# US 12,761,804 B2 · Method of operating memory system and memory system performing the same > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,761,804 B2 - Document ID: US12761804B2 - Status: Granted - Kind: B2 - Application number: 19/170,582 - Filed: 2025-04-04 - Published: 2025-07-17 - Granted: 2026-10-13 - Earliest claimed: 2023-03-21 - Assignees: Samsung Electronics Co., Ltd. - Applicants: Samsung Electronics Co., Ltd. - Inventors: Changho Choi; Young Bong Kim; Eun-Kyung Choi - Web page: https://patentgrep.com/patent/US12761804B2 ## CPC - G06F 3/0613: {in relation to throughput} - G06F 3/064: {Management of blocks} - G06F 3/0679: {Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]} - G06F 12/0246: {in block erasable memory, e.g. flash memory} - G06F 2212/7205: Cleaning, compaction, garbage collection, erase control - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract In a method of operating a memory system disclosed, whether a first condition is satisfied is determined. The first condition is associated with free blocks and garbage collection (GC) target blocks from among a plurality of memory blocks. In response to the first condition being satisfied, a size of a data sample associated with executions of a host input/output request and GC is adjusted. The data sample is generated based on the adjusted size of the data sample. The data sample includes a downscaled current valid page count (VPC) ratio and the first number of previous host input/output request to GC processing ratios. A current host input/output request to GC processing ratio is calculated based on the data sample. The host input/output request and the GC are performed based on the current host input/output request to GC processing ratio. ## Claims ### Claim 1 A method of operating a storage system, the method comprising: determining a first garbage collection (GC) level among a plurality of GC levels based on a level of decrease of free blocks among a plurality of memory blocks; reducing a size of a data sample associated with executions of host input/output requests and GC input/output requests to generate a first data sample having the reduced size, based on the first GC level; and performing the host input/output requests and the GC input/output requests, based on the first data sample. ### Claim 2 (depends on 1) The method of claim 1, wherein the first GC level represents that a number of the free blocks has decreased below a given value. ### Claim 3 (depends on 1) The method of claim 1, wherein the first GC level represents that a ratio of the free blocks to the plurality of memory blocks decreases below a given ratio. ### Claim 4 (depends on 1) The method of claim 1, wherein the plurality of GC levels include: a high-risk level indicating when a number or a ratio of the free blocks decreases below a given number or a given ratio; and a block level indicating when the number or the ratio of the free blocks excessively decreases below the given number or the given ratio. ### Claim 5 (depends on 4) The method of claim 4, wherein the performing the host input/output requests and the GC input/output requests includes: performing the GC input/output requests first and then the host input/output requests, based on the first GC level being the high-risk level. ### Claim 6 (depends on 4) The method of claim 4, wherein the performing the host input/output requests and the GC input/output requests includes: performing the GC input/output requests only and blocking the host input/output requests, based on the first GC level being the block level. ### Claim 7 (depends on 1) The method of claim 1, wherein the size of the data sample is associated with an acquisition speed for acquiring new free blocks. ### Claim 8 (depends on 7) The method of claim 7, wherein the new free blocks are acquired faster as the size of the data sample decreases. ### Claim 9 (depends on 1) The method of claim 1, wherein the first data sample includes: a downscaled current valid page count (VPC) ratio for a current GC block among GC target blocks and a first number of previous host input/output request and GC processing ratios. ### Claim 10 (depends on 9) The method of claim 9, further comprising: determining a second GC level among the plurality of GC levels; maintaining the size of the data sample to generate a second data sample having the maintained size, based on the second GC level; and performing the host input/output requests and the GC input/output requests, based on the second data sample, wherein the second data sample includes the downscaled current VPC ratio and a second number of previous host input/output requests and GC processing ratios, and wherein the second number is greater than the first number. ### Claim 11 A method of operating a memory device, the method comprising: determining a first garbage collection (GC) level among a plurality of GC levels based on a level of decrease of free blocks among a plurality of memory blocks; adjusting a size of a data sample associated with executions of host input/output requests and GC input/output requests to generate the data sample having the adjusted size, based on the first GC level; and performing the host input/output requests and the GC input/output request, based on the data sample. ### Claim 12 (depends on 11) The method of claim 11, wherein the determining the first GC level includes: detecting the level of decrease by determining whether a number or a ratio of the free blocks decreases below a given number or a given ratio; and identifying the first GC level, based on the level of decrease. ### Claim 13 (depends on 11) The method of claim 11, wherein the generating the data sample having the adjusted size includes: adjusting the data sample to have a first size, based on the first GC level being a high-risk level; and adjusting the data sample to have a second size that is smaller than the first size, based on the first GC level being a block level. ### Claim 14 (depends on 13) The method of claim 13, wherein: the high-risk level indicating when a number or a ratio of the free blocks decreases below a given number or a given ratio; and the block level indicating when the number or the ratio of the free blocks excessively decreases below the given number or the given ratio. ### Claim 15 (depends on 11) The method of claim 11, wherein the size of the data sample is associated with an acquisition speed for acquiring new free blocks. ### Claim 16 A method of operating a memory controller, the method comprising: detecting a level of decrease of free blocks among a plurality of memory blocks; determining a first garbage collection (GC) level among a plurality of GC levels, based on the level of decrease; adjusting a data sample associated with executions of host input/output requests and GC input/output requests to generate the data samples having the adjusted size, based on the first GC level; and issuing the host input/output requests and the GC input/output request to a memory device, based on the data sample. ### Claim 17 (depends on 16) The method of claim 16, wherein the first GC level indicates when a number or a ratio of the free blocks decreases below a given number or a given ratio. ### Claim 18 (depends on 16) The method of claim 16, wherein the new free blocks are acquired faster as the size of the data sample decreases. ### Claim 19 (depends on 16) The method of claim 16, wherein the issuing the host input/output requests and the GC input/output requests to the memory device includes: calculating a downscaled current host input/output request to GC processing ratio, based on the data sample; and generating the host input/output requests and the GC input/output requests, based on the downscaled current host input/output request to GC processing ratio. ### Claim 20 (depends on 19) The method of claim 19, wherein the issuing the host input/output requests and the GC input/output requests to the memory device further includes: determining one or more downscaled host tokens and one or more downscaled GC tokens, based on the downscaled current host input/output request to GC processing ratio; deducting the one or more downscaled host tokens one by one whenever the host input/output request is performed; and deducting the one or more downscaled GC tokens one by one whenever the GC is performed.