Garbage collection method based on block grading and nonvolatile storage equipment

Abstract

The embodiment of the invention relates to the field of storage equipment application, and discloses a garbage collection method based on block grading and nonvolatile storage equipment. The garbage collection method based on block grading comprises: determining physical blocks to be recycled in the storage device; if the effective data page exists in the to-be-recycled physical block, obtaining the grade of the to-be-recycled physical block according to the number of times that the effective data page in the to-be-recycled physical block is moved; according to the grade of the to-be-recycledphysical block, moving the data of the effective data page of the to-be-recycled physical block to the physical block of the previous grade; and initializing the physical block to be recycled. By means of the block grading mode, in the garbage collection process, moving of effective data pages can be reduced, the service life of the solid state disk is prolonged, and meanwhile read-write errors inthe data moving process are avoided.

Description

technical field [0001] The invention relates to the application field of storage devices, in particular to a garbage collection method based on block classification and non-volatile storage devices. Background technique [0002] Solid State Drives (SSD) are hard disks made of solid state electronic memory chip arrays. Solid state drives include control units and storage units (FLASH memory chips and DRAM memory chips). In solid state drives, written The physical block of data cannot be rewritten directly, and new data can only be written through garbage collection first. [0003] In order to reduce the write amplification of SSD data, the main research direction is to separate hot and cold data through data grouping, and store them in different physical blocks (Blocks) when data is written. At present, the data life cycle of different logical block addresses (Logical Block Address, LBA) can be predicted through the statistical method, and then the physical blocks stored in ...

AI Technical Summary

Problems solved by technology

At present, the data life cycle of different logical block addresses (Logical Block Address, LBA) can be predicted through the statistical method, and then the physical blocks stored in the data are grouped according to the length of the life cycle. However, the statistics of the data life cycle and Prediction will bring relatively large hardware overhead, and the accuracy of prediction cannot meet product expectations at present.

Images