WO2019080621A1 - 一种数据非对称备份存储方法及存储器 - Google Patents

一种数据非对称备份存储方法及存储器

Info

Publication number
WO2019080621A1
WO2019080621A1 PCT/CN2018/101538 CN2018101538W WO2019080621A1 WO 2019080621 A1 WO2019080621 A1 WO 2019080621A1 CN 2018101538 W CN2018101538 W CN 2018101538W WO 2019080621 A1 WO2019080621 A1 WO 2019080621A1
Authority
WO
WIPO (PCT)
Prior art keywords
storage
data
memory
chip
important data
Prior art date
Application number
PCT/CN2018/101538
Other languages
English (en)
French (fr)
Inventor
林万才
李明伟
Original Assignee
江苏都万电子科技有限公司
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 江苏都万电子科技有限公司 filed Critical 江苏都万电子科技有限公司
Publication of WO2019080621A1 publication Critical patent/WO2019080621A1/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
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0614—Improving the reliability of storage systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14—Error detection or correction of the data by redundancy in operation
    • G06F11/1402—Saving, restoring, recovering or retrying
    • G06F11/1446—Point-in-time backing up or restoration of persistent data
    • G06F11/1448—Management of the data involved in backup or backup restore
    • 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
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0638—Organizing or formatting or addressing of data
    • 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
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/0671—In-line storage system
    • G06F3/0683—Plurality of storage devices
    • G06F3/0685—Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays

Definitions

  • the invention relates to a data storage technology, in particular to a data backup technology, in particular to a data asymmetric backup storage method and a memory using a heterogeneous memory chip to realize low-cost and high-performance data storage.
  • RAID is the abbreviation of English Redundant Array of Independent Disks. Chinese is abbreviated as redundant array of independent disks. RAID is a redundant array of multiple hard disks. Although RAID contains multiple hard drives, it appears as a separate large storage device under the operating system. There are three main benefits to using RAID technology in storage systems:
  • RAID The main purpose of the initial development of RAID was to save costs. At that time, the price of several small-capacity hard disks was lower than that of large-capacity hard disks. At present, the role of RAID in terms of cost savings is not obvious, but RAID can take full advantage of multiple hard disks, achieving speed and throughput far beyond any single hard disk. In addition to performance improvements, RAID also provides good fault tolerance, and can continue to work in the event of a problem with any hard drive, without being affected by the damaged hard drive.
  • RAID technology is divided into several different levels that provide different speeds, security, and price/performance. Choosing the appropriate RAID level based on the actual situation can meet the user's requirements for storage system availability, performance, and capacity. Commonly used RAID levels are as follows: NRAID, JOD, RAID0, RAID1, RAID0+1, RAID3, RAID5, etc. RAID5 and RAID (0+1) are often used today.
  • the above problems are uniformly considered, that is, the advantages of data mutual backup are preserved, and on the basis of the improvement, the backup physical carrier of the same capacity is tailored, and the data is integrated with the non-uniform important data.
  • the feature is to classify important data with general data, specify the storage location when writing, logically control the allocation to write to main memory and small-capacity backup memory, from the application layer, transparent to the user. The actual internal hard cost has been greatly reduced, and the usage rate has also been significantly improved.
  • the object of the present invention is to solve the problem that the storage cost of the isomorphic storage chip as the storage medium is high in the existing data storage, especially when the differential storage is used, the secondary storage utilization rate is not high, and the resource problem is caused, and the invention is used.
  • a data asymmetric backup storage method characterized in that it uses two heterogeneous memory chips as a storage unit together, and stores important data in two storage units of the heterogeneous memory chip: the heterogeneous memory chip It refers to the storage capacity of the memory chip used, the manufacturer, the production batch, the production process, the storage medium, and the interface method. It can increase the storage reliability of important data from 100ppm to 1ppm. Reliability refers to the annual failure rate.
  • a small-capacity memory chip is used as the sub-memory unit, and the memory chip is easily used as the main memory unit, and the capacity ratio of the two is at least 4 times or more.
  • the invention adopts logic control technology, FLASH mass storage technology, solid state storage technology, universal USB, SATA, eMMC bus technology, and encapsulates data backup into an asymmetric storage unit, which is used for distinguishing important data and specifying important data storage.
  • Space when the user needs to write important data, write with the specified address, write the space of the overlapping part of the physical sector number of the primary and secondary storage chips, and the logical control unit (LCU) automatically recognizes the primary and secondary storage after identifying the command.
  • LCU logical control unit
  • the space writes the same data, so that the validity of the data can be compared and fault-tolerant, and the non-important data can be written only to the sector of the main memory chip, thereby limiting the non-essential data occupying the sub-memory space, thereby reducing the storage cost.
  • the user When the user needs to write important data, it can write with address and write space within 16383 sectors. After the logic control unit (LCU) recognizes this command, it automatically writes the space corresponding to the primary and secondary memories to the same data. Therefore, the validity of the data can be compared and fault-tolerant, and the non-important data can be written to sectors other than 16383, thereby limiting the non-essential data occupying the sub-memory space, thereby reducing the storage cost.
  • LCU logic control unit
  • a data asymmetric backup memory characterized in that it is mainly composed of a secondary storage chip, a main storage chip and a logic control unit, wherein the secondary storage chip and the primary storage chip are heterogeneous structure storage chips, and the logic control unit determines The important data is separately written into the storage space of the secondary storage chip and the important data storage space of the primary storage chip, and the logical control unit determines that the non-essential data is written into the non-essential data area of the primary storage unit.
  • the heterogeneous structure refers to that the capacity of the main and sub-memory chips is different, the manufacturer is different, and the storage medium is different, which can improve the storage reliability of important data from 100 ppm to 1 ppm.
  • the data asymmetric backup memory of the present invention is characterized in that it is applied to an on-vehicle memory.
  • the present invention overcomes the shortcomings of the existing backup must have equal capacity backup, especially in the case of high storage cell cost at the current stage, the cost of the single storage space becomes very expensive, and the modern society has higher and higher requirements for data security.
  • the demand for data fault tolerance also puts higher requirements, and asymmetric backup storage effectively solves this existing contradiction!
  • the important data such as driving action, automatic driving state, automatic control command, etc.
  • Other data such as engine, emissions, environmental audio and video are general data.
  • the asymmetric (heterogeneous) RIAD1 redundant storage control method can be reduced from the original 100% increase to only about 4%, and the cost reduction and efficiency improvement effect is very obvious.
  • Increase from 100ppm to 1ppm a 100-fold increase.
  • FIG. 1 is a schematic structural view of a chip memory of the present invention.
  • a data asymmetric backup storage method which uses two heterogeneous storage chips as a storage unit and stores important data in two storage units of a heterogeneous storage chip: the heterogeneous storage chip refers to storage
  • the storage capacity of the chip, the manufacturer, the production batch, the production process, the storage medium, and the interface method can increase the storage reliability of important data from 100ppm to 1ppm.
  • the reliability here refers to Is the annual failure rate. .
  • two small-capacity memory chips generally use a small-capacity memory chip as a secondary storage unit, and a large memory chip is used as a primary storage unit, and the capacity ratio of the two is at least 4 times or more, as shown in FIG. 1 .
  • the storage method of the invention adopts logic control technology, FLASH mass storage technology, main structure USB-OTG technology, solid state storage technology, universal USB, SATA, eMMC bus technology, and encapsulates data backup and storage into an asymmetric storage unit. It is used to store the difference between important data and specify the storage space. When the user needs to write important data, it writes with the specified address and writes the space of the overlapping part of the physical sector number of the primary and secondary storage chips.
  • the logical control unit (LCU) recognizes After this command, the space corresponding to the primary and secondary memories is automatically written to the same data, so that the validity of the data can be compared and fault-tolerant, and the non-important data can be written only to the sector of the primary storage chip, thereby limiting the non- Important data occupies the secondary storage space, which reduces storage costs.
  • the user can write with address, for example, write space within 16383 sectors.
  • the logic control unit (LCU) recognizes this command, it automatically writes the space corresponding to the primary and secondary memories to the same space.
  • the data so that the validity of the data can be compared, fault tolerance, corresponding non-important data can be written to sectors other than 16383, thereby limiting non-critical data occupying the sub-memory space, thereby reducing storage costs.
  • a data asymmetric backup memory which is mainly composed of a secondary storage chip, a main storage chip and a logic control unit, wherein the secondary storage chip and the primary storage chip are heterogeneous storage chips, and the logic control unit determines important data by judging The logic control unit writes the non-essential data into the non-essential data area of the main storage unit by determining to write the storage space of the secondary storage chip and the important data storage space of the primary storage chip, respectively.
  • the heterogeneous structure refers to the difference in capacity of the main and auxiliary memory chips, different manufacturers, production processes, production batches, storage media, etc., and can increase the storage reliability of important data from 100 ppm to 1 ppm.
  • the storage technology in Figure 1 uses logic control technology, FLASH mass storage technology, solid-state storage technology, general USB, SATA, eMMC and other bus technologies to encapsulate data backup and storage into an asymmetric storage unit, when users need to write
  • the specified address is written and written into the space of the overlap of the physical sector number of the primary and secondary storage chips.
  • the logical control unit (LCU) recognizes this command, the space corresponding to the primary and secondary memories is automatically written to the same data.
  • the validity of the data can be compared and fault-tolerant, and the non-important data can be written only to the sector of the main memory chip, thereby limiting the non-essential data occupying the sub-memory space, thereby reducing the storage cost, especially corresponding to the upper 1T capacity storage.
  • Storage costs are especially important in the required work environment.
  • the user When the user needs to write important data, it can be written with an address. For example, the space within 16383 sectors can be written. After the logic control unit (LCU) recognizes this command, the space corresponding to the primary and secondary memories is automatically written. The same data, so that the validity of the data can be compared, fault tolerance, corresponding non-important data can be written to sectors other than 16383, thereby limiting non-essential data occupying the sub-memory space, thereby reducing storage costs.
  • LCU logic control unit
  • the important data (such as driving action, automatic driving state, automatic control command, etc.) of the intelligent car and the future self-driving car account for about 1.5%
  • other Data such as engine, emissions, environmental audio and video are general data.
  • the asymmetric RIAD1 redundant memory control method is used to store important data in the important data area of 8G sub-memory and 256G main memory at the same time, and the cost of 8G memory is only 4% of 256G memory, which is equivalent to only about 4% increase. Cost, the same effect of the two 256G memories when the original symmetric storage is achieved, and the reliability index of important data can be increased from 100ppm to 1ppm, which is 100 times higher.
  • 8G memory and 256G memory are preferably stored in different manufacturers and different storage media to achieve true heterogeneity.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Quality & Reliability (AREA)
  • Techniques For Improving Reliability Of Storages (AREA)

Abstract

一种数据非对称备份存储方法及存储器，其特征是它采用两个异构存储芯片共同作为存储单元，并将重要数据同时存储在二个异构存储芯片的存储单元中：所述的异构存储芯片是指所使用的存储芯片的容量不同、生产厂家不同、存储介质不同，它能将重要数据的可靠性从100ppm提高到1ppm。从而克服了现有备份必须等容量备份的缺点，具有附加成本低，数据存储安全可靠的优点。

Description

一种数据非对称备份存储方法及存储器 技术领域
本发明涉及一种数据存储技术，尤其是一种数据备份技术，具体地说是一种使用异构存储芯片来实现低成本、高可性数据存储的数据非对称备份存储方法及存储器。
背景技术
RAID是英文Redundant Array of Independent Disks的缩写，中文简称为独立磁盘冗余阵列。RAID就是一种由多块硬盘构成的冗余阵列。虽然RAID包含多块硬盘，但是在操作系统下是作为一个独立的大型存储设备出现。利用RAID技术于存储系统的好处主要有以下三种：
1.通过把多个磁盘组织在一起作为一个逻辑卷提供磁盘跨越功能；
2.通过把数据分成多个数据块(Block)并行写入/读出多个磁盘以提高访问磁盘的速度；
3.通过镜像或校验操作提供容错能力。
最初开发RAID的主要目的是节省成本，当时几块小容量硬盘的价格总和要低于大容量的硬盘。目前来看RAID在节省成本方面的作用并不明显，但是RAID可以充分发挥出多块硬盘的优势，实现远远超出任何一块单独硬盘的速度和吞吐量。除了性能上的提高之外，RAID还可以提供良好的容错能力，在任何一块硬盘出现问题的情况下都可以继续工作，不会受到损坏硬盘的影响。
RAID技术分为几种不同的等级，分别可以提供不同的速度，安全性和性价比。根据实际情况选择适当的RAID级别可以满足用户对存储系统可用性、性能和容量的要求。常用的RAID级别有以下几种：NRAID，JOD，RAID0，RAID1，RAID0+1，RAID3，RAID5等。当前经常使用的是RAID5和RAID(0+1)。
在闪存技术出现后，RAID技术和概念扩充到了芯片固态存储上，一般的大容量U盘和SSD都是由多个小容量存储芯片联合起来的RAID0技术。
实际运作中，由于存储芯片IC本身的可靠性(一般MTBF200万小时，芯片失效率为100ppm)，对于存储重要数据的数据存储器，多采用RAID1冗余存储技术，以对称容量的大小进行双份存储以达到互为备份，达到容错和提高可靠性的目的，而实际操作过程中，针对一些特定应用(比如车载存储)并非所有数据均为重要数据，在存储时数据之间没有区别数据重要性等级，以至于所有数据都进行了双备份，从时效性及存储有效性上来讲是一种浪费，同理作为存储，大容量冗余，单体物理存储空间成本也相应上升。
现有发明技术中，将以上的问题进行了统一考虑，即保存了数据互为备份的优点，并在此基础上进行改良，将同等容量的备份物理载体进行裁剪，结合数据全集非均重要数据的特点，将重要数据，与一般数据进行分类，在写入时，指定其存储位置，逻辑控制进行分配写入主存储器以及小容量的备份存储器，从应用层来讲，对用户为透明的操作，实际内部硬性成本大大缩减，使用率也得到了明显的提升。
发明内容
本发明的目的是针对现有的数据存储中大多采用同构存储芯片作为存储介质存在的存储成本高，尤其是差别化存储时，副存储器利用率不高，造成资源的问题，发明一种使用异构存储芯片进行数据存储的数据非对称备份存储方法及存储器
本发明的技术方案之一是：
一种数据非对称备份存储方法，其特征是它采用两个异构存储芯片共同作为存储单元，并将重要数据同时存储在二个异构存储芯片的存储单元中：所述的异构存储芯片是指所使用的存储芯片的存储容量、生产厂家、生产批次、生产工艺、存储介质、接口方式其中一项或多项不同，它能将重要数据的存储可靠性从100ppm提高到1ppm,此处的可靠性指的是年故障率。
使用小容量存储芯片作为副存储单元，大容易存储芯片作为主存储单元，且两者的容量比至少在4倍或以上。
本发明采用逻辑控制技术、FLASH海量存储技术、固态存储技术，通用USB、SATA、eMMC总线技术，将数据备份封装成一种非对称的存储单元，用于对重要数据的区别存储，指定重要数据存储空间；当用户需要写入重要数据时，带指定地址写入，写入主副存储芯片物理扇区号重叠部分的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以只写入主存储芯片的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。
当用户需要写入重要数据时，可带地址写入，写入16383扇区以内的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以写入16383以外的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。
本发明的技术方案之二是：
一种数据非对称备份存储器，其特征是它主要由副存储芯片、主存储芯片和逻辑控制单元组成，所述的副存储芯片和主存储芯片为异构结构的存储芯片，逻辑控制单元通过判断将重要数据分别写入副存储芯片的存储空间和主存储芯片的重要数据存储空间中，逻辑控制单元通过判断将非重要数据写入主存储单元的非重要数据区中。
所述的异构结构是指主副存储芯片的容量不同、生产厂家不同、存储介质不同，它能将重要数据的存储可靠性从100ppm提高到1ppm。
本明的数据非对称备份存储器，其特征是它被应用于车载存储器中。
本发明的有益效果：
本发现克服了现有备份必须等容量备份的缺点，特别是现阶段存储晶元成本高的情况下，单体存储空间成本变的很昂贵，而现代社会对数据安全的要求越来越高，数据容错的需求也提上了更高的要求，非对称的备份存储有效的解决了这一现有矛盾！
以8G+256G的非对称存储在汽车上的应用为例，智能汽车及未来自动驾驶汽车在运行过程中重要数据(如驾驶动作、自动驾驶状态、自动控制指令等)约占1.5％左右，，其他如发动机、排放、环境音视频等数据为一般数据。采用非对称(异构)RIAD1冗余存储控制方式，在成本上可由原来的增加100％减少到仅需增加约4％，降本增效成效十分明显，在重要数据的 可靠性指标上，可以由100ppm提高到1ppm，提高100倍。
附图说明
图1是本发明的芯片存储器的结构示意图。
具体实施方式
下面结合附图和实施例对本发明作进一步的说明。
实施例一。
一种数据非对称备份存储方法，它采用两个异构存储芯片共同作为存储单元，并将重要数据同时存储在二个异构存储芯片的存储单元中：所述的异构存储芯片是指存储芯片的存储容量、生产厂家、生产批次、生产工艺、存储介质、接口方式其中一项或多项不同，它能将重要数据的存储可靠性从100ppm提高到1ppm，此处的可靠性指的是年故障率。。
具体实施时二个异构存储芯片中一般用小容量存储芯片作为副存储单元，大容易存储芯片作为主存储单元，且两者的容量比至少在4倍或以上为最佳，如图1所示。本发明的存储方法采用了逻辑控制技术、FLASH海量存储技术、主结构USB-OTG技术、固态存储技术，通用USB、SATA、eMMC总线技术，将数据备份存存封装成一种非对称的存储单元，用于对重要数据的区别存储，指定存储空间；当用户需要写入重要数据时，带指定地址写入，写入主副存储芯片物理扇区号重叠部分的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以只写入主存储芯片的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。当用户需要写入重要数据时，可带地址写入，如：写入16383扇区以内的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以写入16383以外的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。
实施例二。
如图1所示。
一种数据非对称备份存储器，它主要由副存储芯片、主存储芯片和逻辑控制单元组成，所述的副存储芯片和主存储芯片为异构结构的存储芯片，逻辑控制单元通过判断将重要数据分别写入副存储芯片的存储空间和主存储芯片的重要数据存储空间中，逻辑控制单元通过判断将非重要数据写入主存储单元的非重要数据区中。所述的异构结构是指主副存储芯片的容量不同、生产厂家不同、生产工艺、生产批次、存储介质等不同，它能将重要数据的存储可靠性从100ppm提高到1ppm。图1中的存储技术采用了逻辑控制技术、FLASH海量存储技术、固态存储技术，通用USB、SATA、eMMC等总线技术，将数据备份存存封装成一种非对称的存储单元，当用户需要写入重要数据时，带指定地址写入，写入主副存储芯片物理扇区号重叠部分的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以只写入主存储芯片的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本，特别对应上1T容量存储要求的工作环境，存储成本尤为重要。当用户需要写入重要数据时，可带地址写入，比如：可写入16383扇区以内的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、 副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以写入16383以外的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。
实施例三。
以8G+256G的非对称存储在汽车上的应用为例，智能汽车及未来自动驾驶汽车在运行过程中重要数据(如驾驶动作、自动驾驶状态、自动控制指令等)约占1.5％左右，其他如发动机、排放、环境音视频等数据为一般数据。采用非对称RIAD1冗余存储控制方式，将重要数据同时在8G副存储器和256G主存储器的重要数据区中存储，而8G存储器的成本仅为256G存储器的4％，相当于只要增加约4％的成本，即可实现原来对称存储时两个256G存储器相同的效果，同时在重要数据的可靠性指标上，可以由100ppm提高到1ppm，提高100倍。而且8G存储器和256G存储器最好采用不同生产厂家、不同存储介质的存储器以实现真正意义上的异构。
本发明未涉及部分均与现有技术相同或可采用现有技术加以实现。

Claims (6)

  1. 一种数据非对称备份存储方法，其特征是它采用两个异构存储芯片共同作为存储单元，并将重要数据同时存储在二个异构存储芯片的存储单元中：所述的异构存储芯片是指所使用的存储芯片的存储容量、生产厂家、生产批次、生产工艺、存储介质、接口方式其中一项或多项不同。
  2. 根据权利要求1所述的方法，其特征是使用小容量存储芯片作为副存储单元，大容易存储芯片作为主存储单元，且两者的容量比至少在4倍或以上。
  3. 根据权利要求1所述的方法，其特征是它采用逻辑控制技术、FLASH海量存储技术、固态存储技术，通用USB、SATA、eMMC总线技术，将数据备份封装成一种非对称的存储单元，用于对重要数据的区别存储，指定重要数据存储空间；当用户需要写入重要数据时，带指定地址写入，写入主副存储芯片物理扇区号重叠部分的空间，逻辑控制单元(LCU)识别到此命令后，自动将主、副存储器对应的空间写入相同的数据，从而使数据的有效性可以比对，进行容错，对应非重要数据可以只写入主存储芯片的扇区，从而限制非重要数据占用副存储器空间，从而降低存储成本。
  4. 一种数据非对称备份存储器，其特征是它主要由副存储芯片、主存储芯片和逻辑控制单元组成，所述的副存储芯片和主存储芯片为异构结构的存储芯片，逻辑控制单元通过判断将重要数据分别写入副存储芯片的存储空间和主存储芯片的重要数据存储空间中，逻辑控制单元通过判断将非重要数据写入主存储单元的非重要数据区中。
  5. 根据权利要求1所述的数据非对称备份存储器，其特征是所述的异构结构是指主副存储芯片的存储容量、生产厂家、生产批次、生产工艺、存储介质、接口方式其中一项或多项不同，它能将重要数据的存储可靠性从100ppm提高到1ppm。
  6. 一种权利要求5所述的数据非对称备份存储器，其特征是它被应用于车载存储器中。
PCT/CN2018/101538 2017-10-24 2018-08-21 一种数据非对称备份存储方法及存储器 WO2019080621A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711002549.4 2017-10-24
CN201711002549.4A CN107678697A (zh) 2017-10-24 2017-10-24 一种数据非对称备份存储方法及存储器

Publications (1)

Publication Number Publication Date
WO2019080621A1 true WO2019080621A1 (zh) 2019-05-02

Family

ID=61142130

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/101538 WO2019080621A1 (zh) 2017-10-24 2018-08-21 一种数据非对称备份存储方法及存储器

Country Status (2)

Country Link
CN (1) CN107678697A (zh)
WO (1) WO2019080621A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107678697A (zh) * 2017-10-24 2018-02-09 江苏都万电子科技有限公司 一种数据非对称备份存储方法及存储器
CN111240576A (zh) * 2018-11-28 2020-06-05 瑞昱半导体股份有限公司 计算机系统、存储器管理方法与非暂态计算机可读介质
CN111766840B (zh) * 2020-06-09 2022-05-31 深圳创维－Rgb电子有限公司 一种生产母片的制作、实现方法及存储介质
CN114640688A (zh) * 2022-03-23 2022-06-17 中国兵器装备集团自动化研究所有限公司 一种异构双冗余设备运行状态数据存储装置及访问方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102333103A (zh) * 2010-07-12 2012-01-25 戴元顺 云存储系统
CN104778018A (zh) * 2015-04-23 2015-07-15 南京道熵信息技术有限公司 基于非对称混合型磁盘镜像的宽条带磁盘阵列及存储方法
CN105739920A (zh) * 2016-01-22 2016-07-06 深圳市瑞驰信息技术有限公司 一种自动化分层存储方法及服务器
CN107678697A (zh) * 2017-10-24 2018-02-09 江苏都万电子科技有限公司 一种数据非对称备份存储方法及存储器

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101847110A (zh) * 2009-03-27 2010-09-29 旺玖科技股份有限公司 数据存储系统及其备份方法
CN103455283B (zh) * 2013-08-19 2016-01-20 华中科技大学 一种混合存储系统
CN106897176A (zh) * 2015-12-17 2017-06-27 德阳九鼎智远知识产权运营有限公司 一种高效安全的计算机存储系统
CN106445740A (zh) * 2016-09-22 2017-02-22 深圳市先天海量信息技术有限公司 固态硬盘中nand闪存数据的控制方法及控制系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102333103A (zh) * 2010-07-12 2012-01-25 戴元顺 云存储系统
CN104778018A (zh) * 2015-04-23 2015-07-15 南京道熵信息技术有限公司 基于非对称混合型磁盘镜像的宽条带磁盘阵列及存储方法
CN105739920A (zh) * 2016-01-22 2016-07-06 深圳市瑞驰信息技术有限公司 一种自动化分层存储方法及服务器
CN107678697A (zh) * 2017-10-24 2018-02-09 江苏都万电子科技有限公司 一种数据非对称备份存储方法及存储器

Also Published As

Publication number Publication date
CN107678697A (zh) 2018-02-09

Similar Documents

Publication Publication Date Title
WO2019080621A1 (zh) 一种数据非对称备份存储方法及存储器
US8930671B2 (en) Logical address offset in response to detecting a memory formatting operation
US8984253B2 (en) Transaction log recovery
EP2483782B1 (en) Power interrupt management
US9927999B1 (en) Trim management in solid state drives
WO2016086899A1 (zh) 一种存储设备元数据管理处理方法及系统
US20090327603A1 (en) System including solid state drives paired with hard disk drives in a RAID 1 configuration and a method for providing/implementing said system
US20140156966A1 (en) Storage control system with data management mechanism of parity and method of operation thereof
US11599304B2 (en) Data aggregation in ZNS drive
KR102585883B1 (ko) 메모리 시스템의 동작 방법 및 메모리 시스템
US20150234595A1 (en) Storage device
CN103514095A (zh) 一种数据库写入ssd 的方法和系统
WO2021242317A1 (en) Zns parity swapping to dram
CN103729149A (zh) 一种存储数据的方法
CN103092728A (zh) 一种非易失性内存的磨损错误恢复方法和装置
US10031689B2 (en) Stream management for storage devices
US20090300282A1 (en) Redundant array of independent disks write recovery system
CN102402396B (zh) 复合式储存装置及其复合式储存媒体控制器与编址方法
CN115857790A (zh) 一种存储设备、数据存储方法及存储系统
CN102063271B (zh) 一种磁盘外置Cache基于状态机的写回方法
US20170091048A1 (en) Storage method and device for solid-state drive
JP5908106B2 (ja) 妥当性マスクを記憶する装置および方法ならびに操作装置
CN115525605A (zh) 一种文件系统的传输系统、方法
CN111475112B (zh) 一种提升Oracle数据库性能的装置及数据读写方法
CN103064803B (zh) 一种NAND Flash存储设备的数据读写方法和装置

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: 18870902

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: 18870902

Country of ref document: EP

Kind code of ref document: A1