host will Perform the following steps:
  • the data processing program includes establishing a flash translation layer (FTL table) / average rewritable storage block technology / garbage collection (Garbage collection) / data security / reserved space and other steps, if file compression, the above data processing program is executed after the file is compressed;
  • FTL table flash translation layer
  • Garbage collection garbage collection
  • ECC Error-correcting code
  • the storage address corresponding to the file to be read is obtained, and output to the solid state hard disk;
  • ECC error checking and correction code
  • the data processing program (including the establishment of the flash conversion layer comparison table (FTL table) / average rewritable storage block technology / garbage collection) that should be processed by the control chip of the solid state hard disk has been handed over.
  • the host performs, and the hardware architecture of the solid state hard disk control chip is simplified.
  • the hardware architecture of the solid state hard disk and its control chip of the present invention is further explained below.
  • the solid state hard disk 30 of the present invention includes a control chip 40 and a set of flash memories 50 .
  • the control chip 40 includes:
  • a host control interface 41 is used to connect to a host 20;
  • a boot ROM module 42 is connected to the host control interface 41 and stores a data processing program. When the host control interface 41 is connected to the host 20, the host 20 uploads the data. Processing program
  • a flash memory interface 43 for connecting to the set of flash memories 50;
  • a read/write control unit is connected to the host control interface 41 and the set of flash interfaces 42. After receiving a file access command output by the host, accessing the corresponding file to the set of flash memories through the set of flash interfaces 43 is performed.
  • the read/write control unit includes a read queue management module 44 (Read Queue Management), a write buffer management module 45 (Write Buffer Management), and a data output input control module 46 (UART/GPIO/). IIC).
  • the boot boot ROM module 42, the read queue management module 44, the write buffer management module 45, and the data output input control module 46 are implemented by a non-volatile read only memory (NV-ROM).
  • NV-ROM non-volatile read only memory
  • the solid state hard disk control chip can be simplified to four main hardware architectures as shown in FIG. 8:
  • the read queue management module 44 when the host issues a read command, directly reads the corresponding file directly from the address corresponding to the flash memory according to the address output by the host, and outputs the corresponding file to the host for subsequent error check and correction code. And the decompression step.
  • the write buffer management module 45 (Write Buffer Management), when the host issues a write command, directly writes the file directly into the flash memory according to the storage address of the host pair.
  • the data output is input to the control module 46 for performing data output input queue management.
  • the invention relates to a data management method for a solid-state hard disk architecture, and the complicated data processing program is disposed by the host, and then the solid-state hard disk performs data reading and writing, thereby reducing the computing burden of the control chip and simplifying the hardware structure of the control chip and In addition to reducing development costs, because the software development time is shorter than the hardware development time, the development time of the SSD control chip can be shortened, thereby enhancing the competitiveness of developers.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Stored Programmes (AREA)

Abstract

一种用于储存介质的数据管理方法及固态硬盘(30)与控制芯片(40)，其数据管理方法包含载入预先储存一数据处理程序、上传该数据处理程序至一主机(20)、接收该主机(20)输出的一文件存取指令以及依据该文件存取指令，在一组闪存(50)中存取对应文件，通过上述手段，可将固态硬盘控制芯片(40)需处理的数据处理程序交由主机(20)执行，进而减少固态硬盘控制芯片(40)的运算负担，并简化控制芯片(40)的硬件架构，减少开发成本、缩短开发时程。

Description

一种用于储存介质的数据管理方法及固态硬盘与控制芯片 技术领域
本案是关于一种储存装置的数据管理方法及其储存装置与控制芯片，尤指一种用于储存介质的数据管理方法及其固态硬盘与控制芯片。
背景技术
现有硬盘可大致分为两种：磁性硬盘(Hard Disk Drive)及固态硬盘(Solid State Drive)，两者最大的差异在于磁性硬盘以磁盘作为数据储存媒介，而固态硬盘则以闪存作为储存媒介。
目前电脑主机与固态硬盘之间的数据管理方法，是由主机将欲读写的数据输出至固态硬盘，而固态硬盘必须依据闪存的存取规则，将文件经过特殊的数据处理程序，再对闪存进行数据读写，而其数据处理程序与现有USB随身存储器类似，主要包含了:Data management/Wear leveling/FTL/ECC/Data security/Garbage collection等，以下进一步配合图示说明现有用于固态硬盘架构的数据管理方法。
请参阅图1，依照现有用于固态硬盘架构的数据管理方法，固态硬盘与一主机连接时，将执行以下步骤：
初始化设定，即该主机取得固态硬盘的各种主要资讯，如型号、硬盘容量、空间分割等资讯；
主机输出文件存取指令，即由主机下达将复数文件写入/读取固态硬盘的指令；
固态硬盘执行数据处理程序，通过控制芯片将主机欲写入/读取的文件依据闪存的存取规则进行数据处理程序，即包含上述Data management/Wear leveling/FTL/ECC/Data security等程序处理欲写入/读取的文件，以完成将文件压缩、位址匹配(mapping)、切割等步骤；
在闪存中存取对应文件。
请参阅图2，由于固态硬盘的控制芯片10必须在接收主机20的文件存取指令后，先执行一连串的数据处理程序，因此现有固态硬盘控制芯片10的硬件架构至少须包含有：一直接内存存取控制器11(DMA controller)、一中央程序处理器12(CPU)、一动态随机存储器控制器13(DRAM controller)、一缓冲内存14(Local Buffer Memory)、数据加解密引擎15(Data Security Engine)、 除错引擎16(ECC Engine)、读取队列管理模块17(Read Queue Management)及写入缓冲管理模块18(Write Buffer Management)；而如此复杂的硬件架构，存在有以下缺陷：
1.控制芯片运算量大，容易拖垮数据读写速度。
2.开发时间长、开发成本高：诚如上述，控制芯片负责执行的程序多且繁杂，开发本高，开发所需时程也长，加上控制芯片的设计需因应各家闪存不同的设计与架构，使得控制芯片的开发厂商常需配合闪存变更设计，而就目前控制芯片所需的开发时程，从计画开始到产品上市，通常需花费2年多的时间，对于产品周期短的固态硬盘而言，太长的开发时程将显得相当不利。
目前半导体亦逐渐导入先进制程，未来各家闪存针对NAND Flash的制程、设计及结构将越来越不一致，而固态硬盘的控制芯片的开发需求将变得越来越频繁，太长的开发时程将对厂商相当不利，上述缺陷实有需进一步解决。
发明内容
有鉴于上述固态硬盘的控制芯片有运算量大、降低速度、开发成本高及开发耗时等技术缺陷，本发明是提出一种用于固态硬盘架构的数据管理方法及该固态硬盘与控制芯片。
欲达上述目的，是令该用于储存介质的数据管理方法包含有以下步骤：
预先储存一数据处理程序；
连接一主机；
上传(Upload)该数据处理程序至该主机；
接收该主机输出的一文件存取指令；
依据该文件存取指令，在一组闪存中存取对应文件。
本发明用于固态硬盘架构的数据管理方法中，是令固态硬盘与主机连接时，将原本控制芯片须处理的数据处理程序上传至该主机，由主机执行原本由控制芯片执行的数据处理程序，将欲存取的文件处理后再交由固态硬盘进行存取，如此，即可达到以下功效:
1.减少控制芯片的运算负担：控制芯片只需执行写入及读取的动作，可省去许多数据处理步骤，减少运算负担，提升数据存取速度。
2.简化控制芯片硬件架构：原本需由控制芯片执行的部分数据处理程序交由主机执行后，控制芯片的硬件架构便可简化。而控制芯片的硬件简化后，即可降低开发成本。
3.除上述优点之外，由于软件开发时程较硬件开发时程短，本发明将原本以硬件执行的数据处理程序改为软件，并由主机来执行，即可缩短其开发时程，提升厂商竞争力。
欲达上述目的，是令该固态硬盘包含有一控制芯片与一组闪存，该控制芯片是与该组闪存连接，且包含有：
一主机控制接口(host interface)，系用以与一主机连结；
一引导只读存储器模块(Boot ROM memory)，其与该主机控制接口连接，并储存有一数据处理程序，当该主机控制接口与该主机连接后，供该主机上传该资料处理程序；
一读写控制单元，是与该主机控制接口及该组闪存连接，在接收该主机输出的一文件存取指令后，对该组闪存取对应文件。
又，欲达上述目的，是令该固态硬盘的控制芯片包含有：
一主机控制接口(host interface)，系用以与一主机连结；
一引导只读存储器模块(Boot ROM memory)，其与该主机控制接口连接，并储存有一数据处理程序，当该主机控制接口与该主机连接后，供该主机上传该数据处理程序；
一快闪闪存接口，是用以与一组闪存连接；
一读写控制单元，是与该主机控制接口及该组闪存接口连接，在接收该主机输出的一文件存取指令后，透过该组闪存接口对该组闪存存取对应文件。
附图说明
图1：为现有用于固态硬盘架构的数据管理方法的流程图。
图2：为现有固态硬盘的硬体架构方块图。
图3：为本发明用于固态硬盘架构的数据管理方法的流程图。
图4：为图3其中一步骤的详细流程图。
图5：为图3其中一步骤的另一详细流程图。
图6：为本发明固态硬盘与主机连接后一实施流程图。
图7：为本发明固态硬盘与主机连接后另一实施流程图。
图8：为本发明固态硬盘之硬件架构方块图。
图中，10、控制芯片；11、直接内存存取控制器；12、中央程序处理器；13、动态随机存储器控制器 14、缓冲内存；15、数据加解密引擎；16、除错 引擎 17、读取队列管理模块 18、写入缓冲管理模块 20、主机；30、固态硬盘40、控制芯片；41、主机控制接口 42、启动引导只读存储器模块 43闪存控制接口；44、读取队列管理模块；45、写入缓冲管理模块；50、闪存。
具体实施方式
请配合参阅图3，本发明用于储存介质架构的数据管理方法包含有以下步骤，该储存介质像利用闪存组成的固态硬盘，是由一控制芯片执行，该控制芯片为用于对应连接并控制数据存取一闪存，并与一主机双向沟通执行对应控制与存取程序，该方法的步骤包含：
预先储存一数据处理程序，在本实施例中，该数据处理程序包含对应闪存存取规则的数据处理(Data management)/预留空间(Over-Provision)/平均抹写储存区块技术(Wear leveling)/闪存转换层(FTL)/错误检查和纠正码(ECC)/数据保密(Data security)/垃圾收回(Garbage collection)等数据处理步骤，详细内容在后详述；
连接一主机，该芯片连接在该闪存与该主机之间，在该主机开启时，与该主机连接并讯号传递沟通；
上传(Upload)该数据处理程序至该主机，使该主机得以将欲存取的文件处理为可直接写入或直接读取的文件存取指令，可以较佳的处理方式包含文件压缩、加解密、文件切割为符合闪存的群组(cluster)、存取位址匹配(mapping)、最佳化的数据写入、确认数据写入/读取的正确、执行读取数据的错误检查和纠正、适当的处理垃圾收回、适当的维持预留空间等；
接收该主机依据该数据处理程序处理后所输出的文件存取指令；
依据该数据存取指令，在一组闪存器中存取对应文件。
由于该主机已经将欲存取的文件进行压缩、加解密、切割及位址匹配步骤，故该控制芯片可直接在闪存对应地址中进行数据存取。
本实施例的数据存取方法是由整合在一固态硬盘的控制芯片将例如压缩、加解密、数据切割等演算步骤交由该主机的硬体执行演算(例如中央处理器)，该控制芯片透过固件或软件控制，负责与连接的该闪存的存取沟通，并透过上述步骤将多数的演算程序交由主机运行，如此，可大幅减少该固态硬盘本身所需的控制功能，该控制芯片仅需包含与该闪存与该主机的基础沟 通接口以及储存空间，如此可大幅降低硬件开发时间，减少售价并使可透过微幅调整即可适用于全新不同硬件架构或制程方式的闪存。
上述上传数据处理程序至主机的步骤，可在固态硬盘与主机第一次连接时，由控制芯片进行一初始化程序来实现，以下仅进一步说明固态硬盘的初始化程序，请进一步配合参阅图4，初始化程序包含以下步骤：
开启电源，开始初始化程序；
判断是否以只读存储器(ROM)执行启动引导(Boot)；
若判断以只读存储器执行启动引导，表示这是此固态硬盘的第一次使用，则将上述数据处理程序由一只读存储器上传至该主机，完成初始化程序；
若否，则以一闪存执行启动引导，将上述数据处理程序由一闪存上传至该主机，完成开机初始化程序。
本实施例判断利用只读存储器或闪存开机的目的，是因为该固态硬盘如为与主机首次连接并执行开机程序，则选择以只读存储器完成于前述第一次启动程序后，使该固态硬盘完成格式化、开机磁盘指定，则之后的开机可透过闪存的预设地址(specific address)，取代初始开机所使用的只读存储器开机方式。
本实施例中，上述初始化程序完成后，是进一步执行一组态管理程序，请进一步参阅图5，以下谨进一步说明组态管理程序，其包含以下步骤：
开始组态管理程序；
判断该主机是否连接复数个固态硬盘；
若判断为仅连接单个固态硬盘，则将本固态硬盘设定为单装置(stand-alone device)，并完成组态管理程序；
若判断为多个固态硬盘，则进一步判断是否致使(enable)固态硬盘储存分层(SSD pool)；
若致使固态硬盘储存分层，则将所有固态硬盘配置于同一文件传输层对照表(FTL table)，并且可以跨固态硬盘的执行平均抹写储存区块技术，可以达到延长固态硬盘的寿命以及提升数据读写的速度，以完成组态管理程序；
若不致使固态硬盘储存分层，则进一步判断是否致使带集区组态(RAID0 configuration)；
若致使带集区组态，则将所有固态硬盘配置一带集区，并完成组态管理程序；
若不致使带集区组态，则分别配置复数驱动编号(drive ID)给复数个固态硬盘，并完成组态管理程序。
固态硬盘与主机完成初始程序及组态管理程序后，即可由该主机对本发明固态硬盘进行文件存取，请进一步配合参阅图6，当一主机对本发明固态硬盘进行写入文件时，该主机将执行以下步骤：
确定欲写入的文件；
判断是否进行文件压缩，若否，则直接进入上述数据处理程序，在本实施例中，数据处理程序包含建立闪存转换层(FTL table)/平均抹写储存区块技术/垃圾收回(Garbage collection)/数据保密/预留空间等步骤，若进行文件压缩，则在文件压缩完毕后执行上述数据处理程序；
输出处理后的文件存取指令至固态硬盘；
接收固态硬盘回传错误检查和纠正码(ECC，Error-correcting code)；
执行上述数据处理程序中确认错误检查和纠正码(ECC)的步骤；
判断是否写入错误，若无，完成写入程序，若有，则回到将处理后文件输出至固态硬盘的步骤。
请进一步配合参阅图7，以下进一步说明该主机读取固态硬盘数据的步骤，其包含有：
依据闪存转换层对照表(FTL table)，取得欲读取文件对应的储存地址，并输出至固态硬盘；
接收固态硬盘回传文件；
进行错误检查和纠正码(ECC)确认，判断文件是否有错误，若有则回到接收固态硬盘回传文件的步骤，若无则进入下一步骤；
判断是否为压缩档，若是，则进行解压缩，若否，则完成读取文件的步骤。
诚如上述读写程序所描述，由于原本应由固态硬盘的控制芯片处理的数据处理程序(包含建立闪存转换层对照表(FTL table)/平均抹写储存区块技术/垃圾收回)已交由该主机执行，固态硬盘控制芯片的硬件架构得以简化，以下进一步说明本发明固态硬盘及其控制芯片的硬体架构。
请进一步配合参阅图8，本发明固态硬盘30包含有一控制芯片40及一组闪存50，其中该控制芯片40包含有：
一主机控制接口41(host interface)，是用以与一主机20连接；
一启动引导只读存储器模块42(Boot ROM memory)，其与该主机控制接口41连接，并储存有一数据处理程序，当该主机控制接口41与该主机20连接后，供该主机20上传该数据处理程序；
一闪存接口43，是用以与该组闪存50连接；
一读写控制单元，是与该主机控制接口41及该组闪存接口42连接，在接收该主机输出的一文件存取指令后，透过该组闪存接口43对该组闪存存取对应文件，在本实施例中，该读写控制单元包含一读取队列管理模块44(Read Queue Management)、一写入缓冲管理模块45(Write Buffer Management)及一数据输出输入控制模块46(UART/GPIO/IIC)。
上述启动引导只读存储器模块42、该读取队列管理模块44、该写入缓冲管理模块45及该数据输出输入控制模块46得以一非挥发性只读存储器(NV-ROM)实现。
由于本发明用于固态硬盘架构的数据管理方法中，已将数据处理程序交由主机处理，因此，固态硬盘控制芯片得以如图8所示，简化至四个主要的硬件架构：
启动引导只读存储器模组43，用以储存要上传至主机的数据处理程序；
读取队列管理模块44(Read Queue Management)，当主机下达读取指令时，直接依据主机输出的地址，直接于对应闪存的地址读取对应文件，并输出予主机进行后续的错误检查和纠正码及解压缩步骤。
写入缓冲管理模块45(Write Buffer Management)，当主机下达写入指令时，直接依据主机配对的储存地址，将文件直接写入闪存中。
数据输出输入控制模块46，用以进行数据输出输入队列管理。
本发明用于固态硬盘架构的数据管理方法，将复杂的数据处理程序交由主机处理，再由固态硬盘进行数据读写的配合方式，可达到减少控制芯片的运算负担、简化控制芯片硬件架构及降低开发成本以外，由于软件开发时程较硬件开发时程短，更可缩短固态硬盘控制芯片的开发时程，进而提升开发厂商的竞争力。

Claims (10)

  1. 用于储存介质的数据管理方法，其特征在于，包含有：
    预先储存一数据处理程序；
    连接一主机；
    上传该数据处理程序至该主机；
    接收该主机输出的一文件存取指令；
    依据该文件存取指令，在一组闪存中存取对应文件。
  2. 如权利要求1所述的用于储存介质的数据管理方法，其特征在于，是在连接该主机后，执行一初始化程序及一组态管理程序，以上传该数据处理程序至该主机。
  3. 如权利要求2所述的用于储存介质的数据管理方法，其特征在于，该初始化程序包含：
    开启电源，开始初始化程序；
    判断是否以只读存储器执行启动引导；
    若判断以只读存储器执行启动引导，则将上述数据处理程序由一只读存储器上传至该主机，以完成初始化程序；
    若否，则以一闪存执行启动引导，将上述数据处理程序由一闪存上传至该主机，以完成初始化程序。
  4. 如权利要求2所述的用于储存介质的数据管理方法，其特征在于，该组态管理程序包含：
    开始组态管理程序；
    判断该主机是否连接复数个固态硬盘；
    若判断为仅连接单个固态硬盘，则将本固态硬盘设定为单装置，并完成组态管理程序；
    若判断为多个固态硬盘，则进一步判断是否致使固态硬盘储存分层；
    若致使固态硬盘储存分层，则将所有固态硬盘配置于一文件传输层对照表，以完成组态管理程序；
    若不致使固态硬盘储存分层，则进一步判断是否致使带集区组态；
    若致使带集区组态，则将所有固态硬盘配置一带集区，并完成组态管理程序；
    若不致使带集区组态，则分别配置复数驱动编号给所有固态硬盘，并完成组态管理程序。
  5. 一种固态硬盘，其特征在于，包含有一控制芯片与一组闪存，该控制芯片是与该组闪存连接，且包含有：
    一主机控制接口，是用以与一主机连接；
    一引导只读存储器模块，其与该主机控制接口连接，并储存有一数据处理程序，当该主机控制接口与该主机连接后，供该主机上传该数据处理程序；
    一读写控制单元，是与该主机控制接口及该组闪存连接，在接收该主机输出的一文件存取指令后，对该组闪存存取对应文件。
  6. 如权利要求5所述的固态硬盘，其中，该读写控制单元包含有一读取队列管理模块、一写入缓冲管理模块及一数据输出输入控制模块。
  7. 如权利要求6所述的固态硬盘，其中，该引导只读存储器模块、该读取队列管理模块、该写入缓冲管理模块及该数据输出输入控制模块是以一非挥发性只读存储器实现。
  8. 一种固态硬盘的控制芯片，其特征在于，包含有：
    一主机控制接口，是用以与一主机连接；
    一引导只读存储器模块，其与该主机控制接口连接，并储存有一数据处理程序，当该主机控制接口与该主机连接后，供该主机上传该数据处理程序；
    一闪存接口，是用以与一组闪存连接；
    一读写控制单元，是与该主机控制接口及该组闪存接口连接，在接收该主机输出的一数据存取指令后，透过该组闪存接口对该组闪存存取对应文件。
  9. 如权利要求8所述的固态硬盘的控制芯片，其特征在于，，该读写控制单元包含有一读取队列管理模块、一写入缓冲管理模块及一数据输出输入控制模块。
  10. 如权利要求9所述的固态硬盘的控制芯片，其特征在于，该引导只读存储器模块、该读取队列管理模块、该写入缓冲管理模块及该数据输出输入控制模块是以一非挥发性只读存储器实现。
PCT/CN2015/079794 2015-05-26 2015-05-26 一种用于储存介质的数据管理方法及固态硬盘与控制芯片 WO2016187801A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/079794 WO2016187801A1 (zh) 2015-05-26 2015-05-26 一种用于储存介质的数据管理方法及固态硬盘与控制芯片

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/079794 WO2016187801A1 (zh) 2015-05-26 2015-05-26 一种用于储存介质的数据管理方法及固态硬盘与控制芯片

Publications (1)

Publication Number Publication Date
WO2016187801A1 true WO2016187801A1 (zh) 2016-12-01

Family

ID=57392305

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/079794 WO2016187801A1 (zh) 2015-05-26 2015-05-26 一种用于储存介质的数据管理方法及固态硬盘与控制芯片

Country Status (1)

Country Link
WO (1) WO2016187801A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110515547A (zh) * 2019-08-09 2019-11-29 浪潮电子信息产业股份有限公司 一种固态硬盘及计算机设备
CN112822386A (zh) * 2019-11-15 2021-05-18 合肥美亚光电技术股份有限公司 数据的采集方法及系统
CN113377179A (zh) * 2021-06-10 2021-09-10 深圳忆联信息系统有限公司 降低固态硬盘工作温度的方法、装置、计算机设备及介质
CN113918082A (zh) * 2020-07-08 2022-01-11 慧荣科技股份有限公司 计算机可读取存储介质、配置可靠命令的方法及装置
CN116610257A (zh) * 2023-04-27 2023-08-18 上海威固信息技术股份有限公司 一种高集成的固态硬盘功耗控制方法及固态硬盘

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102253905A (zh) * 2010-05-20 2011-11-23 智微科技股份有限公司 应用于储存装置的数据处理装置和数据存取系统与其相关方法
CN102541760A (zh) * 2012-01-04 2012-07-04 记忆科技(深圳)有限公司 基于固态硬盘的计算机系统
CN102866877A (zh) * 2011-07-08 2013-01-09 三星电子株式会社 存储器控制器及操作方法，及含存储器控制器的电子设备
US20140359192A1 (en) * 2011-05-31 2014-12-04 Micron Technology, Inc. Apparatus including buffer allocation management and related methods
CN104571940A (zh) * 2013-10-17 2015-04-29 财团法人交大思源基金会 储存装置及其相关系统
CN104571933A (zh) * 2013-10-18 2015-04-29 光宝科技股份有限公司 具固态储存元件的电子装置及其相关控制方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102253905A (zh) * 2010-05-20 2011-11-23 智微科技股份有限公司 应用于储存装置的数据处理装置和数据存取系统与其相关方法
US20140359192A1 (en) * 2011-05-31 2014-12-04 Micron Technology, Inc. Apparatus including buffer allocation management and related methods
CN102866877A (zh) * 2011-07-08 2013-01-09 三星电子株式会社 存储器控制器及操作方法，及含存储器控制器的电子设备
CN102541760A (zh) * 2012-01-04 2012-07-04 记忆科技(深圳)有限公司 基于固态硬盘的计算机系统
CN104571940A (zh) * 2013-10-17 2015-04-29 财团法人交大思源基金会 储存装置及其相关系统
CN104571933A (zh) * 2013-10-18 2015-04-29 光宝科技股份有限公司 具固态储存元件的电子装置及其相关控制方法

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110515547A (zh) * 2019-08-09 2019-11-29 浪潮电子信息产业股份有限公司 一种固态硬盘及计算机设备
CN112822386A (zh) * 2019-11-15 2021-05-18 合肥美亚光电技术股份有限公司 数据的采集方法及系统
CN112822386B (zh) * 2019-11-15 2022-04-29 合肥美亚光电技术股份有限公司 数据的采集方法及系统
CN113918082A (zh) * 2020-07-08 2022-01-11 慧荣科技股份有限公司 计算机可读取存储介质、配置可靠命令的方法及装置
CN113918082B (zh) * 2020-07-08 2024-03-08 慧荣科技股份有限公司 计算机可读取存储介质、配置可靠命令的方法及装置
CN113377179A (zh) * 2021-06-10 2021-09-10 深圳忆联信息系统有限公司 降低固态硬盘工作温度的方法、装置、计算机设备及介质
CN116610257A (zh) * 2023-04-27 2023-08-18 上海威固信息技术股份有限公司 一种高集成的固态硬盘功耗控制方法及固态硬盘
CN116610257B (zh) * 2023-04-27 2024-05-17 上海威固信息技术股份有限公司 一种高集成的固态硬盘功耗控制方法及固态硬盘

Similar Documents

Publication Publication Date Title
JP4829365B1 (ja) データ記憶装置及びデータ書き込み方法
TWI704487B (zh) 資料儲存設備及其操作方法
TWI506430B (zh) 映射資訊記錄方法、記憶體控制器與記憶體儲存裝置
TWI557561B (zh) 記憶體管理方法、記憶體控制電路單元與記憶體儲存裝置
WO2016187801A1 (zh) 一种用于储存介质的数据管理方法及固态硬盘与控制芯片
TWI486766B (zh) 資料處理方法、記憶體控制器與記憶體儲存裝置
US11269722B2 (en) Apparatus for diagnosing memory system and operating method thereof
TWI489466B (zh) 記憶體抹除方法、記憶體控制器與記憶體儲存裝置
KR20170012935A (ko) 데이터 저장 장치의 작동 방법과 상기 데이터 저장 장치를 포함하는 시스템의 작동 방법
TWI607309B (zh) 記憶體管理方法、記憶體控制電路單元與記憶體儲存裝置
US10579289B2 (en) Memory management method for configuring super physical units of rewritable non-volatile memory modules, memory control circuit unit and memory storage device
KR20160074025A (ko) 데이터 저장 장치의 동작 방법
TWI492054B (zh) 快閃記憶體的模擬方法與模擬器
TWI523030B (zh) 緩衝記憶體管理方法、記憶體控制器與記憶體儲存裝置
TWI591640B (zh) 記憶體管理方法、記憶體控制電路單元與記憶體儲存裝置
US20150278299A1 (en) External merge sort method and device, and distributed processing device for external merge sort
CN109558266A (zh) 主动错误更正失败处理方法
US9934045B1 (en) Embedded system boot from a storage device
US9817573B2 (en) Smart card management method, memory storage device and memory control circuit unit
US9652403B2 (en) Memory control unit and data storage device including the same
JP2007299249A (ja) Ｎａｎｄ型フラッシュメモリデバイス及びこれを利用したコンピューティングシステムの起動方法
TWI603335B (zh) 映射表載入方法、記憶體控制電路單元與記憶體儲存裝置
TWI503841B (zh) 寫入方法、記憶體控制器與記憶體儲存裝置
TWI417888B (zh) An embedded chip system, a method for burning a wafer, and a computer program product
CN102736985B (zh) 数据合并方法、控制器与储存装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15892882

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15892882

Country of ref document: EP

Kind code of ref document: A1