# US 2021/0334157 A1 · Resiliency scheme to enhance storage performance > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0334157 A1 - Document ID: US20210334157A1 - Status: Published application - Kind: A1 - Application number: 16/857,475 - Filed: 2020-04-24 - Published: 2021-10-28 - Earliest claimed: 2020-04-24 - Assignees: Pure Storage, Inc. - Applicants: Pure Storage, Inc. - Inventors: Hari Kannan; Nenad Miladinovic - Web page: https://patentgrep.com/patent/US20210334157A1 ## CPC - G06F 11/108: {Parity data distribution in semiconductor storages, e.g. in SSD} - G06F 3/061: {Improving I/O performance} - G06F 3/0619: {in relation to data integrity, e.g. data losses, bit errors} - G06F 3/064: {Management of blocks} - G06F 3/067: {Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]} - G06F 3/0688: {Non-volatile semiconductor memory arrays} - G06F 11/106: {Correcting systematically all correctable errors, i.e. scrubbing} - G06F 11/1096: {Parity calculation or recalculation after configuration or reconfiguration of the system} - G06F 11/2069: {Management of state, configuration or failover} - G06F 11/2094: {Redundant storage or storage space (G06F11/2056 takes precedence)} - G06F 12/0246: {in block erasable memory, e.g. flash memory} - G06F 2211/109: Sector level checksum or ECC, i.e. sector or stripe level checksum or ECC in addition to the RAID parity calculation - G06F 2212/7211: Wear leveling - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract A storage system has a resiliency scheme to enhance storage system performance. The storage system composes a RAID stripe. The storage system mixes an ordering of portions of the RAID stripe, based on reliability differences across portions of the solid-state memory. The storage system writes the mixed ordering RAID stripe across the solid-state memory. ## Claims ### Claim 1 A method for a resiliency scheme to enhance storage system performance, performed by a storage system, comprising: composing a RAID stripe of data to be written across solid-state memory of the storage system; mixing an ordering of portions of the RAID stripe of data, based on reliability differences across portions of the solid-state memory; and writing the mixed ordering RAID stripe of data across the solid-state memory. ### Claim 2 (depends on 1) The method of claim 1, wherein: the reliability differences comprise reliability differences among pages of an atomic write page group of the solid-state memory; the mixed ordering of portions of the RAID stripe comprises a first RAID stripe with a first mixed ordering into pages of a first atomic write page group in the solid-state memory; and the storage system is further to write a second, differing mixed ordering of portions of a second RAID stripe into pages of a second atomic write page group in the solid-state memory. ### Claim 3 (depends on 1) The method of claim 1, wherein: the reliability differences comprise reliability differences among pages in an atomic write page group of the solid-state memory; and the mixed ordering of portions of the RAID stripe comprises a first mixed ordering of a first portion of the RAID stripe across a first type of solid-state memory and a second, differing mixed ordering of a second portion of the RAID stripe of data across a second type of solid-state memory. ### Claim 4 (depends on 1) The method of claim 1, further comprising: using an identifier of a storage device of the storage system as a seed for a function that determines the mixed ordering. ### Claim 5 (depends on 1) The method of claim 1, further comprising: reading the mixed ordering RAID stripe of data from the solid-state memory; and rearranging the ordering of the portions of the mixed ordering RAID stripe to reconstitute the RAID stripe of data. ### Claim 6 (depends on 1) The method of claim 1, further comprising: selecting an error correction code (ECC) scheme based on the reliability differences across the portions of the solid-state memory, for the RAID stripe of data. ### Claim 7 (depends on 1) The method of claim 1, further comprising: scrubbing the solid-state memory, through wear monitoring the solid-state memory at varied address intervals or varied address offsets, and rewriting data from worn regions of solid-state memory to further regions of solid-state memory using mixed ordering of portions of the rewritten data. ### Claim 8 A tangible, non-transient, computer readable media having instructions thereon for one or more processors to perform a method, comprising: composing a RAID stripe of data to be written across solid-state memory of a storage system; mixing an ordering of portions of the RAID stripe of data, based on reliability differences across portions of the solid-state memory; and writing the mixed ordering RAID stripe of data across the solid-state memory. ### Claim 9 (depends on 8) The computer readable media of claim 8, wherein: the reliability differences comprise reliability differences among pages of an atomic write page group of the solid-state memory; the mixed ordering of portions of the RAID stripe comprises a first RAID stripe with a first mixed ordering into pages of a first atomic write page group in the solid-state memory; and the storage system is further to write a second, differing mixed ordering of portions of a second RAID stripe into pages of a second atomic write page group in the solid-state memory. ### Claim 10 (depends on 8) The computer readable media of claim 8, wherein: the reliability differences comprise reliability differences among pages in an atomic write page group of the solid-state memory; and the mixed ordering of portions of the RAID stripe comprises a first mixed ordering of a first portion of the RAID stripe across a first type of solid-state memory and a second, differing mixed ordering of a second portion of the RAID stripe of data across a second type of solid-state memory. ### Claim 11 (depends on 8) The computer readable media of claim 8, wherein the method further comprises: reading the mixed ordering RAID stripe of data from the solid-state memory; and rearranging the ordering of the portions of the mixed ordering RAID stripe to reconstitute the RAID stripe of data. ### Claim 12 (depends on 8) The computer readable media of claim 8, wherein the method further comprises: selecting an error correction code (ECC) scheme based on the reliability differences across the portions of the solid-state memory, for the RAID stripe of data. ### Claim 13 (depends on 8) The computer readable media of claim 8, wherein the method further comprises: scrubbing the solid-state memory, through wear monitoring the solid-state memory at varied address intervals or varied address offsets, and rewriting data from worn regions of solid-state memory to further regions of solid-state memory using mixed ordering of portions of the rewritten data. ### Claim 14 A storage system, comprising: solid-state memory; and a processing device to: compose a RAID stripe of data to be written across solid-state memory of the storage system; mix an ordering of portions of the RAID stripe of data, based on reliability differences across portions of the solid-state memory; and write the mixed ordering RAID stripe of data across the solid-state memory. ### Claim 15 (depends on 14) The storage system of claim 14, wherein: the reliability differences comprise reliability differences among pages of an atomic write page group of the solid-state memory; the mixed ordering of portions of the RAID stripe comprises a first RAID stripe with a first mixed ordering into pages of a first atomic write page group in the solid-state memory; and the processing device is further to write a second, differing mixed ordering of portions of a second RAID stripe into pages of a second atomic write page group in the solid-state memory. ### Claim 16 (depends on 14) The storage system of claim 14, wherein: the reliability differences comprise reliability differences among pages in an atomic write page group of the solid-state memory; and the mixed ordering of portions of the RAID stripe comprises a first mixed ordering of a first portion of the RAID stripe across a first type of solid-state memory and a second, differing mixed ordering of a second portion of the RAID stripe of data across a second type of solid-state memory. ### Claim 17 (depends on 14) The storage system of claim 14, wherein the processing device is further to: use an identifier of a storage device of the storage system as a seed for a function that determines the mixed ordering. ### Claim 18 (depends on 14) The storage system of claim 14, wherein the processing device is further to: read the mixed ordering RAID stripe of data from the solid-state memory; and rearrange the ordering of the portions of the mixed ordering RAID stripe to reconstitute the RAID stripe of data. ### Claim 19 (depends on 14) The storage system of claim 14, wherein the processing device is further to: select an error correction code (ECC) scheme based on the reliability differences across the portions of the solid-state memory, for the RAID stripe of data. ### Claim 20 (depends on 14) The storage system of claim 14, wherein the processing device is further to: scrub the solid-state memory, through wear monitoring the solid-state memory at varied address intervals or varied address offsets, and rewriting data from worn regions of solid-state memory to further regions of solid-state memory using mixed ordering of portions of the rewritten data.