Content deleted Content added
→Distributed fault-tolerant file systems: fixed HMDFS ATM |
No edit summary |
||
| (9 intermediate revisions by 3 users not shown) | |||
Line 67:
* [[Reiser4]] – File system that uses [[journaling file system|journaling]], newest version of ReiserFS
* [[Reliance (file system)|Reliance]] – Datalight's transactional file system for high reliability applications
* [[Reliance Nitro]] – Tree-based transactional, [[copy-on-write]] file system developed for high-performance embedded systems, from Datalight (Acquired by [[Tuxera]] in 2019)<ref name=":0">{{Cite web |last=Tuxera |date=2019-06-05 |title=Tuxera acquires mission-critical embedded flash storage leader Datalight |url=https://www.tuxera.com/blog/tuxera-acquires-mission-critical-embedded-flash-storage-leader-datalight/ |access-date=2024-06-28 |website=Tuxera |language=en-US}}</ref>
* [[Remote File System|RFS]] – Native filesystem for [[RTEMS]]<ref name="rfs">{{cite web|title=RTEMS File System|url=http://www.rtems.org/wiki/index.php/RTEMS_File_System|access-date=20 April 2013}}</ref>
* [[SkyFS]] – Developed for [[SkyOS]] to replace BFS as the operating system's main file system. It is based on BFS, but contains many new features.
Line 91:
* [[ReFS|ReFS (Resilient File System)]] – A file system by [[Microsoft]] with built-in resiliency features.
* [[Reliance (file system)|Reliance]] – A transactional file system with [[cyclic redundancy check|CRCs]], created by Datalight.
* [[Reliance Nitro]] – A tree-based transactional, [[copy-on-write]] file system with CRCs, developed for high performance and reliability in embedded systems, from Datalight (Acquired by [[Tuxera]] in 2019)<ref name=":0" />.
* [[ZFS]] – Has checksums for all data; important metadata is always redundant, additional redundancy levels are user-configurable; [[copy-on-write]] and transactional writing ensure metadata consistency; corrupted data can be automatically repaired if a redundant copy is available. Created by [[Sun Microsystems]] for use on [[Solaris (operating system)|Solaris]] 10 and [[OpenSolaris]], ported to [[FreeBSD]] 7.0, [[NetBSD]] (as of August 2009), [[Linux]] and to [[Filesystem in Userspace|FUSE]] (not to be confused with the two zFSes from [[IBM]])
Line 113:
* [[NOVA (filesystem)|NOVA]] – the "non-volatile memory accelerated" file system for persistent main memory.
* [[One File System|OneFS]] – a filesystem utilized by [[Isilon Systems|Isilon]]. It supports selective placement of meta-data directly onto flash SSD.
* [https://www.tuxera.com/products/reliance-velocity-flash-file-system/ Reliance Velocity] - a proprietary flash file system by [[Tuxera]] with high resilience (fail-safe technology) and built-in data integrity. This file system is best suited for embedded applications requiring heavy data workloads over long-term operations. Reliance Velocity can used for all block based media like [[MultiMediaCard|eMMC]], [[Universal Flash Storage|UFS]], eSD, [[SD card]], [[CompactFlash|CF card]], and [[Solid-state drive|SSD]]. It is compatible for [[Linux]], [[Android (operating system)|Android]] and [[QNX]] with portability to other embedded operating systems.
* [https://www.tuxera.com/products/reliance-edge/ Reliance Edge] - a proprietary file system by [[Tuxera]] for resource-constrained embedded systems. It has built-in [[data integrity]] with [[copy-on-write]] transactional technology and [[Deterministic algorithm|deterministic]] operations. This file system can be used for block based media and is configurable for Small [[POSIX]], Full POSIX and can be ported to many [[Real-time operating system|RTOS]] environments. Tuxera has a certified version of this file system called [https://www.tuxera.com/products/reliance-assure/ Reliance Assure]. The source code of Reliance Assure is complaint to [[MISRA C]] and developed following the [[Automotive SPICE|ASPICE]] framework.
* [[Segger Microcontroller Systems]] emFile – filesystem for deeply embedded applications which supports both NAND and NOR flash. Wear leveling, fast read and write, and very low RAM usage.
* [[SPIFFS]] – SPI Flash File System, a wear-leveling filesystem intended for small NOR flash devices.
Line 190 ⟶ 192:
* [[Tahoe-LAFS]] is an open source secure, decentralized, fault-tolerant filesystem utilizing encryption as the basis for a least-authority replicated design.
* A [[FAT12]] and [[FAT16]] (and [[FAT32]]) extension to support automatic file distribution across nodes with extra attributes like ''local'', ''mirror on update'', ''mirror on close'', ''compound on update'', ''compound on close'' in IBM [[4680 OS]] and Toshiba [[4690 OS]]. The distribution attributes are stored on a file-by-file basis in [[Design of the FAT file system#DIR|special entries]] in the directory table.<ref name="IBM_4690_OS_Distribution_Attributes_1">IBM (2003). ''Information about 4690 OS unique file distribution attributes'', IBM document R1001487, 2003-07-30. ({{cite web |url=http://www-01.ibm.com/support/docview.wss?uid=pos1R1001487 |title=IBM Information about 4690 OS unique file distribution attributes - United States |access-date=2014-05-20 |url-status=dead |archive-url=https://web.archive.org/web/20140521070339/http://www-01.ibm.com/support/docview.wss?uid=pos1R1001487 |archive-date=2014-05-21 }}): "[...] file types are stored in the "Reserved bits" portion of the PC-DOS file directory structure [...] only 4690 respects and preserves these attributes. Various non-4690 operating systems take different actions if these bits are turned on [...] when copying from a diskette created on a 4690 system. [...] PC-DOS and Windows 2000 Professional will copy the file without error and zero the bits. OS/2 [...] 1.2 [...] will refuse to copy the file unless [...] first run CHKDSK /F on the file. After [...] CHKDSK, it will copy the file and zero the bits. [...] when [...] copy [...] back to the 4690 system, [...] file will copy as a local file."</ref><ref name="IBM_4690_OS_Distribution_Attributes_2">IBM. ''4690 save and restore file distribution attributes''. IBM document R1000622, 2010-08-31 ({{cite web |url=http://www-01.ibm.com/support/docview.wss?uid=pos1R1000622 |title=IBM 4690 save and restore file distribution attributes - United States |access-date=2014-05-20 |url-status=dead |archive-url=https://web.archive.org/web/20140521070536/http://www-01.ibm.com/support/docview.wss?uid=pos1R1000622 |archive-date=2014-05-21 }}).</ref>
* [[OpenHarmony]] Distributed File System (HMDFS) used for [[Huawei]]'s [[HarmonyOS]] with [[HarmonyOS NEXT]] base and [[OpenHarmony]]-based operating systems, alongside [[EulerOS|openEuler]] server OS that is a cross-device file access where devices can read and edit files on transparently when the two devices are connected to the same network with [[Access token manager]]. Multiple embedded devices connected to the network can automatically synchronise file data with the edge server.<ref>{{Cite web |title=distributed-fs-overview |url=https://docs.openeuler.org/en/docs/22.03_LTS_SP3/docs/Distributed/distributed-fs-overview.html |access-date=2024-05-23 |website=docs.openeuler.org}}</ref>
Line 198 ⟶ 200:
Some of the distributed parallel file systems use an [[object storage device]] (OSD) (in Lustre called OST) for chunks of data together with centralized [[metadata]] servers.
*[[BeeGFS|BeeGFS]] is a hardware-independent parallel file system that features distributed metadata and striping of files across multiple targets, such as NVMe devices or logical volumes.
*[[Lustre (file system)|Lustre]] is an [[Open-source software|open-source]] high-performance distributed parallel file system for Linux, used on many of the largest computers in the world.
*[[Parallel Virtual File System]] (PVFS, PVFS2, [[OrangeFS]]). Developed to store virtual system images, with a focus on non-shared writing optimizations. Available for [[Linux]] under [[GNU General Public License|GPL]].
Line 381 ⟶ 384:
Some of these may be called [[cooperative storage cloud]].
* [[
* [[Scality]] is a distributed filesystem using the [[Chord (peer-to-peer)|Chord]] peer-to-peer protocol.
* [[IPFS]] InterPlanetary File System is p2p, worldwide distributed content-addressable, file-system.
| |||