Method and system for establishing RAID in SSD

A disk and physical page technology, which is applied in the field of RAID formation inside SSD disks, can solve problems such as increased disk write amplification, increased DRAM resources, and reduced efficiency, achieving reduced impact on random performance, reduced DRAM resources, and improved efficiency Effect

Active Publication Date: 2016-05-11
HUNAN GOKE MICROELECTRONICS
View PDF6 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] 1) DRAM resources increase, cost increases, and efficiency decreases; SSD data reading mainly depends on the mapping table, and the RAID formation method based on the logical address of the host needs the mapping table to physically address the stripe verification data, so the mapping table It is necessary to add the RAID verification data addressing part on the basis of the original host capacity addressing space; for SSDs with external DRAM, it will increase the DRAM resources required by the system and increase the cost; for SSDs without external DRAM , due to the need to load the mapping table of RAID parity data from NandFlash, resulting in performance degradation
[0007] 2) The performance of disk random write is degraded: For the RAID formation method based on the logical address of the host, when the member data in the stripe is updated due to the random write service of the host, it is necessary to read the member data of the stripe and recalculate the RAID check value of the stripe , while reading stripe member data needs to occupy NAND Flash bandwidth, and recalculating stripe RAID checksums will take CPU scheduling time; in addition, in order to ensure data consistency, any data update of any member in the stripe will cause the entire stripe to Add mutex with included host logical address
Although the invention with the application number 201210481988.9 distinguishes the case of the stripe update to reduce the number of stripe members to be read, this method still inevitably needs to read the stripe member data
[0008] 3) Increased disk write amplification: For the RAID formation method based on the logical address of the host, the data update of any member in the stripe needs to update the RAID check value (recalculated and written to NAND Flash), which will inevitably lead to disk failure. Increased write amplification and increased NAND Flash wear

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and system for establishing RAID in SSD
  • Method and system for establishing RAID in SSD
  • Method and system for establishing RAID in SSD

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] This embodiment adopts the RAID5 stripe under the situation that there are few physical pages (10 physical pages) and no bad pages in SuperPage, and the specific implementation steps are as follows:

[0045] 1) After obtaining a blank SuperBlock, select a physical block to store the RAID verification data, and record the Die number of the physical block (Die0 in this example) in the block status table of the SuperBlock;

[0046] 2) When the disk executes the write command of the host, the received data is stored in the user data buffer, and the user data buffer is divided according to the NAND Flash physical page size. When the amount of user data received by a buffer reaches the physical page After the maximum amount of user data that can be stored (the physical page size minus the amount of error correction data required for user data), the data in this buffer will be written to Die1 of SuperPage0;

[0047] 3) Add user data buffer buffer0 to the buffer list of the RAI...

Embodiment 2

[0053] This embodiment adopts the RAID5 stripe under the situation that there are few physical pages (10 physical pages) and one bad page in SuperPage, and the specific implementation steps are as follows:

[0054] 1) After obtaining a blank SuperBlock, select a physical block to store the RAID verification data, and record the Die number of the physical block (Die0 in this example) in the block status table of the SuperBlock;

[0055] 2) When the disk executes the host's write command, it stores the received data in the user data buffer divided by the size of the physical page. When the amount of user data received by a buffer reaches the maximum value that can be stored in the physical page After the amount of user data (physical page size minus the amount of error correction data required for user data), the data in the user data buffer will be written to Die1 of SuperPage0;

[0056] 3) Add the user data buffer buffer0 to the buffer list of the RAID calculation task, and judg...

Embodiment 3

[0061] This embodiment adopts the RAID6 stripe adopted in SuperPage in order to improve the reliability of certain data (such as mapping tables), and the specific implementation steps are as follows:

[0062] 1) After obtaining the blank SuperBlock, select two physical blocks to store the RAID verification data, and record the Die numbers of the physical blocks (Die0 and Die1 in this example) in the block status table of the SuperBlock;

[0063] 2) When the disk executes the write command of the host, the received data is stored in the user data buffer, and the user data buffer is divided according to the NAND Flash physical page size. When the amount of user data received by a buffer reaches the physical page After the maximum amount of user data that can be stored (the physical page size minus the amount of error correction data required for user data), the data in the user data buffer will be written to Die2 of SuperPage0;

[0064] 3) Add the user data buffer buffer0 to the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a method for establishing an RAID (Redundant Arrays of Independent Disks) in an SSD (Solid State Drive). The method mainly comprises the following steps: abstracting Nand Flash Blocks into Super Blocks, abstracting Pages into Super Pages, and performing RAID establishment in the Super Pages by taking the Super Pages as RAID stripe units; making the Super Blocks equivalent to the Nand Flash Blocks by an FTL conversion layer: performing writing in sequence from small to big according to Super Page numbers during writing operation and erasing the Nand Flash Blocks in the Super Blocks at the same time during erasing operation; making the Super Pages equivalent to Nand Flash Pages by the FTL conversion layer: performing writing in sequence on the Nand Flash Pages in the Super Pages from small to big according to die numbers during writing operation; in addition, recording bad dies and RAID dies in the Super Blocks by the FTL conversion layer, and skipping dies where bad pages are when data is written for RAID establishment; and after all user data is written in RAID stripes, determining final RAID check data of the stripes and writing the RAID check data into the RAID dies of the Super Pages. According to the method, required DRAM resources are reduced, the cost is low, and the efficiency is high.

Description

technical field [0001] The invention relates to a method and system for constructing an internal RAID of an SSD disk. Background technique [0002] SSD (Solid State Drive) is a data storage device that uses a solid-state electronic memory chip array as a storage medium, controls the chip to read and write data, and manages the storage medium. Due to its fast read and write speed, low power consumption, no noise, Good shock resistance and other characteristics are widely used in consumer electronics, data centers, military, automotive and other fields, showing a trend of replacing mechanical hard drives. [0003] As the storage medium used by the current mainstream SSD, NAND Flash will cause bit flips in the stored data due to particle wear (PECycle), data storage time (Data Retention), high temperature environment and other reasons. Therefore, an error correction algorithm is designed inside the SSD controller. When the number of flipped bits is small, error correction algo...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F3/06
CPCG06F3/0607G06F3/0644G06F3/0679
Inventor 陈意马翼田达海彭鹏杨万云姜黎向平周士兵
Owner HUNAN GOKE MICROELECTRONICS
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap