132results about How to "Reduce write amplification" patented technology

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application. A page is associated with a set of primary ECC codewords, and a page stripe is associated with a set of secondary codewords and primary over secondary parity (PoSP) ECC codewords. Two or more page stripes can form a page grid, wherein the page grid is associated with a group of tertiary ECC codewords, wherein the last page stripe of the page grid has a reduced payload capacity.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application. Predefined gears correspond to different predefined ECC schemes. Based on an observed bit error rate, a gear from a set of predefined gears is selected for use for a particular region of memory. Each gear of the set of predefined gears includes a lower-latency ECC decode option and one or more higher-latency ECC decode options.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application.

Disclosed embodiments are directed to systems and methods for dynamic overprovisioning for data storage systems. In one embodiment, a data storage system can reserve a portion of memory, such as non-volatile solid-state memory, for overprovisioning. Depending on various overprovisioning factors, recovered storage space due to compressing user data can be allocated for storing user data and/or overprovisioning. Utilizing the disclosed dynamic overprovisioning systems and methods can result is more efficient utilization of cache memory, reduction of write amplification, increase in a cache hit rate, and the like. Improved data storage system performance and increased endurance and longevity can thereby be attained.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application.

The invention discloses a caching method and a caching device. The caching method comprises the steps of determining a to-be-deleted data block needing to be deleted from an NV (Non-volatile) storage medium in a caching space according to the access situation of data blocks; selecting and deleting the to-be-deleted data block from the NV storage medium according to a preset data recovery condition. According to the caching method and the caching device, which are disclosed by the invention, the writing times of an NV storage device which is unbalanced in writing and poorer in duration can be reduced, and the service life of the NV storage device can be prolonged.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application. Both primary parity symbols for primary codewords and secondary parity symbols for secondary codewords are generated. The secondary parity symbols are spread out across each page of a group of pages.

The invention discloses a method for selecting a garbage collection target block, and a solid state disk. The method is characterized in that the recovery factor score of each block is calculated according to the proportion and the erasing frequencies of effective pages in each block, and the block with a lowest score is selected as a recovery target block. The effective page ratio and the erasingfrequency weighting score value of each block on the SSD (Solid State Disk) are calculated to comprehensively consider an influence on the read-write performance and the service life of the SSD as far as possible, and the block with a low effective page proportion and few erasing frequencies is selected as the target recovery block of a GC. A wear-leveling thought is used for selecting an optimalblock as a recovery block, block erasing frequencies in the SSD are reduced, data migration frequencies are lowered, write amplification is reduced, and the service life of the SSD is improved.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application.

Disclosed is a method for writing data into a solid-state storage device. The method comprises: receiving a data writing request; based on the data writing request, determining that the data writing request is a first type, a second type or a third type; when the data writing request is the first type, writing data corresponding to the data writing request into a first storage region of the solid-state storage device; when the data writing request is the second type, writing the data corresponding to the data writing request into a second storage region of the solid-state storage device; and when the data writing request is the third type, writing the data corresponding to the data writing request into a third storage region of the solid-state storage device, wherein the first type indicates that the data written according to the data writing request is frequently updated, the second type indicates that the data written according to the data writing request is rarely updated, and the third type indicates that the updating property of the data written according to the data writing request is different from the indication of the first type or the second type. Furthermore, disclosed are a system for writing the data into the solid-state storage device and the storage device.

The invention provides a massive data storage method simultaneously applicable to disk and solid state disk reading and writing features. Full sequencing of records in each block is changed into partial sequencing, a Bloom filter is added to the tail portion of each block, a Log-Structured Append-Tree is created, when the quantity of data stored in each block in the tree reaches a threshold and data in the block is directly added to corresponding child blocks, the data of the child blocks is composed of multiple collating sequences rather than full sequencing is achieved in the blocks in a merging sorting mode; each block in the tree stores one Bloom filter. According to the method, on the condition that no other properties are sacrificed, write amplification is greatly reduced, and the random writing efficiency is greatly improved. Besides, the service life of a solid state disk is better protected and prolonged. In read and write mixed scenes, the random read property is also enhanced, and the method has important market value.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application.

The invention is applicable to the technical field of computer data storage and provides a flash memory array management method and device. The flash memory array management method comprises the steps of dividing a flash memory array into n physical working groups, wherein the same pages of all flash memories in the same physical working group are combined to form a physical combined page, and the same blocks of all flash memories in the same physical working group are combined to form a physical combined block; dividing a logic space into a plurality of logic pages according to the size of the physical combined page, dividing the multiple obtained logic pages into n logic working groups corresponding to the physical working groups, performing re-addressing according to logic addresses of the logic pages of the same logic working group to obtain logic combined pages; managing the logic combined pages in the logic working groups and the physical combined page in the corresponding physical working groups in a dynamic mapping mode so as to achieve reading and writing of the flash memory array. By means of the flash memory array management method and device, the storage space occupied by a mapping table is decreased, and write-in and amplification caused by block mapping are decreased.

The invention discloses a method and a system for reducing garbage collection and write amplification of a key-value separation storage system, and belongs to the field of computer storage, the systemcomprises a memory, an NVM and an SSD; and a first storage area and a second storage area in the NVM manage data according to an LSM-tree and a memory pool respectively; the method comprises the steps of judging whether the size of a to-be-inserted key value pair is greater than or equal to a first threshold value or not, if so, inserting the key value pair into the second storage area, and theninserting a key and a key value address into LSM-tree; otherwise, directly inserting the key value pair into LSM-tree; when the key value pair in the memory pool is deleted, marking the correspondingstorage unit as an idle memory unit; when the data volume in the second storage area is higher than a second threshold value, brushing down the key value pair with the least frequent access to the SSD, and updating the storage address of the key value pair; and setting a flag bit segment on the NVM, and identifying the validity of the key value pair of the data segment on the SSD. Garbage collection and write amplification of the key value storage system can be reduced, and the overall performance of the key value separation storage system is improved.

A novel symmetrical band is disclosed, which may be used in connection with shingled magnetic recording (SMR) in order to reduce write amplification (read-modify-write). Depending on the embodiment, overlapping data tracks diverge from, or converge to the center of each symmetrical band. Associated guard regions may be located at the center, or at the band boundaries, and are shared such that the excess width of a write element is caught by the guard regions from both sides. A symmetrical band may reduce the maximum write amplification by more than half. A hard disk controller may maintain the number of taken or empty tracks on both sides of each symmetrical band substantially equal at every fill level.

The invention discloses a read-write performance optimization method of a data storage device. The method comprise the steps that on the condition that the current total blank block number is smaller than a first preset threshold value and a first average erasing frequency is larger than a second average erasing frequency, screening processing is conducted on all current blank physical blocks, and the screened blank physical block is determined as a current static garbage collection block; screening processing is conducted on all current full user physical blocks in which user input data is fully stored by taking the erasing frequency as a screening basis, and the screened full user physical block is determined as a current physical block to be subjected to static collection; valid data in the current physical block to be subjected to static collection is written in the current static garbage collection block, and then emptying processing is conducted on the current physical block to be subjected to static collection. The method is beneficial for decreasing the write-in amplification rate of the data storage device, and the read-write performance of the data storage device is improved. In addition, the invention further discloses the data storage device and a read-write performance optimization system thereof.

The invention discloses an SSD performance acceleration method. During writing, a mode identification module is introduced; and by converting all-zero user data writing into mapping table writing, write amplification is greatly reduced. In a readout process, an all-zero read buffer region is formatted in a memory in advance. When a host reads previous all-zero written data, the data is determinedas no-map data by querying a mapping table, so that the data is returned to the host directly from the formatted read buffer region. The data does not need to be loaded from NAND, so that a response delay of a command can be greatly improved.

Apparatus and methods provide relatively low uncorrectable bit error rates, low write amplification, long life, fast and efficient retrieval, and efficient storage density such that a solid-state drive (SSD) can be implemented using relatively inexpensive MLC Flash for an enterprise storage application. Data is stored in page stripes. The page stripes can have varying amounts of payload capacity based on selected error correction code strength. Allocation blocks can be divided into journaling cells, correspond to minimum units of data for which a journaling engine or flash translation layer has a logical-to-physical mapping.

The invention discloses a method for reducing write amplification of an NVMe solid state disk. The method is characterized in that the solid state disk is divided into two or more data cold and hot intervals according to the size of access frequency information, and data falling in the same data cold and hot interval is written in a same block; and the access frequency information is subjected tostatistics by a host and written in an access frequency information field of an LBA interval in an NVMe protocol, and the solid state disk reads the access frequency information field carried in eachcommand when receiving a write command of the host. By combining the characteristics of the NVMe protocol with the characteristics of a Nand Flash storage medium the manner of writing data in blocks is innovated, the new technology greatly reduces the triggering frequency of a garbage collection mechanism, namely, the write amplification of the SSD is reduced, the read-write performance of the SSDis improved, and the service life of the SSD is prolonged.

The invention relates to a cold and hot data separation method and device, computer equipment and a storage medium. The method comprises the steps of adopting a file system to determine the cold and hot attributes of to-be-written data; if the to-be-written data is cold data, writing the to-be-written data into a first data block of a memory; if the to-be-written data is hot data, writing the to-be-written data into a second data block of the memory, wherein the first data block and the second data block are physically isolated. According to the method, the cold and hot attributes of the to-be-written data can be accurately determined by utilizing the file system, so that the cold and hot data separation accuracy is improved, physical isolation of the cold and hot data is realized, the influence of the hot data on the cold data during reading or writing operation is further avoided, and the writing amplification of the cold data is greatly reduced.