# US 2015/0106557 A1 · Virtual Memory Device (VMD) Application/Driver for Enhanced Flash Endurance

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2015/0106557 A1
- Document ID: US20150106557A1
- Status: Published application
- Kind: A1
- Application number: 14/575,943
- Filed: 2014-12-18
- Published: 2015-04-16
- Earliest claimed: 2008-06-18
- Assignees: Super Talent Technology, Corp.
- Applicants: Super Talent Technology Corp.
- Inventors: Frank Yu; Abraham C. Ma; Shimon Chen; Yi Syu Yan
- Web page: https://patentgrep.com/patent/US20150106557A1

## CPC

- G11C 11/5678: {using amorphous/crystalline phase transition storage elements}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G11C 13/0004: {comprising amorphous/crystalline phase transition cells}
- G11C 16/3431: {Circuits or methods to detect disturbed nonvolatile memory cells, e.g. which still read as programmed but with threshold less than the program verify threshold or read as erased but with threshold greater than the erase verify threshold, and to reverse the disturbance via a refreshing programming or erasing step}
- G06F 2212/7203: Temporary buffering, e.g. using volatile buffer or dedicated buffer blocks
- G06F 2212/7208: Multiple device management, e.g. distributing data over multiple flash devices
- G06F 2212/7211: Wear leveling
- G11C 29/765: {in solid state disks}
- G11C: STATIC STORES (semiconductor memory devices H10B)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A Virtual-Memory Device (VMD) driver and application execute on a host to increase endurance of flash memory attached to a Super Enhanced Endurance Device (SEED) or Solid-State Drive (SSD). Host accesses to flash are intercepted by the VMD driver using upper and lower-level filter drivers and categorized as data types of paging files, temporary files, meta-data, and user data files, using address ranges and file extensions read from meta-data tables. Paging files and temporary files are optionally written to flash. Full-page and partial-page data are grouped into multi-page meta-pages by data type before storage by the SSD. Ramdisks and caches for storing each data type in the host DRAM are managed and flushed to the SSD by the VMD driver. Write dates are stored for pages or blocks for management functions. A spare/swap area in DRAM reduces flash wear. Reference voltages are adjusted when error correction fails.

## Claims

### Claim 1

A method for refreshing a flash memory comprising:
obtaining a current date;
for a current block in a list of blocks in the flash memory:
(a) reading a block status table for an erase count of the current block and for a write date of the current block that indicates when the current block was last written;
calculating a first time delay as a difference of the write date and the current date;
comparing the first time delay to a retention time;
when the first time delay exceeds the retention time, moving all valid pages in the current block to empty pages in a spare block in a dynamic-random-access memory (DRAM);
when the spare block in the DRAM is full, moving data from the spare block in DRAM to a spare block in the flash memory;
erasing the current block;
selecting another block in the list of blocks of flash memory as the current block and repeating from (a) until all blocks in the list of blocks have been processed,
whereby flash blocks are refreshed.

### Claim 2

A method for refreshing a flash memory comprising:
obtaining a current date;
for a current page in a current block in a list of blocks in the flash memory:
(a) reading a block status table for an erase count of the current block;
(b) reading a page status table for a write date of the current page that indicates when the current page was last written;
calculating a first time delay as a difference of the write date and the current date;
comparing the first time delay to a retention time;
when the first time delay exceeds the retention time, moving the current page in the current block to an empty page in a spare block in a dynamic-random-access memory (DRAM), and marking the current page as garbage page;
when the spare block in the DRAM is full, moving data from the spare block in DRAM to a spare block in the flash memory;
selecting another page in the current block and repeating from (b) until all pages in the current block have been processed;
selecting another block in the list of blocks of flash memory as the current block and repeating from (a) until all blocks in the list of blocks have been processed,
whereby flash pages are refreshed.

### Claim 3 (depends on 2)

The method of claim 2 further comprising:
calculating the retention time as a function of the erase count of the current block.

### Claim 4

A Super Enhanced Endurance Device (SEED) Solid-State Drive (SSD) Endurance Translation Layer (ETL) method to increase endurance of a flash memory having a low specified erase-cycle lifetime comprising:
creating an ETL in a dynamic-random-access memory (DRAM) buffer that is controlled by a controller and using the ETL to provide temporary storage to reduce flash wear;
creating a spare/swap area in the DRAM buffer;
operating a controller to use the spare/swap area in the DRAM buffer to merge valid data in a flash memory with new data to generate combined data;
when over-writing a full page or a partial page of an existing block in the flash memory, writing new data to a spare block in the DRAM buffer related to the existing block in the flash memory, and when the spare block is needed by additional new data for a different block in the flash memory, writing a combination of the spare block with the new data and data from the existing block in the flash memory into a swap block in the flash memory and marking a page status as a garbage page for all pages of the existing block in the flash memory; and
using a backup power supply to power the DRAM buffer and the flash memory and the controller when power is lost, the backup power supply having a sufficient capacity for the controller to copy desired data in the ETL to the flash memory,
whereby a spare/swap function is performed by the controller using the DRAM buffer rather than the flash memory.

### Claim 5

A Super Enhanced Endurance Device (SEED) comprising:
a host interface for receiving host reads and host writes from a host;
a SEED dynamic-random-access memory (DRAM) buffer for storing data;
a controller for controlling access to the flash memory and to the DRAM buffer in response to host reads and host writes received by the host interface, the controller writing host data to the DRAM buffer;
an SEED Solid-State Drive (SSD) Endurance Translation Layer (ETL) implemented in the DRAM buffer and controlled by the controller that uses the ETL to provide temporary storage to reduce flash wear;
a data write cache stored in the DRAM buffer and managed by the controller;
a flash interface for erasing blocks and writing pages in the flash memory, wherein a block is a multiple of pages in size;
a flash memory for storing grouped meta pages from the SSD DRAM buffer;
using a backup power supply to power the DRAM buffer and the flash memory and the controller when power is lost, the backup power supply having a sufficient capacity for the controller to copy desired data in the ETL to the flash memory.

### Claim 6 (depends on 5)

The SEED of claim 5 further comprising:
a spare/swap area in the DRAM buffer;
a controller that uses the spare/swap area in the DRAM buffer to merge valid data in the flash memory with new data to generate combined data when over-writing a full page or a partial page of an existing block in the flash memory by writing new data to a spare block in the DRAM buffer related to the existing block in the flash memory, and when the spare block is needed by additional new data for a different block in the flash memory, writing a combination of the spare block with the new data and data from the existing block in the flash memory into a swap block in the flash memory and marking a page status for pages of the existing block in the flash memory as a garbage page.

### Claim 7 (depends on 5)

The SEED of claim 5 further comprising:
a bad page manager for tracking bad pages of the flash memory,
the bad page manager for performing a routine comprising:
when an erased block in the flash memory has been erased, increasing an erase count in an erase count table for the erased block;
reading a page status table for the erased block, the page status table having a plurality of page entries, each page entry storing a status for a page in the erased block, the status including an indication of a bad page or of a good page;
for each page in the erased block, reading the page entry for the page from the page status table;
when the page status from the page status table is bad, processing a next page;
when the page status from the page status table is good, counting a number of un-erased bits in the page;
when the number of un-erased bits exceeds a first threshold, changing the page status to indicate a bad page in the page entry and increasing a bad page count;
when the number of un-erased bits is between a first threshold and a second threshold, and the page status from the page status table indicates that error-correction code (ECC) protection is needed, processing a next page;
when the number of un-erased bits is between a first threshold and a second threshold, and the page status from the page status table indicates that ECC protection is not needed, changing the page status in the page entry to indicate that ECC protection is needed by marking its page status as protect page and increasing an ECC protection page count;
when the bad page count for the erased block exceeds a third threshold, indicating that the erased block is a bad block and not writing new data to the erased block by marking its page status as bad page,
whereby bad erased blocks are identified by counting bad pages within the erased block and allowing bad pages to exist in a good block to prolong flash endurance.

### Claim 8 (depends on 7)

The SEED of claim 7 further comprising:
an ECC generator that generates an ECC code having more bits per unit of data stored in the flash memory for pages having a lower reliability; and
a LDPC generator, activated by an ECC manager, for generating a Low-Density parity Check (LDPC) code for data having a high reliability in the flash memory;
whereby ECC code is dynamic.

### Claim 9

A Virtual-Memory Device (VMD) driver for executing on a host comprising:
a host Endurance Translation Layer (ETL) executing on the host, the host ETL for increasing endurance of a flash memory having a low specified erase-cycle lifetime;
a host dynamic-random-access memory (DRAM) buffer on the host, the host DRAM buffer controlled by a host controller that uses the host ETL to provide temporary storage on the host to reduce flash wear;
a Super Enhanced Endurance Device (SEED) on a Solid-State Drive (SSD) on a SSD, the SEED for increasing endurance of the flash memory that has the low specified erase-cycle lifetime; and
a SEED Endurance Translation Layer (ETL) in a SEED DRAM buffer that is controlled by a SEED controller, the SEED ETL providing temporary storage to reduce flash wear.

### Claim 10 (depends on 9)

The VMD driver of claim 9 further comprising:
a flush manager for flushing data stored in the host ETL of the host DRAM buffer to the SEED ETL in the SEED DRAM buffer and then to the flash memory of a flash drive system when power is lost;
a resume manager reloader for fetching flushed data from flash memory of the flash drive system and in the SEED ETL in the SEED DRAM buffer and then to the host ETL in the host DRAM buffer when power is restored.

### Claim 11 (depends on 9)

The VMD driver of claim 9 further comprising:
a system transaction manager for creating logs to track a persistency of write activities to the SSD and for periodically setting up checkpoints to remove completely finished transactions; and
a SSD transaction manager for creating logs to tracks a persistency of write activities to the flash memory and for periodically setting up checkpoints to remove completely finished transactions.

### Claim 12 (depends on 9)

The VMD driver of claim 9 further comprising a house cleaning procedure on the host which comprises:
obtaining a last access date of a file;
using a current date to calculate a delay time since the last access date of the file;
when the delay time exceeds a time threshold, deleting the file when an auto-clean setting is enabled and the file has a file type of temporary, log, or paging;
searching for duplicates of the file and deleting the duplicates;
repeating for other files on the host.
